The Neolithic Settlement of Knossos

52
The Neolithic Settlement of Knossos in Crete New Evidence for the Early Occupation of Crete and the Aegean Islands

Transcript of The Neolithic Settlement of Knossos

The Neolithic Settlement of Knossos in Crete

New Evidence for the Early Occupation of Crete and the Aegean Islands

Frontispiece The city the fortifications the harbor and the hinterland of Khandax (Herakleion) in the first half of the 17th century Map by unknown cartographer 17th c Collezione Museo Civico Padua Vikelaia Municipal Library Herakleion

PREHISTORY MONOGRAPHS 42

The Neolithic Settlement of Knossos in Crete

New Evidence for the Early Occupation of Crete and the Aegean Islands

edited by

Nikos Efstratiou Alexandra Karetsou and Maria Ntinou

Published byINSTAP Academic Press

Philadelphia Pennsylvania2013

Design and ProductionINSTAP Academic Press Philadelphia PA

Printing and BindingHoster Bindery Inc Ivyland PA

Library of Congress Cataloging-in-Publication Data

The neolithic settlement of Knossos in Crete new evidence for the early occupation of Crete and the Aegean islands edited by Nikos Efstratiou Alexandra Karetsou and Maria Ntinou pages cm -- (Prehistory monographs 42) Includes bibliographical references and index ISBN 978-1-931534-72-7 (hardcover alk paper) 1 Knossos (Extinct city) 2 Neolithic period--Greece--Crete 3 Land settlement patterns Prehistoric--Greece--Crete 4 Crete (Greece)--Antiquities I Efstratiou Nicholas DF221C8N46 2013 939rsquo18--dc23

2013016076

Copyright copy 2013

INSTAP Academic PressPhiladelphia Pennsylvania

All rights reservedPrinted in the United States of America

To the memory of Professor JD Evans a gentleman of British Archaeology

ndashNikos Efstratiou

Table of Contents

List of Tables in the Text ix

List of Figures in the Text xiii

Preface Alexandra Karetsou xix

Acknowledgments xxiii

Introduction Nikos Efstratiou xxv

1 The Excavation Nikos Efstratiou Alexandra Karetsou and Eleni Banou 1

2 The Stratigraphy and Cultural Phases Nikos Efstratiou 25

3 Fabric Diversity in the Neolithic Ceramics of Knossos Sarantis Dimitriadis 47

4 Neolithic Sedimentology at Knossos Maria-Pilar Fumanal Garciacuteadagger 53

5 The Economy of Neolithic Knossos The Archaeobotanical Data Anaya Sarpaki 63

6 Wood Charcoal Analysis The Local Vegetation Ernestina Badal and Maria Ntinou 95

7 Plant Economy and the Use of Space Evidence from the Opal Phytoliths Marco Madella 119

8 The Knossos Fauna and the Beginning of the Neolithic in the Mediterranean Islands Manuel Peacuterez Ripoll 133

9 The Earliest Settlement on Crete An Archaeozoological Perspective Liora Kolska Horwitz 171

viii THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

10 Radiocarbon Dates from the Neolithic Settlement of Knossos An Overview Yorgos Facorellis and Yiannis Maniatis 193

11 Knossos and the Beginning of the Neolithic in Greece and the Aegean Islands Nikos Efstratiou 201

Index 215

Table 41 Correlation of sedimentology samples with excavation levels and cultural phases 55

Table 42 Munsell color and calculation of statistical parameters of mean size sorting skewness and kurtosis for each of the analyzed sedimentology samples 56

Table 51 Seed list provided to JD Evans by Hans Helbaek (unpublished) 67

Table 52 List of archaeobotanical samples from the 1997 rescue excavation along with relative and absolute dates 67

Table 53 Aceramic Neolithic archaeobotanical sample E 97(30) from Knossos 1997 level 39 retrieved from 16 liters of water-floated soil 68

Table 54 Measurements of Triticum turgidumaestivum from Aceramic and EN levels at Knossos 71

Table 55 Measurements of Triticum dicoccum Triticum monococcum Hordeum vulgare and Lens culinaris 74

Table 56 Early Neolithic I archaeobotanical (seed) samples 75

Table 57 Measurements of Trifolium spp and Leguminosae 78

Table 58 Measurements of Raphanus cf raphanistrum and Linum cf usitatissimum 78

Table 59 Early Neolithic II archaeobotanical (seed) samples 81

List of Tables in the Text

x THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Table 510 Vitis sp measurements from EN II levels sketch of a grape seed showing locations of dimensions 85

Table 511 Middle Neolithic archaeobotanical (seed) samples 86

Table 512 Late Neolithic archaeobotanical (seed) samples 88

Table 61 Inventories of plants growing in different parts of the study area 99

Table 62 Absolute and relative frequencies of taxa identified in the wood charcoal assemblages from Neolithic Knossos 102

Table 63 Presence of plant taxa in wood charcoal assemblages from Neolithic Knossos along with the total number of fragments analyzed and the total number of taxa identified in each level 104

Table 71 Knossos 1997 south profile phytolith counts 122

Table 72 Knossos 1997 west profile phytolith counts 125

Table 81 Number of identified and unidentified specimens by taxa and period 135

Table 82 Measurements of bones from Bos taurus 135

Table 83 Measurements of bones from Ovis aries (Oa) and Capra hircus (Ch) 136

Table 84 Measurements of bones from Sus scrofa domesticus Sus scrofa ferus Capra aegagrus Martes and Meles meles 138

Table 85 Number of identified specimens of Bos OvisCapra and Sus with number of marks caused by dog gnawing 140

Table 86 Early Neolithic I and EN II faunal remains 141

Table 87 Early Neolithic IIMN faunal remains 142

Table 88 Middle Neolithic faunal remains 144

Table 89A Late Neolithic faunal remains 146

Table 89B Late Neolithic faunal remains 149

Table 810 Number of identified specimens of Bos and OvisCapraSus with burn marks 151

Table 811 Number of long bone remains (the diaphysis fragments are not counted here) phalanges and tarsi corresponding to mature and immature bones along with the number of LN tooth remains grouped by age for comparison with the long bones 156

Table 812 Number of mandibles (NM) for goats and sheep from the LN levels classified by age 157

Table 813 Number of mandibles (NM) of Bos taurus classified by age 158

Table 814 Number of maxillary and mandibular remains of Sus scrofa domesticus classified by age 159

Table 815 Number of identified specimens of Bos taurus classified by sex 160

Table 816 Number of identified specimens of Ovis aries (Oa) and Capra hircus (Ch) classified by sex 160

xiLIST OF TABLES IN THE TEXT

Table 817 Chronological representation of the faunal species at Knossos 161

Table 818 Percentages of identified specimens of domestic and wild species at Knossos and other sites 161

Table 819 Representation and abundance of various faunal species at Shillourokambos Ais Yiorkis and Khirokitia 162

Table 820 Introduction and chronological representation of wild animals at various sites in Crete 163

Table 91 Relative frequencies (percentages) of animal species from Knossos 175

Table 92 Schematic representation of the relative chronology (cal bc dates) of sites mentioned in the text 183

Table 101 Summary of the British Museum radiocarbon dates on charcoal from the excavations of JD Evans at Neolithic Knossos sorted by age 194

Table 102 Description of the samples dated in the British Museum Research Laboratory 195

Table 103 Summary of radiocarbon dating results of carbonized samples collected in 1997 from the Neolithic settlement levels at Knossos 196

Frontispiece The city the fortifications the harbor and the hinterland of Khandax (Herakleion) in the first half of the 17th century Map by unknown cartographer 17th c Collezione Museo Civico Padua Vikelaia Municipal Library Herakleion ii

Figure i The Minoan palace and its Neolithic past xxi

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation 2

Figure 12 Trench II (a) view of the Central Court of the Palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the souther profile 3

Figure 13 Plan of the excavation trenches next to the staircase 3

Figure 14 South and west stratigraphic profiles of the trench 4

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench 5

Figure 16 Plans of excavation levels 9 and 10 7

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner 8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner 8

List of Figures in the Text

xiv THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 19 Plan of excavation level 13 9

Figure 110 Plan of excavation level 14 showing hearth in northwest corner 9

Figure 111 View of excavation level 14 showing hearth in northwest corner 10

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south 11

Figure 113 View of excavation level 15 from above 12

Figure 114 View of excavation level 16 from above 12

Figure 115 View of excavation level 16 facing west section 12

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4 12

Figure 117 View of level 17 facing west section 13

Figure 118 View of excavation level 18 from above 13

Figure 119 View of excavation level 19 from above 13

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6 13

Figure 121 View of excavation level 21 from above 14

Figure 122 View of excavation level 23 14

Figure 123 View of excavation level 24 showing wall 7 and grinding stones 14

Figure 124 Plans of excavation levels 22 and 24 15

Figure 125 Plan of excavation level 27 16

Figure 126 View of excavation level 28 from above 16

Figure 127 View of excavation level 28 facing west profile 16

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths 17

Figure 129 View of level 29A showing hearths 1 2 and 3 18

Figure 130 View of excavation level 30 showing hearth 4 18

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure 19

Figure 132 View of the elliptical stone wall from levels 24ndash27 19

Figure 133 View of excavation level 31 19

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7 20

Figure 135 Plan of excavation level 37 showing pits 1 and 2 21

Figure 21 Sedimentological samples of the middle part of the south profile 26

Figure 41 Fine fraction granulometry () of the samples from the west profile 57

Figure 42 The organic content () of the samples from the west profile 57

Figure 43 The carbonate content of the samples from the west profile 57

xvLIST OF FIGURES IN THE TEXT

Figure 44 Morphoscopy of sands without acid treatment 57

Figure 45 Morphoscopy of sands after the elimination of calcareous grains (subsequent to acid treatment) 57

Figure 51 Drawing of Triticum turgidum LT aestivum from the 1997 excavations at Knossos 69

Figure 52 Graphs of measurements and measurement ratios of Triticum turgidumaestivum from Neolithic Knossos 72

Figure 53 Measurements of Triticum turgidumaestivum from Aceramic and Early Neolithic Knossos compared with average values for Erbaba Ramad and Bouqras in the Near East 73

Figure 54 Graphs of measurements and measurement ratios of Lens culinaris from Neolithic Knossos 74

Figure 55 Early sites including those from mainland Greece where Triticum turgidumaestivum is reported 1 Tell Abu Hureyra 2 Tell Halula 3 Tell Aswad 4 Tell Ghoraife 5 Tell Sabi Abyad 6 Servia 7 Cafer Houmlyuumlk 8 Dhali Agridhi 9 Otzaki 10 Sesklo 11 Sitagroi 12 Haccedililar 13 Aşkli Houmlyuumlk 14 Ccedilatal Houmlyuumlk 15 Can Hasan 16 Cayoumlnuuml 17 El Kown 18 Bouqras 19 Tell Ramad 89

Figure 56 Summary of the distribution of all categories of archaeobotanical remains at Neolithic Knossos 90

Figure 61 Climate and topography of Knossos (a) mean annual precipitation in Crete (b) topographic map of the area around Knossos (c) westndasheast topographic section (d) southwestndashnortheast topographic section 96

Figure 62 View of the Knossos valley from Mt Juktas showing present-day vegetation 98

Figure 63 Panoramic view of the site of Knossos showing present-day vegetation 98

Figure 64 Present-day phrygana vegetation on the hills in the study area 98

Figure 65 Present-day vegetation on deep soils in the study area 98

Figure 66 Anatomy of plant taxa identified in wood charcoal assemblages from Neolithic Knossos 106

Figure 67 Wood charcoal diagram from Neolithic Knossos showing relative frequencies of taxa in successive excavation levels 108

Figure 71 Bar chart of phytolith per centage frequencies from the south profile 122

Figure 72 Bar chart of C3 and C

4 phytolith percentage frequencies from the south profile 122

Figure 73 West profile stratigraphy and sampling 123

Figure 74 Bar chart of phytolith percentage frequencies from the west profile 124

Figure 75 Bar chart of C3 and C

4 phytolith percentage frequencies from the west profile 125

Figure 76 Trench section (southwest corner to west face) with phases identified according to phytolith composition and frequencies 127

xvi THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 77 Silica skeleton from grass leaf (long cells and a stoma) from the EN I deposits (sample XXa level 32) 129

Figure 78 Wheat-type silica skeleton from the EN I deposits (sample XXIVb level 32) 129

Figure 79 Silica skeleton from a dicotyledonous plant from the EN I deposits (sample XIV level 30) 129

Figure 710 Millet-type silica skeleton from the EN II deposits (sample Xa level 16) 129

Figure 81 Percentages of the osseous parts of cattle long bones 139

Figure 82 Percentages of the osseous parts belonging to the long bones of middle-sized mammals (goats sheep and pigs) 139

Figure 83 Skeletal fragments of long bones of OvisCapra from level 14 all with dog-gnawing marks 140

Figure 84 Fragments of proximal epiphyses of femur and tibia of Bos taurus with fracture marks caused by impacts from the extraction of marrow level 24 151

Figure 85 Animal bones from level 3 (a) radius diaphysis (b) radius proximal part (c) scapula (d) fragment of femur (e) phalanx I of Capra aegagrus 153

Figure 86 Distal metacarpus of Capra aegagrus and Ovis aries 153

Figure 87 Animal bones (a) ulna in lateral view probably belonging to a wild boar (level 23) (b) Sus scrofa ferus canine fragment (level 10) (c) Sus scrofa domesticus ulna in lateral view (level 14) 154

Figure 88 Diaphysis width range (SD) of Sus scrofa domesticus and Sus scrofa ferus from Zambujal (Portugal) Cerro de la Virgen (Spain) Argissa-Magula and Knossos 154

Figure 89 Meles meles (a) left mandible in lateral view (level 14) (b) lower canine (level 14) (c) left ulna in medial views the proximal epiphysis is not fused (level 3) Martes (d) distal part of humerus in cranial view (level 9) 155

Figure 810 Age classes of the mandibles of Ovis and Capra 157

Figure 811 Age classes of the mandibles of Bos taurus 158

Figure 812 Distal part of metacarpus belonging to a male (possibly ox) of Bos taurus with osseous deformations on the articular surfaces 158

Figure 813 Age classes of Sus scrofa domesticus maxillae and mandibles 159

Figure 814 Correlation of the measurements of phalanx I belonging to Bos taurus 160

Figure 91 Map showing location of sites mentioned in the text 1 Ashkelon 2 lsquoAin Ghazal 3 Atlit Yam 4 Hagoshrim and Tel Ali 5 Ras Shamra 6 Cap Andreas Kastros 7 Khirokitia 8 Tenta 9 Asikli Houmlyuumlk 10 Mersin 11 Can Hasan III 12 Ccedilatalhoumlyuumlk 13 Suberde 14 Haccedililar 15 Nea Nikomedeia 16 Argissa-Magula 17 Sesklo 18 Achilleion 19 Franchthi Cave 20 Sidari Corfu 21 Cave of the Cyclops Youra 22 Melos 23 Santorini 24 Knossos Crete 25 Tel Aray 2 26 Umm el Tlel 27 Qdeir 28 El Kowm 2 173

Figure 92 Adult male agrimi (Capra aegagrus cretica) showing phenotypic resemblance to the wild bezoar goat 177

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

Frontispiece The city the fortifications the harbor and the hinterland of Khandax (Herakleion) in the first half of the 17th century Map by unknown cartographer 17th c Collezione Museo Civico Padua Vikelaia Municipal Library Herakleion

PREHISTORY MONOGRAPHS 42

The Neolithic Settlement of Knossos in Crete

New Evidence for the Early Occupation of Crete and the Aegean Islands

edited by

Nikos Efstratiou Alexandra Karetsou and Maria Ntinou

Published byINSTAP Academic Press

Philadelphia Pennsylvania2013

Design and ProductionINSTAP Academic Press Philadelphia PA

Printing and BindingHoster Bindery Inc Ivyland PA

Library of Congress Cataloging-in-Publication Data

The neolithic settlement of Knossos in Crete new evidence for the early occupation of Crete and the Aegean islands edited by Nikos Efstratiou Alexandra Karetsou and Maria Ntinou pages cm -- (Prehistory monographs 42) Includes bibliographical references and index ISBN 978-1-931534-72-7 (hardcover alk paper) 1 Knossos (Extinct city) 2 Neolithic period--Greece--Crete 3 Land settlement patterns Prehistoric--Greece--Crete 4 Crete (Greece)--Antiquities I Efstratiou Nicholas DF221C8N46 2013 939rsquo18--dc23

2013016076

Copyright copy 2013

INSTAP Academic PressPhiladelphia Pennsylvania

All rights reservedPrinted in the United States of America

To the memory of Professor JD Evans a gentleman of British Archaeology

ndashNikos Efstratiou

Table of Contents

List of Tables in the Text ix

List of Figures in the Text xiii

Preface Alexandra Karetsou xix

Acknowledgments xxiii

Introduction Nikos Efstratiou xxv

1 The Excavation Nikos Efstratiou Alexandra Karetsou and Eleni Banou 1

2 The Stratigraphy and Cultural Phases Nikos Efstratiou 25

3 Fabric Diversity in the Neolithic Ceramics of Knossos Sarantis Dimitriadis 47

4 Neolithic Sedimentology at Knossos Maria-Pilar Fumanal Garciacuteadagger 53

5 The Economy of Neolithic Knossos The Archaeobotanical Data Anaya Sarpaki 63

6 Wood Charcoal Analysis The Local Vegetation Ernestina Badal and Maria Ntinou 95

7 Plant Economy and the Use of Space Evidence from the Opal Phytoliths Marco Madella 119

8 The Knossos Fauna and the Beginning of the Neolithic in the Mediterranean Islands Manuel Peacuterez Ripoll 133

9 The Earliest Settlement on Crete An Archaeozoological Perspective Liora Kolska Horwitz 171

viii THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

10 Radiocarbon Dates from the Neolithic Settlement of Knossos An Overview Yorgos Facorellis and Yiannis Maniatis 193

11 Knossos and the Beginning of the Neolithic in Greece and the Aegean Islands Nikos Efstratiou 201

Index 215

Table 41 Correlation of sedimentology samples with excavation levels and cultural phases 55

Table 42 Munsell color and calculation of statistical parameters of mean size sorting skewness and kurtosis for each of the analyzed sedimentology samples 56

Table 51 Seed list provided to JD Evans by Hans Helbaek (unpublished) 67

Table 52 List of archaeobotanical samples from the 1997 rescue excavation along with relative and absolute dates 67

Table 53 Aceramic Neolithic archaeobotanical sample E 97(30) from Knossos 1997 level 39 retrieved from 16 liters of water-floated soil 68

Table 54 Measurements of Triticum turgidumaestivum from Aceramic and EN levels at Knossos 71

Table 55 Measurements of Triticum dicoccum Triticum monococcum Hordeum vulgare and Lens culinaris 74

Table 56 Early Neolithic I archaeobotanical (seed) samples 75

Table 57 Measurements of Trifolium spp and Leguminosae 78

Table 58 Measurements of Raphanus cf raphanistrum and Linum cf usitatissimum 78

Table 59 Early Neolithic II archaeobotanical (seed) samples 81

List of Tables in the Text

x THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Table 510 Vitis sp measurements from EN II levels sketch of a grape seed showing locations of dimensions 85

Table 511 Middle Neolithic archaeobotanical (seed) samples 86

Table 512 Late Neolithic archaeobotanical (seed) samples 88

Table 61 Inventories of plants growing in different parts of the study area 99

Table 62 Absolute and relative frequencies of taxa identified in the wood charcoal assemblages from Neolithic Knossos 102

Table 63 Presence of plant taxa in wood charcoal assemblages from Neolithic Knossos along with the total number of fragments analyzed and the total number of taxa identified in each level 104

Table 71 Knossos 1997 south profile phytolith counts 122

Table 72 Knossos 1997 west profile phytolith counts 125

Table 81 Number of identified and unidentified specimens by taxa and period 135

Table 82 Measurements of bones from Bos taurus 135

Table 83 Measurements of bones from Ovis aries (Oa) and Capra hircus (Ch) 136

Table 84 Measurements of bones from Sus scrofa domesticus Sus scrofa ferus Capra aegagrus Martes and Meles meles 138

Table 85 Number of identified specimens of Bos OvisCapra and Sus with number of marks caused by dog gnawing 140

Table 86 Early Neolithic I and EN II faunal remains 141

Table 87 Early Neolithic IIMN faunal remains 142

Table 88 Middle Neolithic faunal remains 144

Table 89A Late Neolithic faunal remains 146

Table 89B Late Neolithic faunal remains 149

Table 810 Number of identified specimens of Bos and OvisCapraSus with burn marks 151

Table 811 Number of long bone remains (the diaphysis fragments are not counted here) phalanges and tarsi corresponding to mature and immature bones along with the number of LN tooth remains grouped by age for comparison with the long bones 156

Table 812 Number of mandibles (NM) for goats and sheep from the LN levels classified by age 157

Table 813 Number of mandibles (NM) of Bos taurus classified by age 158

Table 814 Number of maxillary and mandibular remains of Sus scrofa domesticus classified by age 159

Table 815 Number of identified specimens of Bos taurus classified by sex 160

Table 816 Number of identified specimens of Ovis aries (Oa) and Capra hircus (Ch) classified by sex 160

xiLIST OF TABLES IN THE TEXT

Table 817 Chronological representation of the faunal species at Knossos 161

Table 818 Percentages of identified specimens of domestic and wild species at Knossos and other sites 161

Table 819 Representation and abundance of various faunal species at Shillourokambos Ais Yiorkis and Khirokitia 162

Table 820 Introduction and chronological representation of wild animals at various sites in Crete 163

Table 91 Relative frequencies (percentages) of animal species from Knossos 175

Table 92 Schematic representation of the relative chronology (cal bc dates) of sites mentioned in the text 183

Table 101 Summary of the British Museum radiocarbon dates on charcoal from the excavations of JD Evans at Neolithic Knossos sorted by age 194

Table 102 Description of the samples dated in the British Museum Research Laboratory 195

Table 103 Summary of radiocarbon dating results of carbonized samples collected in 1997 from the Neolithic settlement levels at Knossos 196

Frontispiece The city the fortifications the harbor and the hinterland of Khandax (Herakleion) in the first half of the 17th century Map by unknown cartographer 17th c Collezione Museo Civico Padua Vikelaia Municipal Library Herakleion ii

Figure i The Minoan palace and its Neolithic past xxi

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation 2

Figure 12 Trench II (a) view of the Central Court of the Palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the souther profile 3

Figure 13 Plan of the excavation trenches next to the staircase 3

Figure 14 South and west stratigraphic profiles of the trench 4

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench 5

Figure 16 Plans of excavation levels 9 and 10 7

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner 8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner 8

List of Figures in the Text

xiv THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 19 Plan of excavation level 13 9

Figure 110 Plan of excavation level 14 showing hearth in northwest corner 9

Figure 111 View of excavation level 14 showing hearth in northwest corner 10

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south 11

Figure 113 View of excavation level 15 from above 12

Figure 114 View of excavation level 16 from above 12

Figure 115 View of excavation level 16 facing west section 12

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4 12

Figure 117 View of level 17 facing west section 13

Figure 118 View of excavation level 18 from above 13

Figure 119 View of excavation level 19 from above 13

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6 13

Figure 121 View of excavation level 21 from above 14

Figure 122 View of excavation level 23 14

Figure 123 View of excavation level 24 showing wall 7 and grinding stones 14

Figure 124 Plans of excavation levels 22 and 24 15

Figure 125 Plan of excavation level 27 16

Figure 126 View of excavation level 28 from above 16

Figure 127 View of excavation level 28 facing west profile 16

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths 17

Figure 129 View of level 29A showing hearths 1 2 and 3 18

Figure 130 View of excavation level 30 showing hearth 4 18

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure 19

Figure 132 View of the elliptical stone wall from levels 24ndash27 19

Figure 133 View of excavation level 31 19

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7 20

Figure 135 Plan of excavation level 37 showing pits 1 and 2 21

Figure 21 Sedimentological samples of the middle part of the south profile 26

Figure 41 Fine fraction granulometry () of the samples from the west profile 57

Figure 42 The organic content () of the samples from the west profile 57

Figure 43 The carbonate content of the samples from the west profile 57

xvLIST OF FIGURES IN THE TEXT

Figure 44 Morphoscopy of sands without acid treatment 57

Figure 45 Morphoscopy of sands after the elimination of calcareous grains (subsequent to acid treatment) 57

Figure 51 Drawing of Triticum turgidum LT aestivum from the 1997 excavations at Knossos 69

Figure 52 Graphs of measurements and measurement ratios of Triticum turgidumaestivum from Neolithic Knossos 72

Figure 53 Measurements of Triticum turgidumaestivum from Aceramic and Early Neolithic Knossos compared with average values for Erbaba Ramad and Bouqras in the Near East 73

Figure 54 Graphs of measurements and measurement ratios of Lens culinaris from Neolithic Knossos 74

Figure 55 Early sites including those from mainland Greece where Triticum turgidumaestivum is reported 1 Tell Abu Hureyra 2 Tell Halula 3 Tell Aswad 4 Tell Ghoraife 5 Tell Sabi Abyad 6 Servia 7 Cafer Houmlyuumlk 8 Dhali Agridhi 9 Otzaki 10 Sesklo 11 Sitagroi 12 Haccedililar 13 Aşkli Houmlyuumlk 14 Ccedilatal Houmlyuumlk 15 Can Hasan 16 Cayoumlnuuml 17 El Kown 18 Bouqras 19 Tell Ramad 89

Figure 56 Summary of the distribution of all categories of archaeobotanical remains at Neolithic Knossos 90

Figure 61 Climate and topography of Knossos (a) mean annual precipitation in Crete (b) topographic map of the area around Knossos (c) westndasheast topographic section (d) southwestndashnortheast topographic section 96

Figure 62 View of the Knossos valley from Mt Juktas showing present-day vegetation 98

Figure 63 Panoramic view of the site of Knossos showing present-day vegetation 98

Figure 64 Present-day phrygana vegetation on the hills in the study area 98

Figure 65 Present-day vegetation on deep soils in the study area 98

Figure 66 Anatomy of plant taxa identified in wood charcoal assemblages from Neolithic Knossos 106

Figure 67 Wood charcoal diagram from Neolithic Knossos showing relative frequencies of taxa in successive excavation levels 108

Figure 71 Bar chart of phytolith per centage frequencies from the south profile 122

Figure 72 Bar chart of C3 and C

4 phytolith percentage frequencies from the south profile 122

Figure 73 West profile stratigraphy and sampling 123

Figure 74 Bar chart of phytolith percentage frequencies from the west profile 124

Figure 75 Bar chart of C3 and C

4 phytolith percentage frequencies from the west profile 125

Figure 76 Trench section (southwest corner to west face) with phases identified according to phytolith composition and frequencies 127

xvi THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 77 Silica skeleton from grass leaf (long cells and a stoma) from the EN I deposits (sample XXa level 32) 129

Figure 78 Wheat-type silica skeleton from the EN I deposits (sample XXIVb level 32) 129

Figure 79 Silica skeleton from a dicotyledonous plant from the EN I deposits (sample XIV level 30) 129

Figure 710 Millet-type silica skeleton from the EN II deposits (sample Xa level 16) 129

Figure 81 Percentages of the osseous parts of cattle long bones 139

Figure 82 Percentages of the osseous parts belonging to the long bones of middle-sized mammals (goats sheep and pigs) 139

Figure 83 Skeletal fragments of long bones of OvisCapra from level 14 all with dog-gnawing marks 140

Figure 84 Fragments of proximal epiphyses of femur and tibia of Bos taurus with fracture marks caused by impacts from the extraction of marrow level 24 151

Figure 85 Animal bones from level 3 (a) radius diaphysis (b) radius proximal part (c) scapula (d) fragment of femur (e) phalanx I of Capra aegagrus 153

Figure 86 Distal metacarpus of Capra aegagrus and Ovis aries 153

Figure 87 Animal bones (a) ulna in lateral view probably belonging to a wild boar (level 23) (b) Sus scrofa ferus canine fragment (level 10) (c) Sus scrofa domesticus ulna in lateral view (level 14) 154

Figure 88 Diaphysis width range (SD) of Sus scrofa domesticus and Sus scrofa ferus from Zambujal (Portugal) Cerro de la Virgen (Spain) Argissa-Magula and Knossos 154

Figure 89 Meles meles (a) left mandible in lateral view (level 14) (b) lower canine (level 14) (c) left ulna in medial views the proximal epiphysis is not fused (level 3) Martes (d) distal part of humerus in cranial view (level 9) 155

Figure 810 Age classes of the mandibles of Ovis and Capra 157

Figure 811 Age classes of the mandibles of Bos taurus 158

Figure 812 Distal part of metacarpus belonging to a male (possibly ox) of Bos taurus with osseous deformations on the articular surfaces 158

Figure 813 Age classes of Sus scrofa domesticus maxillae and mandibles 159

Figure 814 Correlation of the measurements of phalanx I belonging to Bos taurus 160

Figure 91 Map showing location of sites mentioned in the text 1 Ashkelon 2 lsquoAin Ghazal 3 Atlit Yam 4 Hagoshrim and Tel Ali 5 Ras Shamra 6 Cap Andreas Kastros 7 Khirokitia 8 Tenta 9 Asikli Houmlyuumlk 10 Mersin 11 Can Hasan III 12 Ccedilatalhoumlyuumlk 13 Suberde 14 Haccedililar 15 Nea Nikomedeia 16 Argissa-Magula 17 Sesklo 18 Achilleion 19 Franchthi Cave 20 Sidari Corfu 21 Cave of the Cyclops Youra 22 Melos 23 Santorini 24 Knossos Crete 25 Tel Aray 2 26 Umm el Tlel 27 Qdeir 28 El Kowm 2 173

Figure 92 Adult male agrimi (Capra aegagrus cretica) showing phenotypic resemblance to the wild bezoar goat 177

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

PREHISTORY MONOGRAPHS 42

The Neolithic Settlement of Knossos in Crete

New Evidence for the Early Occupation of Crete and the Aegean Islands

edited by

Nikos Efstratiou Alexandra Karetsou and Maria Ntinou

Published byINSTAP Academic Press

Philadelphia Pennsylvania2013

Design and ProductionINSTAP Academic Press Philadelphia PA

Printing and BindingHoster Bindery Inc Ivyland PA

Library of Congress Cataloging-in-Publication Data

The neolithic settlement of Knossos in Crete new evidence for the early occupation of Crete and the Aegean islands edited by Nikos Efstratiou Alexandra Karetsou and Maria Ntinou pages cm -- (Prehistory monographs 42) Includes bibliographical references and index ISBN 978-1-931534-72-7 (hardcover alk paper) 1 Knossos (Extinct city) 2 Neolithic period--Greece--Crete 3 Land settlement patterns Prehistoric--Greece--Crete 4 Crete (Greece)--Antiquities I Efstratiou Nicholas DF221C8N46 2013 939rsquo18--dc23

2013016076

Copyright copy 2013

INSTAP Academic PressPhiladelphia Pennsylvania

All rights reservedPrinted in the United States of America

To the memory of Professor JD Evans a gentleman of British Archaeology

ndashNikos Efstratiou

Table of Contents

List of Tables in the Text ix

List of Figures in the Text xiii

Preface Alexandra Karetsou xix

Acknowledgments xxiii

Introduction Nikos Efstratiou xxv

1 The Excavation Nikos Efstratiou Alexandra Karetsou and Eleni Banou 1

2 The Stratigraphy and Cultural Phases Nikos Efstratiou 25

3 Fabric Diversity in the Neolithic Ceramics of Knossos Sarantis Dimitriadis 47

4 Neolithic Sedimentology at Knossos Maria-Pilar Fumanal Garciacuteadagger 53

5 The Economy of Neolithic Knossos The Archaeobotanical Data Anaya Sarpaki 63

6 Wood Charcoal Analysis The Local Vegetation Ernestina Badal and Maria Ntinou 95

7 Plant Economy and the Use of Space Evidence from the Opal Phytoliths Marco Madella 119

8 The Knossos Fauna and the Beginning of the Neolithic in the Mediterranean Islands Manuel Peacuterez Ripoll 133

9 The Earliest Settlement on Crete An Archaeozoological Perspective Liora Kolska Horwitz 171

viii THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

10 Radiocarbon Dates from the Neolithic Settlement of Knossos An Overview Yorgos Facorellis and Yiannis Maniatis 193

11 Knossos and the Beginning of the Neolithic in Greece and the Aegean Islands Nikos Efstratiou 201

Index 215

Table 41 Correlation of sedimentology samples with excavation levels and cultural phases 55

Table 42 Munsell color and calculation of statistical parameters of mean size sorting skewness and kurtosis for each of the analyzed sedimentology samples 56

Table 51 Seed list provided to JD Evans by Hans Helbaek (unpublished) 67

Table 52 List of archaeobotanical samples from the 1997 rescue excavation along with relative and absolute dates 67

Table 53 Aceramic Neolithic archaeobotanical sample E 97(30) from Knossos 1997 level 39 retrieved from 16 liters of water-floated soil 68

Table 54 Measurements of Triticum turgidumaestivum from Aceramic and EN levels at Knossos 71

Table 55 Measurements of Triticum dicoccum Triticum monococcum Hordeum vulgare and Lens culinaris 74

Table 56 Early Neolithic I archaeobotanical (seed) samples 75

Table 57 Measurements of Trifolium spp and Leguminosae 78

Table 58 Measurements of Raphanus cf raphanistrum and Linum cf usitatissimum 78

Table 59 Early Neolithic II archaeobotanical (seed) samples 81

List of Tables in the Text

x THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Table 510 Vitis sp measurements from EN II levels sketch of a grape seed showing locations of dimensions 85

Table 511 Middle Neolithic archaeobotanical (seed) samples 86

Table 512 Late Neolithic archaeobotanical (seed) samples 88

Table 61 Inventories of plants growing in different parts of the study area 99

Table 62 Absolute and relative frequencies of taxa identified in the wood charcoal assemblages from Neolithic Knossos 102

Table 63 Presence of plant taxa in wood charcoal assemblages from Neolithic Knossos along with the total number of fragments analyzed and the total number of taxa identified in each level 104

Table 71 Knossos 1997 south profile phytolith counts 122

Table 72 Knossos 1997 west profile phytolith counts 125

Table 81 Number of identified and unidentified specimens by taxa and period 135

Table 82 Measurements of bones from Bos taurus 135

Table 83 Measurements of bones from Ovis aries (Oa) and Capra hircus (Ch) 136

Table 84 Measurements of bones from Sus scrofa domesticus Sus scrofa ferus Capra aegagrus Martes and Meles meles 138

Table 85 Number of identified specimens of Bos OvisCapra and Sus with number of marks caused by dog gnawing 140

Table 86 Early Neolithic I and EN II faunal remains 141

Table 87 Early Neolithic IIMN faunal remains 142

Table 88 Middle Neolithic faunal remains 144

Table 89A Late Neolithic faunal remains 146

Table 89B Late Neolithic faunal remains 149

Table 810 Number of identified specimens of Bos and OvisCapraSus with burn marks 151

Table 811 Number of long bone remains (the diaphysis fragments are not counted here) phalanges and tarsi corresponding to mature and immature bones along with the number of LN tooth remains grouped by age for comparison with the long bones 156

Table 812 Number of mandibles (NM) for goats and sheep from the LN levels classified by age 157

Table 813 Number of mandibles (NM) of Bos taurus classified by age 158

Table 814 Number of maxillary and mandibular remains of Sus scrofa domesticus classified by age 159

Table 815 Number of identified specimens of Bos taurus classified by sex 160

Table 816 Number of identified specimens of Ovis aries (Oa) and Capra hircus (Ch) classified by sex 160

xiLIST OF TABLES IN THE TEXT

Table 817 Chronological representation of the faunal species at Knossos 161

Table 818 Percentages of identified specimens of domestic and wild species at Knossos and other sites 161

Table 819 Representation and abundance of various faunal species at Shillourokambos Ais Yiorkis and Khirokitia 162

Table 820 Introduction and chronological representation of wild animals at various sites in Crete 163

Table 91 Relative frequencies (percentages) of animal species from Knossos 175

Table 92 Schematic representation of the relative chronology (cal bc dates) of sites mentioned in the text 183

Table 101 Summary of the British Museum radiocarbon dates on charcoal from the excavations of JD Evans at Neolithic Knossos sorted by age 194

Table 102 Description of the samples dated in the British Museum Research Laboratory 195

Table 103 Summary of radiocarbon dating results of carbonized samples collected in 1997 from the Neolithic settlement levels at Knossos 196

Frontispiece The city the fortifications the harbor and the hinterland of Khandax (Herakleion) in the first half of the 17th century Map by unknown cartographer 17th c Collezione Museo Civico Padua Vikelaia Municipal Library Herakleion ii

Figure i The Minoan palace and its Neolithic past xxi

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation 2

Figure 12 Trench II (a) view of the Central Court of the Palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the souther profile 3

Figure 13 Plan of the excavation trenches next to the staircase 3

Figure 14 South and west stratigraphic profiles of the trench 4

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench 5

Figure 16 Plans of excavation levels 9 and 10 7

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner 8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner 8

List of Figures in the Text

xiv THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 19 Plan of excavation level 13 9

Figure 110 Plan of excavation level 14 showing hearth in northwest corner 9

Figure 111 View of excavation level 14 showing hearth in northwest corner 10

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south 11

Figure 113 View of excavation level 15 from above 12

Figure 114 View of excavation level 16 from above 12

Figure 115 View of excavation level 16 facing west section 12

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4 12

Figure 117 View of level 17 facing west section 13

Figure 118 View of excavation level 18 from above 13

Figure 119 View of excavation level 19 from above 13

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6 13

Figure 121 View of excavation level 21 from above 14

Figure 122 View of excavation level 23 14

Figure 123 View of excavation level 24 showing wall 7 and grinding stones 14

Figure 124 Plans of excavation levels 22 and 24 15

Figure 125 Plan of excavation level 27 16

Figure 126 View of excavation level 28 from above 16

Figure 127 View of excavation level 28 facing west profile 16

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths 17

Figure 129 View of level 29A showing hearths 1 2 and 3 18

Figure 130 View of excavation level 30 showing hearth 4 18

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure 19

Figure 132 View of the elliptical stone wall from levels 24ndash27 19

Figure 133 View of excavation level 31 19

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7 20

Figure 135 Plan of excavation level 37 showing pits 1 and 2 21

Figure 21 Sedimentological samples of the middle part of the south profile 26

Figure 41 Fine fraction granulometry () of the samples from the west profile 57

Figure 42 The organic content () of the samples from the west profile 57

Figure 43 The carbonate content of the samples from the west profile 57

xvLIST OF FIGURES IN THE TEXT

Figure 44 Morphoscopy of sands without acid treatment 57

Figure 45 Morphoscopy of sands after the elimination of calcareous grains (subsequent to acid treatment) 57

Figure 51 Drawing of Triticum turgidum LT aestivum from the 1997 excavations at Knossos 69

Figure 52 Graphs of measurements and measurement ratios of Triticum turgidumaestivum from Neolithic Knossos 72

Figure 53 Measurements of Triticum turgidumaestivum from Aceramic and Early Neolithic Knossos compared with average values for Erbaba Ramad and Bouqras in the Near East 73

Figure 54 Graphs of measurements and measurement ratios of Lens culinaris from Neolithic Knossos 74

Figure 55 Early sites including those from mainland Greece where Triticum turgidumaestivum is reported 1 Tell Abu Hureyra 2 Tell Halula 3 Tell Aswad 4 Tell Ghoraife 5 Tell Sabi Abyad 6 Servia 7 Cafer Houmlyuumlk 8 Dhali Agridhi 9 Otzaki 10 Sesklo 11 Sitagroi 12 Haccedililar 13 Aşkli Houmlyuumlk 14 Ccedilatal Houmlyuumlk 15 Can Hasan 16 Cayoumlnuuml 17 El Kown 18 Bouqras 19 Tell Ramad 89

Figure 56 Summary of the distribution of all categories of archaeobotanical remains at Neolithic Knossos 90

Figure 61 Climate and topography of Knossos (a) mean annual precipitation in Crete (b) topographic map of the area around Knossos (c) westndasheast topographic section (d) southwestndashnortheast topographic section 96

Figure 62 View of the Knossos valley from Mt Juktas showing present-day vegetation 98

Figure 63 Panoramic view of the site of Knossos showing present-day vegetation 98

Figure 64 Present-day phrygana vegetation on the hills in the study area 98

Figure 65 Present-day vegetation on deep soils in the study area 98

Figure 66 Anatomy of plant taxa identified in wood charcoal assemblages from Neolithic Knossos 106

Figure 67 Wood charcoal diagram from Neolithic Knossos showing relative frequencies of taxa in successive excavation levels 108

Figure 71 Bar chart of phytolith per centage frequencies from the south profile 122

Figure 72 Bar chart of C3 and C

4 phytolith percentage frequencies from the south profile 122

Figure 73 West profile stratigraphy and sampling 123

Figure 74 Bar chart of phytolith percentage frequencies from the west profile 124

Figure 75 Bar chart of C3 and C

4 phytolith percentage frequencies from the west profile 125

Figure 76 Trench section (southwest corner to west face) with phases identified according to phytolith composition and frequencies 127

xvi THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 77 Silica skeleton from grass leaf (long cells and a stoma) from the EN I deposits (sample XXa level 32) 129

Figure 78 Wheat-type silica skeleton from the EN I deposits (sample XXIVb level 32) 129

Figure 79 Silica skeleton from a dicotyledonous plant from the EN I deposits (sample XIV level 30) 129

Figure 710 Millet-type silica skeleton from the EN II deposits (sample Xa level 16) 129

Figure 81 Percentages of the osseous parts of cattle long bones 139

Figure 82 Percentages of the osseous parts belonging to the long bones of middle-sized mammals (goats sheep and pigs) 139

Figure 83 Skeletal fragments of long bones of OvisCapra from level 14 all with dog-gnawing marks 140

Figure 84 Fragments of proximal epiphyses of femur and tibia of Bos taurus with fracture marks caused by impacts from the extraction of marrow level 24 151

Figure 85 Animal bones from level 3 (a) radius diaphysis (b) radius proximal part (c) scapula (d) fragment of femur (e) phalanx I of Capra aegagrus 153

Figure 86 Distal metacarpus of Capra aegagrus and Ovis aries 153

Figure 87 Animal bones (a) ulna in lateral view probably belonging to a wild boar (level 23) (b) Sus scrofa ferus canine fragment (level 10) (c) Sus scrofa domesticus ulna in lateral view (level 14) 154

Figure 88 Diaphysis width range (SD) of Sus scrofa domesticus and Sus scrofa ferus from Zambujal (Portugal) Cerro de la Virgen (Spain) Argissa-Magula and Knossos 154

Figure 89 Meles meles (a) left mandible in lateral view (level 14) (b) lower canine (level 14) (c) left ulna in medial views the proximal epiphysis is not fused (level 3) Martes (d) distal part of humerus in cranial view (level 9) 155

Figure 810 Age classes of the mandibles of Ovis and Capra 157

Figure 811 Age classes of the mandibles of Bos taurus 158

Figure 812 Distal part of metacarpus belonging to a male (possibly ox) of Bos taurus with osseous deformations on the articular surfaces 158

Figure 813 Age classes of Sus scrofa domesticus maxillae and mandibles 159

Figure 814 Correlation of the measurements of phalanx I belonging to Bos taurus 160

Figure 91 Map showing location of sites mentioned in the text 1 Ashkelon 2 lsquoAin Ghazal 3 Atlit Yam 4 Hagoshrim and Tel Ali 5 Ras Shamra 6 Cap Andreas Kastros 7 Khirokitia 8 Tenta 9 Asikli Houmlyuumlk 10 Mersin 11 Can Hasan III 12 Ccedilatalhoumlyuumlk 13 Suberde 14 Haccedililar 15 Nea Nikomedeia 16 Argissa-Magula 17 Sesklo 18 Achilleion 19 Franchthi Cave 20 Sidari Corfu 21 Cave of the Cyclops Youra 22 Melos 23 Santorini 24 Knossos Crete 25 Tel Aray 2 26 Umm el Tlel 27 Qdeir 28 El Kowm 2 173

Figure 92 Adult male agrimi (Capra aegagrus cretica) showing phenotypic resemblance to the wild bezoar goat 177

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

Design and ProductionINSTAP Academic Press Philadelphia PA

Printing and BindingHoster Bindery Inc Ivyland PA

Library of Congress Cataloging-in-Publication Data

The neolithic settlement of Knossos in Crete new evidence for the early occupation of Crete and the Aegean islands edited by Nikos Efstratiou Alexandra Karetsou and Maria Ntinou pages cm -- (Prehistory monographs 42) Includes bibliographical references and index ISBN 978-1-931534-72-7 (hardcover alk paper) 1 Knossos (Extinct city) 2 Neolithic period--Greece--Crete 3 Land settlement patterns Prehistoric--Greece--Crete 4 Crete (Greece)--Antiquities I Efstratiou Nicholas DF221C8N46 2013 939rsquo18--dc23

2013016076

Copyright copy 2013

INSTAP Academic PressPhiladelphia Pennsylvania

All rights reservedPrinted in the United States of America

To the memory of Professor JD Evans a gentleman of British Archaeology

ndashNikos Efstratiou

Table of Contents

List of Tables in the Text ix

List of Figures in the Text xiii

Preface Alexandra Karetsou xix

Acknowledgments xxiii

Introduction Nikos Efstratiou xxv

1 The Excavation Nikos Efstratiou Alexandra Karetsou and Eleni Banou 1

2 The Stratigraphy and Cultural Phases Nikos Efstratiou 25

3 Fabric Diversity in the Neolithic Ceramics of Knossos Sarantis Dimitriadis 47

4 Neolithic Sedimentology at Knossos Maria-Pilar Fumanal Garciacuteadagger 53

5 The Economy of Neolithic Knossos The Archaeobotanical Data Anaya Sarpaki 63

6 Wood Charcoal Analysis The Local Vegetation Ernestina Badal and Maria Ntinou 95

7 Plant Economy and the Use of Space Evidence from the Opal Phytoliths Marco Madella 119

8 The Knossos Fauna and the Beginning of the Neolithic in the Mediterranean Islands Manuel Peacuterez Ripoll 133

9 The Earliest Settlement on Crete An Archaeozoological Perspective Liora Kolska Horwitz 171

viii THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

10 Radiocarbon Dates from the Neolithic Settlement of Knossos An Overview Yorgos Facorellis and Yiannis Maniatis 193

11 Knossos and the Beginning of the Neolithic in Greece and the Aegean Islands Nikos Efstratiou 201

Index 215

Table 41 Correlation of sedimentology samples with excavation levels and cultural phases 55

Table 42 Munsell color and calculation of statistical parameters of mean size sorting skewness and kurtosis for each of the analyzed sedimentology samples 56

Table 51 Seed list provided to JD Evans by Hans Helbaek (unpublished) 67

Table 52 List of archaeobotanical samples from the 1997 rescue excavation along with relative and absolute dates 67

Table 53 Aceramic Neolithic archaeobotanical sample E 97(30) from Knossos 1997 level 39 retrieved from 16 liters of water-floated soil 68

Table 54 Measurements of Triticum turgidumaestivum from Aceramic and EN levels at Knossos 71

Table 55 Measurements of Triticum dicoccum Triticum monococcum Hordeum vulgare and Lens culinaris 74

Table 56 Early Neolithic I archaeobotanical (seed) samples 75

Table 57 Measurements of Trifolium spp and Leguminosae 78

Table 58 Measurements of Raphanus cf raphanistrum and Linum cf usitatissimum 78

Table 59 Early Neolithic II archaeobotanical (seed) samples 81

List of Tables in the Text

x THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Table 510 Vitis sp measurements from EN II levels sketch of a grape seed showing locations of dimensions 85

Table 511 Middle Neolithic archaeobotanical (seed) samples 86

Table 512 Late Neolithic archaeobotanical (seed) samples 88

Table 61 Inventories of plants growing in different parts of the study area 99

Table 62 Absolute and relative frequencies of taxa identified in the wood charcoal assemblages from Neolithic Knossos 102

Table 63 Presence of plant taxa in wood charcoal assemblages from Neolithic Knossos along with the total number of fragments analyzed and the total number of taxa identified in each level 104

Table 71 Knossos 1997 south profile phytolith counts 122

Table 72 Knossos 1997 west profile phytolith counts 125

Table 81 Number of identified and unidentified specimens by taxa and period 135

Table 82 Measurements of bones from Bos taurus 135

Table 83 Measurements of bones from Ovis aries (Oa) and Capra hircus (Ch) 136

Table 84 Measurements of bones from Sus scrofa domesticus Sus scrofa ferus Capra aegagrus Martes and Meles meles 138

Table 85 Number of identified specimens of Bos OvisCapra and Sus with number of marks caused by dog gnawing 140

Table 86 Early Neolithic I and EN II faunal remains 141

Table 87 Early Neolithic IIMN faunal remains 142

Table 88 Middle Neolithic faunal remains 144

Table 89A Late Neolithic faunal remains 146

Table 89B Late Neolithic faunal remains 149

Table 810 Number of identified specimens of Bos and OvisCapraSus with burn marks 151

Table 811 Number of long bone remains (the diaphysis fragments are not counted here) phalanges and tarsi corresponding to mature and immature bones along with the number of LN tooth remains grouped by age for comparison with the long bones 156

Table 812 Number of mandibles (NM) for goats and sheep from the LN levels classified by age 157

Table 813 Number of mandibles (NM) of Bos taurus classified by age 158

Table 814 Number of maxillary and mandibular remains of Sus scrofa domesticus classified by age 159

Table 815 Number of identified specimens of Bos taurus classified by sex 160

Table 816 Number of identified specimens of Ovis aries (Oa) and Capra hircus (Ch) classified by sex 160

xiLIST OF TABLES IN THE TEXT

Table 817 Chronological representation of the faunal species at Knossos 161

Table 818 Percentages of identified specimens of domestic and wild species at Knossos and other sites 161

Table 819 Representation and abundance of various faunal species at Shillourokambos Ais Yiorkis and Khirokitia 162

Table 820 Introduction and chronological representation of wild animals at various sites in Crete 163

Table 91 Relative frequencies (percentages) of animal species from Knossos 175

Table 92 Schematic representation of the relative chronology (cal bc dates) of sites mentioned in the text 183

Table 101 Summary of the British Museum radiocarbon dates on charcoal from the excavations of JD Evans at Neolithic Knossos sorted by age 194

Table 102 Description of the samples dated in the British Museum Research Laboratory 195

Table 103 Summary of radiocarbon dating results of carbonized samples collected in 1997 from the Neolithic settlement levels at Knossos 196

Frontispiece The city the fortifications the harbor and the hinterland of Khandax (Herakleion) in the first half of the 17th century Map by unknown cartographer 17th c Collezione Museo Civico Padua Vikelaia Municipal Library Herakleion ii

Figure i The Minoan palace and its Neolithic past xxi

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation 2

Figure 12 Trench II (a) view of the Central Court of the Palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the souther profile 3

Figure 13 Plan of the excavation trenches next to the staircase 3

Figure 14 South and west stratigraphic profiles of the trench 4

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench 5

Figure 16 Plans of excavation levels 9 and 10 7

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner 8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner 8

List of Figures in the Text

xiv THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 19 Plan of excavation level 13 9

Figure 110 Plan of excavation level 14 showing hearth in northwest corner 9

Figure 111 View of excavation level 14 showing hearth in northwest corner 10

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south 11

Figure 113 View of excavation level 15 from above 12

Figure 114 View of excavation level 16 from above 12

Figure 115 View of excavation level 16 facing west section 12

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4 12

Figure 117 View of level 17 facing west section 13

Figure 118 View of excavation level 18 from above 13

Figure 119 View of excavation level 19 from above 13

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6 13

Figure 121 View of excavation level 21 from above 14

Figure 122 View of excavation level 23 14

Figure 123 View of excavation level 24 showing wall 7 and grinding stones 14

Figure 124 Plans of excavation levels 22 and 24 15

Figure 125 Plan of excavation level 27 16

Figure 126 View of excavation level 28 from above 16

Figure 127 View of excavation level 28 facing west profile 16

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths 17

Figure 129 View of level 29A showing hearths 1 2 and 3 18

Figure 130 View of excavation level 30 showing hearth 4 18

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure 19

Figure 132 View of the elliptical stone wall from levels 24ndash27 19

Figure 133 View of excavation level 31 19

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7 20

Figure 135 Plan of excavation level 37 showing pits 1 and 2 21

Figure 21 Sedimentological samples of the middle part of the south profile 26

Figure 41 Fine fraction granulometry () of the samples from the west profile 57

Figure 42 The organic content () of the samples from the west profile 57

Figure 43 The carbonate content of the samples from the west profile 57

xvLIST OF FIGURES IN THE TEXT

Figure 44 Morphoscopy of sands without acid treatment 57

Figure 45 Morphoscopy of sands after the elimination of calcareous grains (subsequent to acid treatment) 57

Figure 51 Drawing of Triticum turgidum LT aestivum from the 1997 excavations at Knossos 69

Figure 52 Graphs of measurements and measurement ratios of Triticum turgidumaestivum from Neolithic Knossos 72

Figure 53 Measurements of Triticum turgidumaestivum from Aceramic and Early Neolithic Knossos compared with average values for Erbaba Ramad and Bouqras in the Near East 73

Figure 54 Graphs of measurements and measurement ratios of Lens culinaris from Neolithic Knossos 74

Figure 55 Early sites including those from mainland Greece where Triticum turgidumaestivum is reported 1 Tell Abu Hureyra 2 Tell Halula 3 Tell Aswad 4 Tell Ghoraife 5 Tell Sabi Abyad 6 Servia 7 Cafer Houmlyuumlk 8 Dhali Agridhi 9 Otzaki 10 Sesklo 11 Sitagroi 12 Haccedililar 13 Aşkli Houmlyuumlk 14 Ccedilatal Houmlyuumlk 15 Can Hasan 16 Cayoumlnuuml 17 El Kown 18 Bouqras 19 Tell Ramad 89

Figure 56 Summary of the distribution of all categories of archaeobotanical remains at Neolithic Knossos 90

Figure 61 Climate and topography of Knossos (a) mean annual precipitation in Crete (b) topographic map of the area around Knossos (c) westndasheast topographic section (d) southwestndashnortheast topographic section 96

Figure 62 View of the Knossos valley from Mt Juktas showing present-day vegetation 98

Figure 63 Panoramic view of the site of Knossos showing present-day vegetation 98

Figure 64 Present-day phrygana vegetation on the hills in the study area 98

Figure 65 Present-day vegetation on deep soils in the study area 98

Figure 66 Anatomy of plant taxa identified in wood charcoal assemblages from Neolithic Knossos 106

Figure 67 Wood charcoal diagram from Neolithic Knossos showing relative frequencies of taxa in successive excavation levels 108

Figure 71 Bar chart of phytolith per centage frequencies from the south profile 122

Figure 72 Bar chart of C3 and C

4 phytolith percentage frequencies from the south profile 122

Figure 73 West profile stratigraphy and sampling 123

Figure 74 Bar chart of phytolith percentage frequencies from the west profile 124

Figure 75 Bar chart of C3 and C

4 phytolith percentage frequencies from the west profile 125

Figure 76 Trench section (southwest corner to west face) with phases identified according to phytolith composition and frequencies 127

xvi THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 77 Silica skeleton from grass leaf (long cells and a stoma) from the EN I deposits (sample XXa level 32) 129

Figure 78 Wheat-type silica skeleton from the EN I deposits (sample XXIVb level 32) 129

Figure 79 Silica skeleton from a dicotyledonous plant from the EN I deposits (sample XIV level 30) 129

Figure 710 Millet-type silica skeleton from the EN II deposits (sample Xa level 16) 129

Figure 81 Percentages of the osseous parts of cattle long bones 139

Figure 82 Percentages of the osseous parts belonging to the long bones of middle-sized mammals (goats sheep and pigs) 139

Figure 83 Skeletal fragments of long bones of OvisCapra from level 14 all with dog-gnawing marks 140

Figure 84 Fragments of proximal epiphyses of femur and tibia of Bos taurus with fracture marks caused by impacts from the extraction of marrow level 24 151

Figure 85 Animal bones from level 3 (a) radius diaphysis (b) radius proximal part (c) scapula (d) fragment of femur (e) phalanx I of Capra aegagrus 153

Figure 86 Distal metacarpus of Capra aegagrus and Ovis aries 153

Figure 87 Animal bones (a) ulna in lateral view probably belonging to a wild boar (level 23) (b) Sus scrofa ferus canine fragment (level 10) (c) Sus scrofa domesticus ulna in lateral view (level 14) 154

Figure 88 Diaphysis width range (SD) of Sus scrofa domesticus and Sus scrofa ferus from Zambujal (Portugal) Cerro de la Virgen (Spain) Argissa-Magula and Knossos 154

Figure 89 Meles meles (a) left mandible in lateral view (level 14) (b) lower canine (level 14) (c) left ulna in medial views the proximal epiphysis is not fused (level 3) Martes (d) distal part of humerus in cranial view (level 9) 155

Figure 810 Age classes of the mandibles of Ovis and Capra 157

Figure 811 Age classes of the mandibles of Bos taurus 158

Figure 812 Distal part of metacarpus belonging to a male (possibly ox) of Bos taurus with osseous deformations on the articular surfaces 158

Figure 813 Age classes of Sus scrofa domesticus maxillae and mandibles 159

Figure 814 Correlation of the measurements of phalanx I belonging to Bos taurus 160

Figure 91 Map showing location of sites mentioned in the text 1 Ashkelon 2 lsquoAin Ghazal 3 Atlit Yam 4 Hagoshrim and Tel Ali 5 Ras Shamra 6 Cap Andreas Kastros 7 Khirokitia 8 Tenta 9 Asikli Houmlyuumlk 10 Mersin 11 Can Hasan III 12 Ccedilatalhoumlyuumlk 13 Suberde 14 Haccedililar 15 Nea Nikomedeia 16 Argissa-Magula 17 Sesklo 18 Achilleion 19 Franchthi Cave 20 Sidari Corfu 21 Cave of the Cyclops Youra 22 Melos 23 Santorini 24 Knossos Crete 25 Tel Aray 2 26 Umm el Tlel 27 Qdeir 28 El Kowm 2 173

Figure 92 Adult male agrimi (Capra aegagrus cretica) showing phenotypic resemblance to the wild bezoar goat 177

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

To the memory of Professor JD Evans a gentleman of British Archaeology

ndashNikos Efstratiou

Table of Contents

List of Tables in the Text ix

List of Figures in the Text xiii

Preface Alexandra Karetsou xix

Acknowledgments xxiii

Introduction Nikos Efstratiou xxv

1 The Excavation Nikos Efstratiou Alexandra Karetsou and Eleni Banou 1

2 The Stratigraphy and Cultural Phases Nikos Efstratiou 25

3 Fabric Diversity in the Neolithic Ceramics of Knossos Sarantis Dimitriadis 47

4 Neolithic Sedimentology at Knossos Maria-Pilar Fumanal Garciacuteadagger 53

5 The Economy of Neolithic Knossos The Archaeobotanical Data Anaya Sarpaki 63

6 Wood Charcoal Analysis The Local Vegetation Ernestina Badal and Maria Ntinou 95

7 Plant Economy and the Use of Space Evidence from the Opal Phytoliths Marco Madella 119

8 The Knossos Fauna and the Beginning of the Neolithic in the Mediterranean Islands Manuel Peacuterez Ripoll 133

9 The Earliest Settlement on Crete An Archaeozoological Perspective Liora Kolska Horwitz 171

viii THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

10 Radiocarbon Dates from the Neolithic Settlement of Knossos An Overview Yorgos Facorellis and Yiannis Maniatis 193

11 Knossos and the Beginning of the Neolithic in Greece and the Aegean Islands Nikos Efstratiou 201

Index 215

Table 41 Correlation of sedimentology samples with excavation levels and cultural phases 55

Table 42 Munsell color and calculation of statistical parameters of mean size sorting skewness and kurtosis for each of the analyzed sedimentology samples 56

Table 51 Seed list provided to JD Evans by Hans Helbaek (unpublished) 67

Table 52 List of archaeobotanical samples from the 1997 rescue excavation along with relative and absolute dates 67

Table 53 Aceramic Neolithic archaeobotanical sample E 97(30) from Knossos 1997 level 39 retrieved from 16 liters of water-floated soil 68

Table 54 Measurements of Triticum turgidumaestivum from Aceramic and EN levels at Knossos 71

Table 55 Measurements of Triticum dicoccum Triticum monococcum Hordeum vulgare and Lens culinaris 74

Table 56 Early Neolithic I archaeobotanical (seed) samples 75

Table 57 Measurements of Trifolium spp and Leguminosae 78

Table 58 Measurements of Raphanus cf raphanistrum and Linum cf usitatissimum 78

Table 59 Early Neolithic II archaeobotanical (seed) samples 81

List of Tables in the Text

x THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Table 510 Vitis sp measurements from EN II levels sketch of a grape seed showing locations of dimensions 85

Table 511 Middle Neolithic archaeobotanical (seed) samples 86

Table 512 Late Neolithic archaeobotanical (seed) samples 88

Table 61 Inventories of plants growing in different parts of the study area 99

Table 62 Absolute and relative frequencies of taxa identified in the wood charcoal assemblages from Neolithic Knossos 102

Table 63 Presence of plant taxa in wood charcoal assemblages from Neolithic Knossos along with the total number of fragments analyzed and the total number of taxa identified in each level 104

Table 71 Knossos 1997 south profile phytolith counts 122

Table 72 Knossos 1997 west profile phytolith counts 125

Table 81 Number of identified and unidentified specimens by taxa and period 135

Table 82 Measurements of bones from Bos taurus 135

Table 83 Measurements of bones from Ovis aries (Oa) and Capra hircus (Ch) 136

Table 84 Measurements of bones from Sus scrofa domesticus Sus scrofa ferus Capra aegagrus Martes and Meles meles 138

Table 85 Number of identified specimens of Bos OvisCapra and Sus with number of marks caused by dog gnawing 140

Table 86 Early Neolithic I and EN II faunal remains 141

Table 87 Early Neolithic IIMN faunal remains 142

Table 88 Middle Neolithic faunal remains 144

Table 89A Late Neolithic faunal remains 146

Table 89B Late Neolithic faunal remains 149

Table 810 Number of identified specimens of Bos and OvisCapraSus with burn marks 151

Table 811 Number of long bone remains (the diaphysis fragments are not counted here) phalanges and tarsi corresponding to mature and immature bones along with the number of LN tooth remains grouped by age for comparison with the long bones 156

Table 812 Number of mandibles (NM) for goats and sheep from the LN levels classified by age 157

Table 813 Number of mandibles (NM) of Bos taurus classified by age 158

Table 814 Number of maxillary and mandibular remains of Sus scrofa domesticus classified by age 159

Table 815 Number of identified specimens of Bos taurus classified by sex 160

Table 816 Number of identified specimens of Ovis aries (Oa) and Capra hircus (Ch) classified by sex 160

xiLIST OF TABLES IN THE TEXT

Table 817 Chronological representation of the faunal species at Knossos 161

Table 818 Percentages of identified specimens of domestic and wild species at Knossos and other sites 161

Table 819 Representation and abundance of various faunal species at Shillourokambos Ais Yiorkis and Khirokitia 162

Table 820 Introduction and chronological representation of wild animals at various sites in Crete 163

Table 91 Relative frequencies (percentages) of animal species from Knossos 175

Table 92 Schematic representation of the relative chronology (cal bc dates) of sites mentioned in the text 183

Table 101 Summary of the British Museum radiocarbon dates on charcoal from the excavations of JD Evans at Neolithic Knossos sorted by age 194

Table 102 Description of the samples dated in the British Museum Research Laboratory 195

Table 103 Summary of radiocarbon dating results of carbonized samples collected in 1997 from the Neolithic settlement levels at Knossos 196

Frontispiece The city the fortifications the harbor and the hinterland of Khandax (Herakleion) in the first half of the 17th century Map by unknown cartographer 17th c Collezione Museo Civico Padua Vikelaia Municipal Library Herakleion ii

Figure i The Minoan palace and its Neolithic past xxi

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation 2

Figure 12 Trench II (a) view of the Central Court of the Palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the souther profile 3

Figure 13 Plan of the excavation trenches next to the staircase 3

Figure 14 South and west stratigraphic profiles of the trench 4

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench 5

Figure 16 Plans of excavation levels 9 and 10 7

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner 8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner 8

List of Figures in the Text

xiv THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 19 Plan of excavation level 13 9

Figure 110 Plan of excavation level 14 showing hearth in northwest corner 9

Figure 111 View of excavation level 14 showing hearth in northwest corner 10

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south 11

Figure 113 View of excavation level 15 from above 12

Figure 114 View of excavation level 16 from above 12

Figure 115 View of excavation level 16 facing west section 12

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4 12

Figure 117 View of level 17 facing west section 13

Figure 118 View of excavation level 18 from above 13

Figure 119 View of excavation level 19 from above 13

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6 13

Figure 121 View of excavation level 21 from above 14

Figure 122 View of excavation level 23 14

Figure 123 View of excavation level 24 showing wall 7 and grinding stones 14

Figure 124 Plans of excavation levels 22 and 24 15

Figure 125 Plan of excavation level 27 16

Figure 126 View of excavation level 28 from above 16

Figure 127 View of excavation level 28 facing west profile 16

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths 17

Figure 129 View of level 29A showing hearths 1 2 and 3 18

Figure 130 View of excavation level 30 showing hearth 4 18

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure 19

Figure 132 View of the elliptical stone wall from levels 24ndash27 19

Figure 133 View of excavation level 31 19

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7 20

Figure 135 Plan of excavation level 37 showing pits 1 and 2 21

Figure 21 Sedimentological samples of the middle part of the south profile 26

Figure 41 Fine fraction granulometry () of the samples from the west profile 57

Figure 42 The organic content () of the samples from the west profile 57

Figure 43 The carbonate content of the samples from the west profile 57

xvLIST OF FIGURES IN THE TEXT

Figure 44 Morphoscopy of sands without acid treatment 57

Figure 45 Morphoscopy of sands after the elimination of calcareous grains (subsequent to acid treatment) 57

Figure 51 Drawing of Triticum turgidum LT aestivum from the 1997 excavations at Knossos 69

Figure 52 Graphs of measurements and measurement ratios of Triticum turgidumaestivum from Neolithic Knossos 72

Figure 53 Measurements of Triticum turgidumaestivum from Aceramic and Early Neolithic Knossos compared with average values for Erbaba Ramad and Bouqras in the Near East 73

Figure 54 Graphs of measurements and measurement ratios of Lens culinaris from Neolithic Knossos 74

Figure 55 Early sites including those from mainland Greece where Triticum turgidumaestivum is reported 1 Tell Abu Hureyra 2 Tell Halula 3 Tell Aswad 4 Tell Ghoraife 5 Tell Sabi Abyad 6 Servia 7 Cafer Houmlyuumlk 8 Dhali Agridhi 9 Otzaki 10 Sesklo 11 Sitagroi 12 Haccedililar 13 Aşkli Houmlyuumlk 14 Ccedilatal Houmlyuumlk 15 Can Hasan 16 Cayoumlnuuml 17 El Kown 18 Bouqras 19 Tell Ramad 89

Figure 56 Summary of the distribution of all categories of archaeobotanical remains at Neolithic Knossos 90

Figure 61 Climate and topography of Knossos (a) mean annual precipitation in Crete (b) topographic map of the area around Knossos (c) westndasheast topographic section (d) southwestndashnortheast topographic section 96

Figure 62 View of the Knossos valley from Mt Juktas showing present-day vegetation 98

Figure 63 Panoramic view of the site of Knossos showing present-day vegetation 98

Figure 64 Present-day phrygana vegetation on the hills in the study area 98

Figure 65 Present-day vegetation on deep soils in the study area 98

Figure 66 Anatomy of plant taxa identified in wood charcoal assemblages from Neolithic Knossos 106

Figure 67 Wood charcoal diagram from Neolithic Knossos showing relative frequencies of taxa in successive excavation levels 108

Figure 71 Bar chart of phytolith per centage frequencies from the south profile 122

Figure 72 Bar chart of C3 and C

4 phytolith percentage frequencies from the south profile 122

Figure 73 West profile stratigraphy and sampling 123

Figure 74 Bar chart of phytolith percentage frequencies from the west profile 124

Figure 75 Bar chart of C3 and C

4 phytolith percentage frequencies from the west profile 125

Figure 76 Trench section (southwest corner to west face) with phases identified according to phytolith composition and frequencies 127

xvi THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 77 Silica skeleton from grass leaf (long cells and a stoma) from the EN I deposits (sample XXa level 32) 129

Figure 78 Wheat-type silica skeleton from the EN I deposits (sample XXIVb level 32) 129

Figure 79 Silica skeleton from a dicotyledonous plant from the EN I deposits (sample XIV level 30) 129

Figure 710 Millet-type silica skeleton from the EN II deposits (sample Xa level 16) 129

Figure 81 Percentages of the osseous parts of cattle long bones 139

Figure 82 Percentages of the osseous parts belonging to the long bones of middle-sized mammals (goats sheep and pigs) 139

Figure 83 Skeletal fragments of long bones of OvisCapra from level 14 all with dog-gnawing marks 140

Figure 84 Fragments of proximal epiphyses of femur and tibia of Bos taurus with fracture marks caused by impacts from the extraction of marrow level 24 151

Figure 85 Animal bones from level 3 (a) radius diaphysis (b) radius proximal part (c) scapula (d) fragment of femur (e) phalanx I of Capra aegagrus 153

Figure 86 Distal metacarpus of Capra aegagrus and Ovis aries 153

Figure 87 Animal bones (a) ulna in lateral view probably belonging to a wild boar (level 23) (b) Sus scrofa ferus canine fragment (level 10) (c) Sus scrofa domesticus ulna in lateral view (level 14) 154

Figure 88 Diaphysis width range (SD) of Sus scrofa domesticus and Sus scrofa ferus from Zambujal (Portugal) Cerro de la Virgen (Spain) Argissa-Magula and Knossos 154

Figure 89 Meles meles (a) left mandible in lateral view (level 14) (b) lower canine (level 14) (c) left ulna in medial views the proximal epiphysis is not fused (level 3) Martes (d) distal part of humerus in cranial view (level 9) 155

Figure 810 Age classes of the mandibles of Ovis and Capra 157

Figure 811 Age classes of the mandibles of Bos taurus 158

Figure 812 Distal part of metacarpus belonging to a male (possibly ox) of Bos taurus with osseous deformations on the articular surfaces 158

Figure 813 Age classes of Sus scrofa domesticus maxillae and mandibles 159

Figure 814 Correlation of the measurements of phalanx I belonging to Bos taurus 160

Figure 91 Map showing location of sites mentioned in the text 1 Ashkelon 2 lsquoAin Ghazal 3 Atlit Yam 4 Hagoshrim and Tel Ali 5 Ras Shamra 6 Cap Andreas Kastros 7 Khirokitia 8 Tenta 9 Asikli Houmlyuumlk 10 Mersin 11 Can Hasan III 12 Ccedilatalhoumlyuumlk 13 Suberde 14 Haccedililar 15 Nea Nikomedeia 16 Argissa-Magula 17 Sesklo 18 Achilleion 19 Franchthi Cave 20 Sidari Corfu 21 Cave of the Cyclops Youra 22 Melos 23 Santorini 24 Knossos Crete 25 Tel Aray 2 26 Umm el Tlel 27 Qdeir 28 El Kowm 2 173

Figure 92 Adult male agrimi (Capra aegagrus cretica) showing phenotypic resemblance to the wild bezoar goat 177

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

Table of Contents

List of Tables in the Text ix

List of Figures in the Text xiii

Preface Alexandra Karetsou xix

Acknowledgments xxiii

Introduction Nikos Efstratiou xxv

1 The Excavation Nikos Efstratiou Alexandra Karetsou and Eleni Banou 1

2 The Stratigraphy and Cultural Phases Nikos Efstratiou 25

3 Fabric Diversity in the Neolithic Ceramics of Knossos Sarantis Dimitriadis 47

4 Neolithic Sedimentology at Knossos Maria-Pilar Fumanal Garciacuteadagger 53

5 The Economy of Neolithic Knossos The Archaeobotanical Data Anaya Sarpaki 63

6 Wood Charcoal Analysis The Local Vegetation Ernestina Badal and Maria Ntinou 95

7 Plant Economy and the Use of Space Evidence from the Opal Phytoliths Marco Madella 119

8 The Knossos Fauna and the Beginning of the Neolithic in the Mediterranean Islands Manuel Peacuterez Ripoll 133

9 The Earliest Settlement on Crete An Archaeozoological Perspective Liora Kolska Horwitz 171

viii THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

10 Radiocarbon Dates from the Neolithic Settlement of Knossos An Overview Yorgos Facorellis and Yiannis Maniatis 193

11 Knossos and the Beginning of the Neolithic in Greece and the Aegean Islands Nikos Efstratiou 201

Index 215

Table 41 Correlation of sedimentology samples with excavation levels and cultural phases 55

Table 42 Munsell color and calculation of statistical parameters of mean size sorting skewness and kurtosis for each of the analyzed sedimentology samples 56

Table 51 Seed list provided to JD Evans by Hans Helbaek (unpublished) 67

Table 52 List of archaeobotanical samples from the 1997 rescue excavation along with relative and absolute dates 67

Table 53 Aceramic Neolithic archaeobotanical sample E 97(30) from Knossos 1997 level 39 retrieved from 16 liters of water-floated soil 68

Table 54 Measurements of Triticum turgidumaestivum from Aceramic and EN levels at Knossos 71

Table 55 Measurements of Triticum dicoccum Triticum monococcum Hordeum vulgare and Lens culinaris 74

Table 56 Early Neolithic I archaeobotanical (seed) samples 75

Table 57 Measurements of Trifolium spp and Leguminosae 78

Table 58 Measurements of Raphanus cf raphanistrum and Linum cf usitatissimum 78

Table 59 Early Neolithic II archaeobotanical (seed) samples 81

List of Tables in the Text

x THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Table 510 Vitis sp measurements from EN II levels sketch of a grape seed showing locations of dimensions 85

Table 511 Middle Neolithic archaeobotanical (seed) samples 86

Table 512 Late Neolithic archaeobotanical (seed) samples 88

Table 61 Inventories of plants growing in different parts of the study area 99

Table 62 Absolute and relative frequencies of taxa identified in the wood charcoal assemblages from Neolithic Knossos 102

Table 63 Presence of plant taxa in wood charcoal assemblages from Neolithic Knossos along with the total number of fragments analyzed and the total number of taxa identified in each level 104

Table 71 Knossos 1997 south profile phytolith counts 122

Table 72 Knossos 1997 west profile phytolith counts 125

Table 81 Number of identified and unidentified specimens by taxa and period 135

Table 82 Measurements of bones from Bos taurus 135

Table 83 Measurements of bones from Ovis aries (Oa) and Capra hircus (Ch) 136

Table 84 Measurements of bones from Sus scrofa domesticus Sus scrofa ferus Capra aegagrus Martes and Meles meles 138

Table 85 Number of identified specimens of Bos OvisCapra and Sus with number of marks caused by dog gnawing 140

Table 86 Early Neolithic I and EN II faunal remains 141

Table 87 Early Neolithic IIMN faunal remains 142

Table 88 Middle Neolithic faunal remains 144

Table 89A Late Neolithic faunal remains 146

Table 89B Late Neolithic faunal remains 149

Table 810 Number of identified specimens of Bos and OvisCapraSus with burn marks 151

Table 811 Number of long bone remains (the diaphysis fragments are not counted here) phalanges and tarsi corresponding to mature and immature bones along with the number of LN tooth remains grouped by age for comparison with the long bones 156

Table 812 Number of mandibles (NM) for goats and sheep from the LN levels classified by age 157

Table 813 Number of mandibles (NM) of Bos taurus classified by age 158

Table 814 Number of maxillary and mandibular remains of Sus scrofa domesticus classified by age 159

Table 815 Number of identified specimens of Bos taurus classified by sex 160

Table 816 Number of identified specimens of Ovis aries (Oa) and Capra hircus (Ch) classified by sex 160

xiLIST OF TABLES IN THE TEXT

Table 817 Chronological representation of the faunal species at Knossos 161

Table 818 Percentages of identified specimens of domestic and wild species at Knossos and other sites 161

Table 819 Representation and abundance of various faunal species at Shillourokambos Ais Yiorkis and Khirokitia 162

Table 820 Introduction and chronological representation of wild animals at various sites in Crete 163

Table 91 Relative frequencies (percentages) of animal species from Knossos 175

Table 92 Schematic representation of the relative chronology (cal bc dates) of sites mentioned in the text 183

Table 101 Summary of the British Museum radiocarbon dates on charcoal from the excavations of JD Evans at Neolithic Knossos sorted by age 194

Table 102 Description of the samples dated in the British Museum Research Laboratory 195

Table 103 Summary of radiocarbon dating results of carbonized samples collected in 1997 from the Neolithic settlement levels at Knossos 196

Frontispiece The city the fortifications the harbor and the hinterland of Khandax (Herakleion) in the first half of the 17th century Map by unknown cartographer 17th c Collezione Museo Civico Padua Vikelaia Municipal Library Herakleion ii

Figure i The Minoan palace and its Neolithic past xxi

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation 2

Figure 12 Trench II (a) view of the Central Court of the Palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the souther profile 3

Figure 13 Plan of the excavation trenches next to the staircase 3

Figure 14 South and west stratigraphic profiles of the trench 4

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench 5

Figure 16 Plans of excavation levels 9 and 10 7

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner 8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner 8

List of Figures in the Text

xiv THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 19 Plan of excavation level 13 9

Figure 110 Plan of excavation level 14 showing hearth in northwest corner 9

Figure 111 View of excavation level 14 showing hearth in northwest corner 10

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south 11

Figure 113 View of excavation level 15 from above 12

Figure 114 View of excavation level 16 from above 12

Figure 115 View of excavation level 16 facing west section 12

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4 12

Figure 117 View of level 17 facing west section 13

Figure 118 View of excavation level 18 from above 13

Figure 119 View of excavation level 19 from above 13

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6 13

Figure 121 View of excavation level 21 from above 14

Figure 122 View of excavation level 23 14

Figure 123 View of excavation level 24 showing wall 7 and grinding stones 14

Figure 124 Plans of excavation levels 22 and 24 15

Figure 125 Plan of excavation level 27 16

Figure 126 View of excavation level 28 from above 16

Figure 127 View of excavation level 28 facing west profile 16

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths 17

Figure 129 View of level 29A showing hearths 1 2 and 3 18

Figure 130 View of excavation level 30 showing hearth 4 18

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure 19

Figure 132 View of the elliptical stone wall from levels 24ndash27 19

Figure 133 View of excavation level 31 19

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7 20

Figure 135 Plan of excavation level 37 showing pits 1 and 2 21

Figure 21 Sedimentological samples of the middle part of the south profile 26

Figure 41 Fine fraction granulometry () of the samples from the west profile 57

Figure 42 The organic content () of the samples from the west profile 57

Figure 43 The carbonate content of the samples from the west profile 57

xvLIST OF FIGURES IN THE TEXT

Figure 44 Morphoscopy of sands without acid treatment 57

Figure 45 Morphoscopy of sands after the elimination of calcareous grains (subsequent to acid treatment) 57

Figure 51 Drawing of Triticum turgidum LT aestivum from the 1997 excavations at Knossos 69

Figure 52 Graphs of measurements and measurement ratios of Triticum turgidumaestivum from Neolithic Knossos 72

Figure 53 Measurements of Triticum turgidumaestivum from Aceramic and Early Neolithic Knossos compared with average values for Erbaba Ramad and Bouqras in the Near East 73

Figure 54 Graphs of measurements and measurement ratios of Lens culinaris from Neolithic Knossos 74

Figure 55 Early sites including those from mainland Greece where Triticum turgidumaestivum is reported 1 Tell Abu Hureyra 2 Tell Halula 3 Tell Aswad 4 Tell Ghoraife 5 Tell Sabi Abyad 6 Servia 7 Cafer Houmlyuumlk 8 Dhali Agridhi 9 Otzaki 10 Sesklo 11 Sitagroi 12 Haccedililar 13 Aşkli Houmlyuumlk 14 Ccedilatal Houmlyuumlk 15 Can Hasan 16 Cayoumlnuuml 17 El Kown 18 Bouqras 19 Tell Ramad 89

Figure 56 Summary of the distribution of all categories of archaeobotanical remains at Neolithic Knossos 90

Figure 61 Climate and topography of Knossos (a) mean annual precipitation in Crete (b) topographic map of the area around Knossos (c) westndasheast topographic section (d) southwestndashnortheast topographic section 96

Figure 62 View of the Knossos valley from Mt Juktas showing present-day vegetation 98

Figure 63 Panoramic view of the site of Knossos showing present-day vegetation 98

Figure 64 Present-day phrygana vegetation on the hills in the study area 98

Figure 65 Present-day vegetation on deep soils in the study area 98

Figure 66 Anatomy of plant taxa identified in wood charcoal assemblages from Neolithic Knossos 106

Figure 67 Wood charcoal diagram from Neolithic Knossos showing relative frequencies of taxa in successive excavation levels 108

Figure 71 Bar chart of phytolith per centage frequencies from the south profile 122

Figure 72 Bar chart of C3 and C

4 phytolith percentage frequencies from the south profile 122

Figure 73 West profile stratigraphy and sampling 123

Figure 74 Bar chart of phytolith percentage frequencies from the west profile 124

Figure 75 Bar chart of C3 and C

4 phytolith percentage frequencies from the west profile 125

Figure 76 Trench section (southwest corner to west face) with phases identified according to phytolith composition and frequencies 127

xvi THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 77 Silica skeleton from grass leaf (long cells and a stoma) from the EN I deposits (sample XXa level 32) 129

Figure 78 Wheat-type silica skeleton from the EN I deposits (sample XXIVb level 32) 129

Figure 79 Silica skeleton from a dicotyledonous plant from the EN I deposits (sample XIV level 30) 129

Figure 710 Millet-type silica skeleton from the EN II deposits (sample Xa level 16) 129

Figure 81 Percentages of the osseous parts of cattle long bones 139

Figure 82 Percentages of the osseous parts belonging to the long bones of middle-sized mammals (goats sheep and pigs) 139

Figure 83 Skeletal fragments of long bones of OvisCapra from level 14 all with dog-gnawing marks 140

Figure 84 Fragments of proximal epiphyses of femur and tibia of Bos taurus with fracture marks caused by impacts from the extraction of marrow level 24 151

Figure 85 Animal bones from level 3 (a) radius diaphysis (b) radius proximal part (c) scapula (d) fragment of femur (e) phalanx I of Capra aegagrus 153

Figure 86 Distal metacarpus of Capra aegagrus and Ovis aries 153

Figure 87 Animal bones (a) ulna in lateral view probably belonging to a wild boar (level 23) (b) Sus scrofa ferus canine fragment (level 10) (c) Sus scrofa domesticus ulna in lateral view (level 14) 154

Figure 88 Diaphysis width range (SD) of Sus scrofa domesticus and Sus scrofa ferus from Zambujal (Portugal) Cerro de la Virgen (Spain) Argissa-Magula and Knossos 154

Figure 89 Meles meles (a) left mandible in lateral view (level 14) (b) lower canine (level 14) (c) left ulna in medial views the proximal epiphysis is not fused (level 3) Martes (d) distal part of humerus in cranial view (level 9) 155

Figure 810 Age classes of the mandibles of Ovis and Capra 157

Figure 811 Age classes of the mandibles of Bos taurus 158

Figure 812 Distal part of metacarpus belonging to a male (possibly ox) of Bos taurus with osseous deformations on the articular surfaces 158

Figure 813 Age classes of Sus scrofa domesticus maxillae and mandibles 159

Figure 814 Correlation of the measurements of phalanx I belonging to Bos taurus 160

Figure 91 Map showing location of sites mentioned in the text 1 Ashkelon 2 lsquoAin Ghazal 3 Atlit Yam 4 Hagoshrim and Tel Ali 5 Ras Shamra 6 Cap Andreas Kastros 7 Khirokitia 8 Tenta 9 Asikli Houmlyuumlk 10 Mersin 11 Can Hasan III 12 Ccedilatalhoumlyuumlk 13 Suberde 14 Haccedililar 15 Nea Nikomedeia 16 Argissa-Magula 17 Sesklo 18 Achilleion 19 Franchthi Cave 20 Sidari Corfu 21 Cave of the Cyclops Youra 22 Melos 23 Santorini 24 Knossos Crete 25 Tel Aray 2 26 Umm el Tlel 27 Qdeir 28 El Kowm 2 173

Figure 92 Adult male agrimi (Capra aegagrus cretica) showing phenotypic resemblance to the wild bezoar goat 177

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

viii THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

10 Radiocarbon Dates from the Neolithic Settlement of Knossos An Overview Yorgos Facorellis and Yiannis Maniatis 193

11 Knossos and the Beginning of the Neolithic in Greece and the Aegean Islands Nikos Efstratiou 201

Index 215

Table 41 Correlation of sedimentology samples with excavation levels and cultural phases 55

Table 42 Munsell color and calculation of statistical parameters of mean size sorting skewness and kurtosis for each of the analyzed sedimentology samples 56

Table 51 Seed list provided to JD Evans by Hans Helbaek (unpublished) 67

Table 52 List of archaeobotanical samples from the 1997 rescue excavation along with relative and absolute dates 67

Table 53 Aceramic Neolithic archaeobotanical sample E 97(30) from Knossos 1997 level 39 retrieved from 16 liters of water-floated soil 68

Table 54 Measurements of Triticum turgidumaestivum from Aceramic and EN levels at Knossos 71

Table 55 Measurements of Triticum dicoccum Triticum monococcum Hordeum vulgare and Lens culinaris 74

Table 56 Early Neolithic I archaeobotanical (seed) samples 75

Table 57 Measurements of Trifolium spp and Leguminosae 78

Table 58 Measurements of Raphanus cf raphanistrum and Linum cf usitatissimum 78

Table 59 Early Neolithic II archaeobotanical (seed) samples 81

List of Tables in the Text

x THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Table 510 Vitis sp measurements from EN II levels sketch of a grape seed showing locations of dimensions 85

Table 511 Middle Neolithic archaeobotanical (seed) samples 86

Table 512 Late Neolithic archaeobotanical (seed) samples 88

Table 61 Inventories of plants growing in different parts of the study area 99

Table 62 Absolute and relative frequencies of taxa identified in the wood charcoal assemblages from Neolithic Knossos 102

Table 63 Presence of plant taxa in wood charcoal assemblages from Neolithic Knossos along with the total number of fragments analyzed and the total number of taxa identified in each level 104

Table 71 Knossos 1997 south profile phytolith counts 122

Table 72 Knossos 1997 west profile phytolith counts 125

Table 81 Number of identified and unidentified specimens by taxa and period 135

Table 82 Measurements of bones from Bos taurus 135

Table 83 Measurements of bones from Ovis aries (Oa) and Capra hircus (Ch) 136

Table 84 Measurements of bones from Sus scrofa domesticus Sus scrofa ferus Capra aegagrus Martes and Meles meles 138

Table 85 Number of identified specimens of Bos OvisCapra and Sus with number of marks caused by dog gnawing 140

Table 86 Early Neolithic I and EN II faunal remains 141

Table 87 Early Neolithic IIMN faunal remains 142

Table 88 Middle Neolithic faunal remains 144

Table 89A Late Neolithic faunal remains 146

Table 89B Late Neolithic faunal remains 149

Table 810 Number of identified specimens of Bos and OvisCapraSus with burn marks 151

Table 811 Number of long bone remains (the diaphysis fragments are not counted here) phalanges and tarsi corresponding to mature and immature bones along with the number of LN tooth remains grouped by age for comparison with the long bones 156

Table 812 Number of mandibles (NM) for goats and sheep from the LN levels classified by age 157

Table 813 Number of mandibles (NM) of Bos taurus classified by age 158

Table 814 Number of maxillary and mandibular remains of Sus scrofa domesticus classified by age 159

Table 815 Number of identified specimens of Bos taurus classified by sex 160

Table 816 Number of identified specimens of Ovis aries (Oa) and Capra hircus (Ch) classified by sex 160

xiLIST OF TABLES IN THE TEXT

Table 817 Chronological representation of the faunal species at Knossos 161

Table 818 Percentages of identified specimens of domestic and wild species at Knossos and other sites 161

Table 819 Representation and abundance of various faunal species at Shillourokambos Ais Yiorkis and Khirokitia 162

Table 820 Introduction and chronological representation of wild animals at various sites in Crete 163

Table 91 Relative frequencies (percentages) of animal species from Knossos 175

Table 92 Schematic representation of the relative chronology (cal bc dates) of sites mentioned in the text 183

Table 101 Summary of the British Museum radiocarbon dates on charcoal from the excavations of JD Evans at Neolithic Knossos sorted by age 194

Table 102 Description of the samples dated in the British Museum Research Laboratory 195

Table 103 Summary of radiocarbon dating results of carbonized samples collected in 1997 from the Neolithic settlement levels at Knossos 196

Frontispiece The city the fortifications the harbor and the hinterland of Khandax (Herakleion) in the first half of the 17th century Map by unknown cartographer 17th c Collezione Museo Civico Padua Vikelaia Municipal Library Herakleion ii

Figure i The Minoan palace and its Neolithic past xxi

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation 2

Figure 12 Trench II (a) view of the Central Court of the Palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the souther profile 3

Figure 13 Plan of the excavation trenches next to the staircase 3

Figure 14 South and west stratigraphic profiles of the trench 4

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench 5

Figure 16 Plans of excavation levels 9 and 10 7

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner 8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner 8

List of Figures in the Text

xiv THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 19 Plan of excavation level 13 9

Figure 110 Plan of excavation level 14 showing hearth in northwest corner 9

Figure 111 View of excavation level 14 showing hearth in northwest corner 10

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south 11

Figure 113 View of excavation level 15 from above 12

Figure 114 View of excavation level 16 from above 12

Figure 115 View of excavation level 16 facing west section 12

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4 12

Figure 117 View of level 17 facing west section 13

Figure 118 View of excavation level 18 from above 13

Figure 119 View of excavation level 19 from above 13

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6 13

Figure 121 View of excavation level 21 from above 14

Figure 122 View of excavation level 23 14

Figure 123 View of excavation level 24 showing wall 7 and grinding stones 14

Figure 124 Plans of excavation levels 22 and 24 15

Figure 125 Plan of excavation level 27 16

Figure 126 View of excavation level 28 from above 16

Figure 127 View of excavation level 28 facing west profile 16

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths 17

Figure 129 View of level 29A showing hearths 1 2 and 3 18

Figure 130 View of excavation level 30 showing hearth 4 18

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure 19

Figure 132 View of the elliptical stone wall from levels 24ndash27 19

Figure 133 View of excavation level 31 19

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7 20

Figure 135 Plan of excavation level 37 showing pits 1 and 2 21

Figure 21 Sedimentological samples of the middle part of the south profile 26

Figure 41 Fine fraction granulometry () of the samples from the west profile 57

Figure 42 The organic content () of the samples from the west profile 57

Figure 43 The carbonate content of the samples from the west profile 57

xvLIST OF FIGURES IN THE TEXT

Figure 44 Morphoscopy of sands without acid treatment 57

Figure 45 Morphoscopy of sands after the elimination of calcareous grains (subsequent to acid treatment) 57

Figure 51 Drawing of Triticum turgidum LT aestivum from the 1997 excavations at Knossos 69

Figure 52 Graphs of measurements and measurement ratios of Triticum turgidumaestivum from Neolithic Knossos 72

Figure 53 Measurements of Triticum turgidumaestivum from Aceramic and Early Neolithic Knossos compared with average values for Erbaba Ramad and Bouqras in the Near East 73

Figure 54 Graphs of measurements and measurement ratios of Lens culinaris from Neolithic Knossos 74

Figure 55 Early sites including those from mainland Greece where Triticum turgidumaestivum is reported 1 Tell Abu Hureyra 2 Tell Halula 3 Tell Aswad 4 Tell Ghoraife 5 Tell Sabi Abyad 6 Servia 7 Cafer Houmlyuumlk 8 Dhali Agridhi 9 Otzaki 10 Sesklo 11 Sitagroi 12 Haccedililar 13 Aşkli Houmlyuumlk 14 Ccedilatal Houmlyuumlk 15 Can Hasan 16 Cayoumlnuuml 17 El Kown 18 Bouqras 19 Tell Ramad 89

Figure 56 Summary of the distribution of all categories of archaeobotanical remains at Neolithic Knossos 90

Figure 61 Climate and topography of Knossos (a) mean annual precipitation in Crete (b) topographic map of the area around Knossos (c) westndasheast topographic section (d) southwestndashnortheast topographic section 96

Figure 62 View of the Knossos valley from Mt Juktas showing present-day vegetation 98

Figure 63 Panoramic view of the site of Knossos showing present-day vegetation 98

Figure 64 Present-day phrygana vegetation on the hills in the study area 98

Figure 65 Present-day vegetation on deep soils in the study area 98

Figure 66 Anatomy of plant taxa identified in wood charcoal assemblages from Neolithic Knossos 106

Figure 67 Wood charcoal diagram from Neolithic Knossos showing relative frequencies of taxa in successive excavation levels 108

Figure 71 Bar chart of phytolith per centage frequencies from the south profile 122

Figure 72 Bar chart of C3 and C

4 phytolith percentage frequencies from the south profile 122

Figure 73 West profile stratigraphy and sampling 123

Figure 74 Bar chart of phytolith percentage frequencies from the west profile 124

Figure 75 Bar chart of C3 and C

4 phytolith percentage frequencies from the west profile 125

Figure 76 Trench section (southwest corner to west face) with phases identified according to phytolith composition and frequencies 127

xvi THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 77 Silica skeleton from grass leaf (long cells and a stoma) from the EN I deposits (sample XXa level 32) 129

Figure 78 Wheat-type silica skeleton from the EN I deposits (sample XXIVb level 32) 129

Figure 79 Silica skeleton from a dicotyledonous plant from the EN I deposits (sample XIV level 30) 129

Figure 710 Millet-type silica skeleton from the EN II deposits (sample Xa level 16) 129

Figure 81 Percentages of the osseous parts of cattle long bones 139

Figure 82 Percentages of the osseous parts belonging to the long bones of middle-sized mammals (goats sheep and pigs) 139

Figure 83 Skeletal fragments of long bones of OvisCapra from level 14 all with dog-gnawing marks 140

Figure 84 Fragments of proximal epiphyses of femur and tibia of Bos taurus with fracture marks caused by impacts from the extraction of marrow level 24 151

Figure 85 Animal bones from level 3 (a) radius diaphysis (b) radius proximal part (c) scapula (d) fragment of femur (e) phalanx I of Capra aegagrus 153

Figure 86 Distal metacarpus of Capra aegagrus and Ovis aries 153

Figure 87 Animal bones (a) ulna in lateral view probably belonging to a wild boar (level 23) (b) Sus scrofa ferus canine fragment (level 10) (c) Sus scrofa domesticus ulna in lateral view (level 14) 154

Figure 88 Diaphysis width range (SD) of Sus scrofa domesticus and Sus scrofa ferus from Zambujal (Portugal) Cerro de la Virgen (Spain) Argissa-Magula and Knossos 154

Figure 89 Meles meles (a) left mandible in lateral view (level 14) (b) lower canine (level 14) (c) left ulna in medial views the proximal epiphysis is not fused (level 3) Martes (d) distal part of humerus in cranial view (level 9) 155

Figure 810 Age classes of the mandibles of Ovis and Capra 157

Figure 811 Age classes of the mandibles of Bos taurus 158

Figure 812 Distal part of metacarpus belonging to a male (possibly ox) of Bos taurus with osseous deformations on the articular surfaces 158

Figure 813 Age classes of Sus scrofa domesticus maxillae and mandibles 159

Figure 814 Correlation of the measurements of phalanx I belonging to Bos taurus 160

Figure 91 Map showing location of sites mentioned in the text 1 Ashkelon 2 lsquoAin Ghazal 3 Atlit Yam 4 Hagoshrim and Tel Ali 5 Ras Shamra 6 Cap Andreas Kastros 7 Khirokitia 8 Tenta 9 Asikli Houmlyuumlk 10 Mersin 11 Can Hasan III 12 Ccedilatalhoumlyuumlk 13 Suberde 14 Haccedililar 15 Nea Nikomedeia 16 Argissa-Magula 17 Sesklo 18 Achilleion 19 Franchthi Cave 20 Sidari Corfu 21 Cave of the Cyclops Youra 22 Melos 23 Santorini 24 Knossos Crete 25 Tel Aray 2 26 Umm el Tlel 27 Qdeir 28 El Kowm 2 173

Figure 92 Adult male agrimi (Capra aegagrus cretica) showing phenotypic resemblance to the wild bezoar goat 177

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

Table 41 Correlation of sedimentology samples with excavation levels and cultural phases 55

Table 42 Munsell color and calculation of statistical parameters of mean size sorting skewness and kurtosis for each of the analyzed sedimentology samples 56

Table 51 Seed list provided to JD Evans by Hans Helbaek (unpublished) 67

Table 52 List of archaeobotanical samples from the 1997 rescue excavation along with relative and absolute dates 67

Table 53 Aceramic Neolithic archaeobotanical sample E 97(30) from Knossos 1997 level 39 retrieved from 16 liters of water-floated soil 68

Table 54 Measurements of Triticum turgidumaestivum from Aceramic and EN levels at Knossos 71

Table 55 Measurements of Triticum dicoccum Triticum monococcum Hordeum vulgare and Lens culinaris 74

Table 56 Early Neolithic I archaeobotanical (seed) samples 75

Table 57 Measurements of Trifolium spp and Leguminosae 78

Table 58 Measurements of Raphanus cf raphanistrum and Linum cf usitatissimum 78

Table 59 Early Neolithic II archaeobotanical (seed) samples 81

List of Tables in the Text

x THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Table 510 Vitis sp measurements from EN II levels sketch of a grape seed showing locations of dimensions 85

Table 511 Middle Neolithic archaeobotanical (seed) samples 86

Table 512 Late Neolithic archaeobotanical (seed) samples 88

Table 61 Inventories of plants growing in different parts of the study area 99

Table 62 Absolute and relative frequencies of taxa identified in the wood charcoal assemblages from Neolithic Knossos 102

Table 63 Presence of plant taxa in wood charcoal assemblages from Neolithic Knossos along with the total number of fragments analyzed and the total number of taxa identified in each level 104

Table 71 Knossos 1997 south profile phytolith counts 122

Table 72 Knossos 1997 west profile phytolith counts 125

Table 81 Number of identified and unidentified specimens by taxa and period 135

Table 82 Measurements of bones from Bos taurus 135

Table 83 Measurements of bones from Ovis aries (Oa) and Capra hircus (Ch) 136

Table 84 Measurements of bones from Sus scrofa domesticus Sus scrofa ferus Capra aegagrus Martes and Meles meles 138

Table 85 Number of identified specimens of Bos OvisCapra and Sus with number of marks caused by dog gnawing 140

Table 86 Early Neolithic I and EN II faunal remains 141

Table 87 Early Neolithic IIMN faunal remains 142

Table 88 Middle Neolithic faunal remains 144

Table 89A Late Neolithic faunal remains 146

Table 89B Late Neolithic faunal remains 149

Table 810 Number of identified specimens of Bos and OvisCapraSus with burn marks 151

Table 811 Number of long bone remains (the diaphysis fragments are not counted here) phalanges and tarsi corresponding to mature and immature bones along with the number of LN tooth remains grouped by age for comparison with the long bones 156

Table 812 Number of mandibles (NM) for goats and sheep from the LN levels classified by age 157

Table 813 Number of mandibles (NM) of Bos taurus classified by age 158

Table 814 Number of maxillary and mandibular remains of Sus scrofa domesticus classified by age 159

Table 815 Number of identified specimens of Bos taurus classified by sex 160

Table 816 Number of identified specimens of Ovis aries (Oa) and Capra hircus (Ch) classified by sex 160

xiLIST OF TABLES IN THE TEXT

Table 817 Chronological representation of the faunal species at Knossos 161

Table 818 Percentages of identified specimens of domestic and wild species at Knossos and other sites 161

Table 819 Representation and abundance of various faunal species at Shillourokambos Ais Yiorkis and Khirokitia 162

Table 820 Introduction and chronological representation of wild animals at various sites in Crete 163

Table 91 Relative frequencies (percentages) of animal species from Knossos 175

Table 92 Schematic representation of the relative chronology (cal bc dates) of sites mentioned in the text 183

Table 101 Summary of the British Museum radiocarbon dates on charcoal from the excavations of JD Evans at Neolithic Knossos sorted by age 194

Table 102 Description of the samples dated in the British Museum Research Laboratory 195

Table 103 Summary of radiocarbon dating results of carbonized samples collected in 1997 from the Neolithic settlement levels at Knossos 196

Frontispiece The city the fortifications the harbor and the hinterland of Khandax (Herakleion) in the first half of the 17th century Map by unknown cartographer 17th c Collezione Museo Civico Padua Vikelaia Municipal Library Herakleion ii

Figure i The Minoan palace and its Neolithic past xxi

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation 2

Figure 12 Trench II (a) view of the Central Court of the Palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the souther profile 3

Figure 13 Plan of the excavation trenches next to the staircase 3

Figure 14 South and west stratigraphic profiles of the trench 4

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench 5

Figure 16 Plans of excavation levels 9 and 10 7

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner 8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner 8

List of Figures in the Text

xiv THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 19 Plan of excavation level 13 9

Figure 110 Plan of excavation level 14 showing hearth in northwest corner 9

Figure 111 View of excavation level 14 showing hearth in northwest corner 10

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south 11

Figure 113 View of excavation level 15 from above 12

Figure 114 View of excavation level 16 from above 12

Figure 115 View of excavation level 16 facing west section 12

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4 12

Figure 117 View of level 17 facing west section 13

Figure 118 View of excavation level 18 from above 13

Figure 119 View of excavation level 19 from above 13

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6 13

Figure 121 View of excavation level 21 from above 14

Figure 122 View of excavation level 23 14

Figure 123 View of excavation level 24 showing wall 7 and grinding stones 14

Figure 124 Plans of excavation levels 22 and 24 15

Figure 125 Plan of excavation level 27 16

Figure 126 View of excavation level 28 from above 16

Figure 127 View of excavation level 28 facing west profile 16

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths 17

Figure 129 View of level 29A showing hearths 1 2 and 3 18

Figure 130 View of excavation level 30 showing hearth 4 18

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure 19

Figure 132 View of the elliptical stone wall from levels 24ndash27 19

Figure 133 View of excavation level 31 19

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7 20

Figure 135 Plan of excavation level 37 showing pits 1 and 2 21

Figure 21 Sedimentological samples of the middle part of the south profile 26

Figure 41 Fine fraction granulometry () of the samples from the west profile 57

Figure 42 The organic content () of the samples from the west profile 57

Figure 43 The carbonate content of the samples from the west profile 57

xvLIST OF FIGURES IN THE TEXT

Figure 44 Morphoscopy of sands without acid treatment 57

Figure 45 Morphoscopy of sands after the elimination of calcareous grains (subsequent to acid treatment) 57

Figure 51 Drawing of Triticum turgidum LT aestivum from the 1997 excavations at Knossos 69

Figure 52 Graphs of measurements and measurement ratios of Triticum turgidumaestivum from Neolithic Knossos 72

Figure 53 Measurements of Triticum turgidumaestivum from Aceramic and Early Neolithic Knossos compared with average values for Erbaba Ramad and Bouqras in the Near East 73

Figure 54 Graphs of measurements and measurement ratios of Lens culinaris from Neolithic Knossos 74

Figure 55 Early sites including those from mainland Greece where Triticum turgidumaestivum is reported 1 Tell Abu Hureyra 2 Tell Halula 3 Tell Aswad 4 Tell Ghoraife 5 Tell Sabi Abyad 6 Servia 7 Cafer Houmlyuumlk 8 Dhali Agridhi 9 Otzaki 10 Sesklo 11 Sitagroi 12 Haccedililar 13 Aşkli Houmlyuumlk 14 Ccedilatal Houmlyuumlk 15 Can Hasan 16 Cayoumlnuuml 17 El Kown 18 Bouqras 19 Tell Ramad 89

Figure 56 Summary of the distribution of all categories of archaeobotanical remains at Neolithic Knossos 90

Figure 61 Climate and topography of Knossos (a) mean annual precipitation in Crete (b) topographic map of the area around Knossos (c) westndasheast topographic section (d) southwestndashnortheast topographic section 96

Figure 62 View of the Knossos valley from Mt Juktas showing present-day vegetation 98

Figure 63 Panoramic view of the site of Knossos showing present-day vegetation 98

Figure 64 Present-day phrygana vegetation on the hills in the study area 98

Figure 65 Present-day vegetation on deep soils in the study area 98

Figure 66 Anatomy of plant taxa identified in wood charcoal assemblages from Neolithic Knossos 106

Figure 67 Wood charcoal diagram from Neolithic Knossos showing relative frequencies of taxa in successive excavation levels 108

Figure 71 Bar chart of phytolith per centage frequencies from the south profile 122

Figure 72 Bar chart of C3 and C

4 phytolith percentage frequencies from the south profile 122

Figure 73 West profile stratigraphy and sampling 123

Figure 74 Bar chart of phytolith percentage frequencies from the west profile 124

Figure 75 Bar chart of C3 and C

4 phytolith percentage frequencies from the west profile 125

Figure 76 Trench section (southwest corner to west face) with phases identified according to phytolith composition and frequencies 127

xvi THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 77 Silica skeleton from grass leaf (long cells and a stoma) from the EN I deposits (sample XXa level 32) 129

Figure 78 Wheat-type silica skeleton from the EN I deposits (sample XXIVb level 32) 129

Figure 79 Silica skeleton from a dicotyledonous plant from the EN I deposits (sample XIV level 30) 129

Figure 710 Millet-type silica skeleton from the EN II deposits (sample Xa level 16) 129

Figure 81 Percentages of the osseous parts of cattle long bones 139

Figure 82 Percentages of the osseous parts belonging to the long bones of middle-sized mammals (goats sheep and pigs) 139

Figure 83 Skeletal fragments of long bones of OvisCapra from level 14 all with dog-gnawing marks 140

Figure 84 Fragments of proximal epiphyses of femur and tibia of Bos taurus with fracture marks caused by impacts from the extraction of marrow level 24 151

Figure 85 Animal bones from level 3 (a) radius diaphysis (b) radius proximal part (c) scapula (d) fragment of femur (e) phalanx I of Capra aegagrus 153

Figure 86 Distal metacarpus of Capra aegagrus and Ovis aries 153

Figure 87 Animal bones (a) ulna in lateral view probably belonging to a wild boar (level 23) (b) Sus scrofa ferus canine fragment (level 10) (c) Sus scrofa domesticus ulna in lateral view (level 14) 154

Figure 88 Diaphysis width range (SD) of Sus scrofa domesticus and Sus scrofa ferus from Zambujal (Portugal) Cerro de la Virgen (Spain) Argissa-Magula and Knossos 154

Figure 89 Meles meles (a) left mandible in lateral view (level 14) (b) lower canine (level 14) (c) left ulna in medial views the proximal epiphysis is not fused (level 3) Martes (d) distal part of humerus in cranial view (level 9) 155

Figure 810 Age classes of the mandibles of Ovis and Capra 157

Figure 811 Age classes of the mandibles of Bos taurus 158

Figure 812 Distal part of metacarpus belonging to a male (possibly ox) of Bos taurus with osseous deformations on the articular surfaces 158

Figure 813 Age classes of Sus scrofa domesticus maxillae and mandibles 159

Figure 814 Correlation of the measurements of phalanx I belonging to Bos taurus 160

Figure 91 Map showing location of sites mentioned in the text 1 Ashkelon 2 lsquoAin Ghazal 3 Atlit Yam 4 Hagoshrim and Tel Ali 5 Ras Shamra 6 Cap Andreas Kastros 7 Khirokitia 8 Tenta 9 Asikli Houmlyuumlk 10 Mersin 11 Can Hasan III 12 Ccedilatalhoumlyuumlk 13 Suberde 14 Haccedililar 15 Nea Nikomedeia 16 Argissa-Magula 17 Sesklo 18 Achilleion 19 Franchthi Cave 20 Sidari Corfu 21 Cave of the Cyclops Youra 22 Melos 23 Santorini 24 Knossos Crete 25 Tel Aray 2 26 Umm el Tlel 27 Qdeir 28 El Kowm 2 173

Figure 92 Adult male agrimi (Capra aegagrus cretica) showing phenotypic resemblance to the wild bezoar goat 177

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

x THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Table 510 Vitis sp measurements from EN II levels sketch of a grape seed showing locations of dimensions 85

Table 511 Middle Neolithic archaeobotanical (seed) samples 86

Table 512 Late Neolithic archaeobotanical (seed) samples 88

Table 61 Inventories of plants growing in different parts of the study area 99

Table 62 Absolute and relative frequencies of taxa identified in the wood charcoal assemblages from Neolithic Knossos 102

Table 63 Presence of plant taxa in wood charcoal assemblages from Neolithic Knossos along with the total number of fragments analyzed and the total number of taxa identified in each level 104

Table 71 Knossos 1997 south profile phytolith counts 122

Table 72 Knossos 1997 west profile phytolith counts 125

Table 81 Number of identified and unidentified specimens by taxa and period 135

Table 82 Measurements of bones from Bos taurus 135

Table 83 Measurements of bones from Ovis aries (Oa) and Capra hircus (Ch) 136

Table 84 Measurements of bones from Sus scrofa domesticus Sus scrofa ferus Capra aegagrus Martes and Meles meles 138

Table 85 Number of identified specimens of Bos OvisCapra and Sus with number of marks caused by dog gnawing 140

Table 86 Early Neolithic I and EN II faunal remains 141

Table 87 Early Neolithic IIMN faunal remains 142

Table 88 Middle Neolithic faunal remains 144

Table 89A Late Neolithic faunal remains 146

Table 89B Late Neolithic faunal remains 149

Table 810 Number of identified specimens of Bos and OvisCapraSus with burn marks 151

Table 811 Number of long bone remains (the diaphysis fragments are not counted here) phalanges and tarsi corresponding to mature and immature bones along with the number of LN tooth remains grouped by age for comparison with the long bones 156

Table 812 Number of mandibles (NM) for goats and sheep from the LN levels classified by age 157

Table 813 Number of mandibles (NM) of Bos taurus classified by age 158

Table 814 Number of maxillary and mandibular remains of Sus scrofa domesticus classified by age 159

Table 815 Number of identified specimens of Bos taurus classified by sex 160

Table 816 Number of identified specimens of Ovis aries (Oa) and Capra hircus (Ch) classified by sex 160

xiLIST OF TABLES IN THE TEXT

Table 817 Chronological representation of the faunal species at Knossos 161

Table 818 Percentages of identified specimens of domestic and wild species at Knossos and other sites 161

Table 819 Representation and abundance of various faunal species at Shillourokambos Ais Yiorkis and Khirokitia 162

Table 820 Introduction and chronological representation of wild animals at various sites in Crete 163

Table 91 Relative frequencies (percentages) of animal species from Knossos 175

Table 92 Schematic representation of the relative chronology (cal bc dates) of sites mentioned in the text 183

Table 101 Summary of the British Museum radiocarbon dates on charcoal from the excavations of JD Evans at Neolithic Knossos sorted by age 194

Table 102 Description of the samples dated in the British Museum Research Laboratory 195

Table 103 Summary of radiocarbon dating results of carbonized samples collected in 1997 from the Neolithic settlement levels at Knossos 196

Frontispiece The city the fortifications the harbor and the hinterland of Khandax (Herakleion) in the first half of the 17th century Map by unknown cartographer 17th c Collezione Museo Civico Padua Vikelaia Municipal Library Herakleion ii

Figure i The Minoan palace and its Neolithic past xxi

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation 2

Figure 12 Trench II (a) view of the Central Court of the Palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the souther profile 3

Figure 13 Plan of the excavation trenches next to the staircase 3

Figure 14 South and west stratigraphic profiles of the trench 4

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench 5

Figure 16 Plans of excavation levels 9 and 10 7

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner 8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner 8

List of Figures in the Text

xiv THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 19 Plan of excavation level 13 9

Figure 110 Plan of excavation level 14 showing hearth in northwest corner 9

Figure 111 View of excavation level 14 showing hearth in northwest corner 10

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south 11

Figure 113 View of excavation level 15 from above 12

Figure 114 View of excavation level 16 from above 12

Figure 115 View of excavation level 16 facing west section 12

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4 12

Figure 117 View of level 17 facing west section 13

Figure 118 View of excavation level 18 from above 13

Figure 119 View of excavation level 19 from above 13

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6 13

Figure 121 View of excavation level 21 from above 14

Figure 122 View of excavation level 23 14

Figure 123 View of excavation level 24 showing wall 7 and grinding stones 14

Figure 124 Plans of excavation levels 22 and 24 15

Figure 125 Plan of excavation level 27 16

Figure 126 View of excavation level 28 from above 16

Figure 127 View of excavation level 28 facing west profile 16

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths 17

Figure 129 View of level 29A showing hearths 1 2 and 3 18

Figure 130 View of excavation level 30 showing hearth 4 18

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure 19

Figure 132 View of the elliptical stone wall from levels 24ndash27 19

Figure 133 View of excavation level 31 19

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7 20

Figure 135 Plan of excavation level 37 showing pits 1 and 2 21

Figure 21 Sedimentological samples of the middle part of the south profile 26

Figure 41 Fine fraction granulometry () of the samples from the west profile 57

Figure 42 The organic content () of the samples from the west profile 57

Figure 43 The carbonate content of the samples from the west profile 57

xvLIST OF FIGURES IN THE TEXT

Figure 44 Morphoscopy of sands without acid treatment 57

Figure 45 Morphoscopy of sands after the elimination of calcareous grains (subsequent to acid treatment) 57

Figure 51 Drawing of Triticum turgidum LT aestivum from the 1997 excavations at Knossos 69

Figure 52 Graphs of measurements and measurement ratios of Triticum turgidumaestivum from Neolithic Knossos 72

Figure 53 Measurements of Triticum turgidumaestivum from Aceramic and Early Neolithic Knossos compared with average values for Erbaba Ramad and Bouqras in the Near East 73

Figure 54 Graphs of measurements and measurement ratios of Lens culinaris from Neolithic Knossos 74

Figure 55 Early sites including those from mainland Greece where Triticum turgidumaestivum is reported 1 Tell Abu Hureyra 2 Tell Halula 3 Tell Aswad 4 Tell Ghoraife 5 Tell Sabi Abyad 6 Servia 7 Cafer Houmlyuumlk 8 Dhali Agridhi 9 Otzaki 10 Sesklo 11 Sitagroi 12 Haccedililar 13 Aşkli Houmlyuumlk 14 Ccedilatal Houmlyuumlk 15 Can Hasan 16 Cayoumlnuuml 17 El Kown 18 Bouqras 19 Tell Ramad 89

Figure 56 Summary of the distribution of all categories of archaeobotanical remains at Neolithic Knossos 90

Figure 61 Climate and topography of Knossos (a) mean annual precipitation in Crete (b) topographic map of the area around Knossos (c) westndasheast topographic section (d) southwestndashnortheast topographic section 96

Figure 62 View of the Knossos valley from Mt Juktas showing present-day vegetation 98

Figure 63 Panoramic view of the site of Knossos showing present-day vegetation 98

Figure 64 Present-day phrygana vegetation on the hills in the study area 98

Figure 65 Present-day vegetation on deep soils in the study area 98

Figure 66 Anatomy of plant taxa identified in wood charcoal assemblages from Neolithic Knossos 106

Figure 67 Wood charcoal diagram from Neolithic Knossos showing relative frequencies of taxa in successive excavation levels 108

Figure 71 Bar chart of phytolith per centage frequencies from the south profile 122

Figure 72 Bar chart of C3 and C

4 phytolith percentage frequencies from the south profile 122

Figure 73 West profile stratigraphy and sampling 123

Figure 74 Bar chart of phytolith percentage frequencies from the west profile 124

Figure 75 Bar chart of C3 and C

4 phytolith percentage frequencies from the west profile 125

Figure 76 Trench section (southwest corner to west face) with phases identified according to phytolith composition and frequencies 127

xvi THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 77 Silica skeleton from grass leaf (long cells and a stoma) from the EN I deposits (sample XXa level 32) 129

Figure 78 Wheat-type silica skeleton from the EN I deposits (sample XXIVb level 32) 129

Figure 79 Silica skeleton from a dicotyledonous plant from the EN I deposits (sample XIV level 30) 129

Figure 710 Millet-type silica skeleton from the EN II deposits (sample Xa level 16) 129

Figure 81 Percentages of the osseous parts of cattle long bones 139

Figure 82 Percentages of the osseous parts belonging to the long bones of middle-sized mammals (goats sheep and pigs) 139

Figure 83 Skeletal fragments of long bones of OvisCapra from level 14 all with dog-gnawing marks 140

Figure 84 Fragments of proximal epiphyses of femur and tibia of Bos taurus with fracture marks caused by impacts from the extraction of marrow level 24 151

Figure 85 Animal bones from level 3 (a) radius diaphysis (b) radius proximal part (c) scapula (d) fragment of femur (e) phalanx I of Capra aegagrus 153

Figure 86 Distal metacarpus of Capra aegagrus and Ovis aries 153

Figure 87 Animal bones (a) ulna in lateral view probably belonging to a wild boar (level 23) (b) Sus scrofa ferus canine fragment (level 10) (c) Sus scrofa domesticus ulna in lateral view (level 14) 154

Figure 88 Diaphysis width range (SD) of Sus scrofa domesticus and Sus scrofa ferus from Zambujal (Portugal) Cerro de la Virgen (Spain) Argissa-Magula and Knossos 154

Figure 89 Meles meles (a) left mandible in lateral view (level 14) (b) lower canine (level 14) (c) left ulna in medial views the proximal epiphysis is not fused (level 3) Martes (d) distal part of humerus in cranial view (level 9) 155

Figure 810 Age classes of the mandibles of Ovis and Capra 157

Figure 811 Age classes of the mandibles of Bos taurus 158

Figure 812 Distal part of metacarpus belonging to a male (possibly ox) of Bos taurus with osseous deformations on the articular surfaces 158

Figure 813 Age classes of Sus scrofa domesticus maxillae and mandibles 159

Figure 814 Correlation of the measurements of phalanx I belonging to Bos taurus 160

Figure 91 Map showing location of sites mentioned in the text 1 Ashkelon 2 lsquoAin Ghazal 3 Atlit Yam 4 Hagoshrim and Tel Ali 5 Ras Shamra 6 Cap Andreas Kastros 7 Khirokitia 8 Tenta 9 Asikli Houmlyuumlk 10 Mersin 11 Can Hasan III 12 Ccedilatalhoumlyuumlk 13 Suberde 14 Haccedililar 15 Nea Nikomedeia 16 Argissa-Magula 17 Sesklo 18 Achilleion 19 Franchthi Cave 20 Sidari Corfu 21 Cave of the Cyclops Youra 22 Melos 23 Santorini 24 Knossos Crete 25 Tel Aray 2 26 Umm el Tlel 27 Qdeir 28 El Kowm 2 173

Figure 92 Adult male agrimi (Capra aegagrus cretica) showing phenotypic resemblance to the wild bezoar goat 177

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

xiLIST OF TABLES IN THE TEXT

Table 817 Chronological representation of the faunal species at Knossos 161

Table 818 Percentages of identified specimens of domestic and wild species at Knossos and other sites 161

Table 819 Representation and abundance of various faunal species at Shillourokambos Ais Yiorkis and Khirokitia 162

Table 820 Introduction and chronological representation of wild animals at various sites in Crete 163

Table 91 Relative frequencies (percentages) of animal species from Knossos 175

Table 92 Schematic representation of the relative chronology (cal bc dates) of sites mentioned in the text 183

Table 101 Summary of the British Museum radiocarbon dates on charcoal from the excavations of JD Evans at Neolithic Knossos sorted by age 194

Table 102 Description of the samples dated in the British Museum Research Laboratory 195

Table 103 Summary of radiocarbon dating results of carbonized samples collected in 1997 from the Neolithic settlement levels at Knossos 196

Frontispiece The city the fortifications the harbor and the hinterland of Khandax (Herakleion) in the first half of the 17th century Map by unknown cartographer 17th c Collezione Museo Civico Padua Vikelaia Municipal Library Herakleion ii

Figure i The Minoan palace and its Neolithic past xxi

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation 2

Figure 12 Trench II (a) view of the Central Court of the Palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the souther profile 3

Figure 13 Plan of the excavation trenches next to the staircase 3

Figure 14 South and west stratigraphic profiles of the trench 4

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench 5

Figure 16 Plans of excavation levels 9 and 10 7

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner 8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner 8

List of Figures in the Text

xiv THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 19 Plan of excavation level 13 9

Figure 110 Plan of excavation level 14 showing hearth in northwest corner 9

Figure 111 View of excavation level 14 showing hearth in northwest corner 10

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south 11

Figure 113 View of excavation level 15 from above 12

Figure 114 View of excavation level 16 from above 12

Figure 115 View of excavation level 16 facing west section 12

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4 12

Figure 117 View of level 17 facing west section 13

Figure 118 View of excavation level 18 from above 13

Figure 119 View of excavation level 19 from above 13

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6 13

Figure 121 View of excavation level 21 from above 14

Figure 122 View of excavation level 23 14

Figure 123 View of excavation level 24 showing wall 7 and grinding stones 14

Figure 124 Plans of excavation levels 22 and 24 15

Figure 125 Plan of excavation level 27 16

Figure 126 View of excavation level 28 from above 16

Figure 127 View of excavation level 28 facing west profile 16

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths 17

Figure 129 View of level 29A showing hearths 1 2 and 3 18

Figure 130 View of excavation level 30 showing hearth 4 18

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure 19

Figure 132 View of the elliptical stone wall from levels 24ndash27 19

Figure 133 View of excavation level 31 19

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7 20

Figure 135 Plan of excavation level 37 showing pits 1 and 2 21

Figure 21 Sedimentological samples of the middle part of the south profile 26

Figure 41 Fine fraction granulometry () of the samples from the west profile 57

Figure 42 The organic content () of the samples from the west profile 57

Figure 43 The carbonate content of the samples from the west profile 57

xvLIST OF FIGURES IN THE TEXT

Figure 44 Morphoscopy of sands without acid treatment 57

Figure 45 Morphoscopy of sands after the elimination of calcareous grains (subsequent to acid treatment) 57

Figure 51 Drawing of Triticum turgidum LT aestivum from the 1997 excavations at Knossos 69

Figure 52 Graphs of measurements and measurement ratios of Triticum turgidumaestivum from Neolithic Knossos 72

Figure 53 Measurements of Triticum turgidumaestivum from Aceramic and Early Neolithic Knossos compared with average values for Erbaba Ramad and Bouqras in the Near East 73

Figure 54 Graphs of measurements and measurement ratios of Lens culinaris from Neolithic Knossos 74

Figure 55 Early sites including those from mainland Greece where Triticum turgidumaestivum is reported 1 Tell Abu Hureyra 2 Tell Halula 3 Tell Aswad 4 Tell Ghoraife 5 Tell Sabi Abyad 6 Servia 7 Cafer Houmlyuumlk 8 Dhali Agridhi 9 Otzaki 10 Sesklo 11 Sitagroi 12 Haccedililar 13 Aşkli Houmlyuumlk 14 Ccedilatal Houmlyuumlk 15 Can Hasan 16 Cayoumlnuuml 17 El Kown 18 Bouqras 19 Tell Ramad 89

Figure 56 Summary of the distribution of all categories of archaeobotanical remains at Neolithic Knossos 90

Figure 61 Climate and topography of Knossos (a) mean annual precipitation in Crete (b) topographic map of the area around Knossos (c) westndasheast topographic section (d) southwestndashnortheast topographic section 96

Figure 62 View of the Knossos valley from Mt Juktas showing present-day vegetation 98

Figure 63 Panoramic view of the site of Knossos showing present-day vegetation 98

Figure 64 Present-day phrygana vegetation on the hills in the study area 98

Figure 65 Present-day vegetation on deep soils in the study area 98

Figure 66 Anatomy of plant taxa identified in wood charcoal assemblages from Neolithic Knossos 106

Figure 67 Wood charcoal diagram from Neolithic Knossos showing relative frequencies of taxa in successive excavation levels 108

Figure 71 Bar chart of phytolith per centage frequencies from the south profile 122

Figure 72 Bar chart of C3 and C

4 phytolith percentage frequencies from the south profile 122

Figure 73 West profile stratigraphy and sampling 123

Figure 74 Bar chart of phytolith percentage frequencies from the west profile 124

Figure 75 Bar chart of C3 and C

4 phytolith percentage frequencies from the west profile 125

Figure 76 Trench section (southwest corner to west face) with phases identified according to phytolith composition and frequencies 127

xvi THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 77 Silica skeleton from grass leaf (long cells and a stoma) from the EN I deposits (sample XXa level 32) 129

Figure 78 Wheat-type silica skeleton from the EN I deposits (sample XXIVb level 32) 129

Figure 79 Silica skeleton from a dicotyledonous plant from the EN I deposits (sample XIV level 30) 129

Figure 710 Millet-type silica skeleton from the EN II deposits (sample Xa level 16) 129

Figure 81 Percentages of the osseous parts of cattle long bones 139

Figure 82 Percentages of the osseous parts belonging to the long bones of middle-sized mammals (goats sheep and pigs) 139

Figure 83 Skeletal fragments of long bones of OvisCapra from level 14 all with dog-gnawing marks 140

Figure 84 Fragments of proximal epiphyses of femur and tibia of Bos taurus with fracture marks caused by impacts from the extraction of marrow level 24 151

Figure 85 Animal bones from level 3 (a) radius diaphysis (b) radius proximal part (c) scapula (d) fragment of femur (e) phalanx I of Capra aegagrus 153

Figure 86 Distal metacarpus of Capra aegagrus and Ovis aries 153

Figure 87 Animal bones (a) ulna in lateral view probably belonging to a wild boar (level 23) (b) Sus scrofa ferus canine fragment (level 10) (c) Sus scrofa domesticus ulna in lateral view (level 14) 154

Figure 88 Diaphysis width range (SD) of Sus scrofa domesticus and Sus scrofa ferus from Zambujal (Portugal) Cerro de la Virgen (Spain) Argissa-Magula and Knossos 154

Figure 89 Meles meles (a) left mandible in lateral view (level 14) (b) lower canine (level 14) (c) left ulna in medial views the proximal epiphysis is not fused (level 3) Martes (d) distal part of humerus in cranial view (level 9) 155

Figure 810 Age classes of the mandibles of Ovis and Capra 157

Figure 811 Age classes of the mandibles of Bos taurus 158

Figure 812 Distal part of metacarpus belonging to a male (possibly ox) of Bos taurus with osseous deformations on the articular surfaces 158

Figure 813 Age classes of Sus scrofa domesticus maxillae and mandibles 159

Figure 814 Correlation of the measurements of phalanx I belonging to Bos taurus 160

Figure 91 Map showing location of sites mentioned in the text 1 Ashkelon 2 lsquoAin Ghazal 3 Atlit Yam 4 Hagoshrim and Tel Ali 5 Ras Shamra 6 Cap Andreas Kastros 7 Khirokitia 8 Tenta 9 Asikli Houmlyuumlk 10 Mersin 11 Can Hasan III 12 Ccedilatalhoumlyuumlk 13 Suberde 14 Haccedililar 15 Nea Nikomedeia 16 Argissa-Magula 17 Sesklo 18 Achilleion 19 Franchthi Cave 20 Sidari Corfu 21 Cave of the Cyclops Youra 22 Melos 23 Santorini 24 Knossos Crete 25 Tel Aray 2 26 Umm el Tlel 27 Qdeir 28 El Kowm 2 173

Figure 92 Adult male agrimi (Capra aegagrus cretica) showing phenotypic resemblance to the wild bezoar goat 177

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

Frontispiece The city the fortifications the harbor and the hinterland of Khandax (Herakleion) in the first half of the 17th century Map by unknown cartographer 17th c Collezione Museo Civico Padua Vikelaia Municipal Library Herakleion ii

Figure i The Minoan palace and its Neolithic past xxi

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation 2

Figure 12 Trench II (a) view of the Central Court of the Palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the souther profile 3

Figure 13 Plan of the excavation trenches next to the staircase 3

Figure 14 South and west stratigraphic profiles of the trench 4

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench 5

Figure 16 Plans of excavation levels 9 and 10 7

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner 8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner 8

List of Figures in the Text

xiv THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 19 Plan of excavation level 13 9

Figure 110 Plan of excavation level 14 showing hearth in northwest corner 9

Figure 111 View of excavation level 14 showing hearth in northwest corner 10

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south 11

Figure 113 View of excavation level 15 from above 12

Figure 114 View of excavation level 16 from above 12

Figure 115 View of excavation level 16 facing west section 12

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4 12

Figure 117 View of level 17 facing west section 13

Figure 118 View of excavation level 18 from above 13

Figure 119 View of excavation level 19 from above 13

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6 13

Figure 121 View of excavation level 21 from above 14

Figure 122 View of excavation level 23 14

Figure 123 View of excavation level 24 showing wall 7 and grinding stones 14

Figure 124 Plans of excavation levels 22 and 24 15

Figure 125 Plan of excavation level 27 16

Figure 126 View of excavation level 28 from above 16

Figure 127 View of excavation level 28 facing west profile 16

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths 17

Figure 129 View of level 29A showing hearths 1 2 and 3 18

Figure 130 View of excavation level 30 showing hearth 4 18

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure 19

Figure 132 View of the elliptical stone wall from levels 24ndash27 19

Figure 133 View of excavation level 31 19

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7 20

Figure 135 Plan of excavation level 37 showing pits 1 and 2 21

Figure 21 Sedimentological samples of the middle part of the south profile 26

Figure 41 Fine fraction granulometry () of the samples from the west profile 57

Figure 42 The organic content () of the samples from the west profile 57

Figure 43 The carbonate content of the samples from the west profile 57

xvLIST OF FIGURES IN THE TEXT

Figure 44 Morphoscopy of sands without acid treatment 57

Figure 45 Morphoscopy of sands after the elimination of calcareous grains (subsequent to acid treatment) 57

Figure 51 Drawing of Triticum turgidum LT aestivum from the 1997 excavations at Knossos 69

Figure 52 Graphs of measurements and measurement ratios of Triticum turgidumaestivum from Neolithic Knossos 72

Figure 53 Measurements of Triticum turgidumaestivum from Aceramic and Early Neolithic Knossos compared with average values for Erbaba Ramad and Bouqras in the Near East 73

Figure 54 Graphs of measurements and measurement ratios of Lens culinaris from Neolithic Knossos 74

Figure 55 Early sites including those from mainland Greece where Triticum turgidumaestivum is reported 1 Tell Abu Hureyra 2 Tell Halula 3 Tell Aswad 4 Tell Ghoraife 5 Tell Sabi Abyad 6 Servia 7 Cafer Houmlyuumlk 8 Dhali Agridhi 9 Otzaki 10 Sesklo 11 Sitagroi 12 Haccedililar 13 Aşkli Houmlyuumlk 14 Ccedilatal Houmlyuumlk 15 Can Hasan 16 Cayoumlnuuml 17 El Kown 18 Bouqras 19 Tell Ramad 89

Figure 56 Summary of the distribution of all categories of archaeobotanical remains at Neolithic Knossos 90

Figure 61 Climate and topography of Knossos (a) mean annual precipitation in Crete (b) topographic map of the area around Knossos (c) westndasheast topographic section (d) southwestndashnortheast topographic section 96

Figure 62 View of the Knossos valley from Mt Juktas showing present-day vegetation 98

Figure 63 Panoramic view of the site of Knossos showing present-day vegetation 98

Figure 64 Present-day phrygana vegetation on the hills in the study area 98

Figure 65 Present-day vegetation on deep soils in the study area 98

Figure 66 Anatomy of plant taxa identified in wood charcoal assemblages from Neolithic Knossos 106

Figure 67 Wood charcoal diagram from Neolithic Knossos showing relative frequencies of taxa in successive excavation levels 108

Figure 71 Bar chart of phytolith per centage frequencies from the south profile 122

Figure 72 Bar chart of C3 and C

4 phytolith percentage frequencies from the south profile 122

Figure 73 West profile stratigraphy and sampling 123

Figure 74 Bar chart of phytolith percentage frequencies from the west profile 124

Figure 75 Bar chart of C3 and C

4 phytolith percentage frequencies from the west profile 125

Figure 76 Trench section (southwest corner to west face) with phases identified according to phytolith composition and frequencies 127

xvi THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 77 Silica skeleton from grass leaf (long cells and a stoma) from the EN I deposits (sample XXa level 32) 129

Figure 78 Wheat-type silica skeleton from the EN I deposits (sample XXIVb level 32) 129

Figure 79 Silica skeleton from a dicotyledonous plant from the EN I deposits (sample XIV level 30) 129

Figure 710 Millet-type silica skeleton from the EN II deposits (sample Xa level 16) 129

Figure 81 Percentages of the osseous parts of cattle long bones 139

Figure 82 Percentages of the osseous parts belonging to the long bones of middle-sized mammals (goats sheep and pigs) 139

Figure 83 Skeletal fragments of long bones of OvisCapra from level 14 all with dog-gnawing marks 140

Figure 84 Fragments of proximal epiphyses of femur and tibia of Bos taurus with fracture marks caused by impacts from the extraction of marrow level 24 151

Figure 85 Animal bones from level 3 (a) radius diaphysis (b) radius proximal part (c) scapula (d) fragment of femur (e) phalanx I of Capra aegagrus 153

Figure 86 Distal metacarpus of Capra aegagrus and Ovis aries 153

Figure 87 Animal bones (a) ulna in lateral view probably belonging to a wild boar (level 23) (b) Sus scrofa ferus canine fragment (level 10) (c) Sus scrofa domesticus ulna in lateral view (level 14) 154

Figure 88 Diaphysis width range (SD) of Sus scrofa domesticus and Sus scrofa ferus from Zambujal (Portugal) Cerro de la Virgen (Spain) Argissa-Magula and Knossos 154

Figure 89 Meles meles (a) left mandible in lateral view (level 14) (b) lower canine (level 14) (c) left ulna in medial views the proximal epiphysis is not fused (level 3) Martes (d) distal part of humerus in cranial view (level 9) 155

Figure 810 Age classes of the mandibles of Ovis and Capra 157

Figure 811 Age classes of the mandibles of Bos taurus 158

Figure 812 Distal part of metacarpus belonging to a male (possibly ox) of Bos taurus with osseous deformations on the articular surfaces 158

Figure 813 Age classes of Sus scrofa domesticus maxillae and mandibles 159

Figure 814 Correlation of the measurements of phalanx I belonging to Bos taurus 160

Figure 91 Map showing location of sites mentioned in the text 1 Ashkelon 2 lsquoAin Ghazal 3 Atlit Yam 4 Hagoshrim and Tel Ali 5 Ras Shamra 6 Cap Andreas Kastros 7 Khirokitia 8 Tenta 9 Asikli Houmlyuumlk 10 Mersin 11 Can Hasan III 12 Ccedilatalhoumlyuumlk 13 Suberde 14 Haccedililar 15 Nea Nikomedeia 16 Argissa-Magula 17 Sesklo 18 Achilleion 19 Franchthi Cave 20 Sidari Corfu 21 Cave of the Cyclops Youra 22 Melos 23 Santorini 24 Knossos Crete 25 Tel Aray 2 26 Umm el Tlel 27 Qdeir 28 El Kowm 2 173

Figure 92 Adult male agrimi (Capra aegagrus cretica) showing phenotypic resemblance to the wild bezoar goat 177

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

xiv THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 19 Plan of excavation level 13 9

Figure 110 Plan of excavation level 14 showing hearth in northwest corner 9

Figure 111 View of excavation level 14 showing hearth in northwest corner 10

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south 11

Figure 113 View of excavation level 15 from above 12

Figure 114 View of excavation level 16 from above 12

Figure 115 View of excavation level 16 facing west section 12

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4 12

Figure 117 View of level 17 facing west section 13

Figure 118 View of excavation level 18 from above 13

Figure 119 View of excavation level 19 from above 13

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6 13

Figure 121 View of excavation level 21 from above 14

Figure 122 View of excavation level 23 14

Figure 123 View of excavation level 24 showing wall 7 and grinding stones 14

Figure 124 Plans of excavation levels 22 and 24 15

Figure 125 Plan of excavation level 27 16

Figure 126 View of excavation level 28 from above 16

Figure 127 View of excavation level 28 facing west profile 16

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths 17

Figure 129 View of level 29A showing hearths 1 2 and 3 18

Figure 130 View of excavation level 30 showing hearth 4 18

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure 19

Figure 132 View of the elliptical stone wall from levels 24ndash27 19

Figure 133 View of excavation level 31 19

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7 20

Figure 135 Plan of excavation level 37 showing pits 1 and 2 21

Figure 21 Sedimentological samples of the middle part of the south profile 26

Figure 41 Fine fraction granulometry () of the samples from the west profile 57

Figure 42 The organic content () of the samples from the west profile 57

Figure 43 The carbonate content of the samples from the west profile 57

xvLIST OF FIGURES IN THE TEXT

Figure 44 Morphoscopy of sands without acid treatment 57

Figure 45 Morphoscopy of sands after the elimination of calcareous grains (subsequent to acid treatment) 57

Figure 51 Drawing of Triticum turgidum LT aestivum from the 1997 excavations at Knossos 69

Figure 52 Graphs of measurements and measurement ratios of Triticum turgidumaestivum from Neolithic Knossos 72

Figure 53 Measurements of Triticum turgidumaestivum from Aceramic and Early Neolithic Knossos compared with average values for Erbaba Ramad and Bouqras in the Near East 73

Figure 54 Graphs of measurements and measurement ratios of Lens culinaris from Neolithic Knossos 74

Figure 55 Early sites including those from mainland Greece where Triticum turgidumaestivum is reported 1 Tell Abu Hureyra 2 Tell Halula 3 Tell Aswad 4 Tell Ghoraife 5 Tell Sabi Abyad 6 Servia 7 Cafer Houmlyuumlk 8 Dhali Agridhi 9 Otzaki 10 Sesklo 11 Sitagroi 12 Haccedililar 13 Aşkli Houmlyuumlk 14 Ccedilatal Houmlyuumlk 15 Can Hasan 16 Cayoumlnuuml 17 El Kown 18 Bouqras 19 Tell Ramad 89

Figure 56 Summary of the distribution of all categories of archaeobotanical remains at Neolithic Knossos 90

Figure 61 Climate and topography of Knossos (a) mean annual precipitation in Crete (b) topographic map of the area around Knossos (c) westndasheast topographic section (d) southwestndashnortheast topographic section 96

Figure 62 View of the Knossos valley from Mt Juktas showing present-day vegetation 98

Figure 63 Panoramic view of the site of Knossos showing present-day vegetation 98

Figure 64 Present-day phrygana vegetation on the hills in the study area 98

Figure 65 Present-day vegetation on deep soils in the study area 98

Figure 66 Anatomy of plant taxa identified in wood charcoal assemblages from Neolithic Knossos 106

Figure 67 Wood charcoal diagram from Neolithic Knossos showing relative frequencies of taxa in successive excavation levels 108

Figure 71 Bar chart of phytolith per centage frequencies from the south profile 122

Figure 72 Bar chart of C3 and C

4 phytolith percentage frequencies from the south profile 122

Figure 73 West profile stratigraphy and sampling 123

Figure 74 Bar chart of phytolith percentage frequencies from the west profile 124

Figure 75 Bar chart of C3 and C

4 phytolith percentage frequencies from the west profile 125

Figure 76 Trench section (southwest corner to west face) with phases identified according to phytolith composition and frequencies 127

xvi THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 77 Silica skeleton from grass leaf (long cells and a stoma) from the EN I deposits (sample XXa level 32) 129

Figure 78 Wheat-type silica skeleton from the EN I deposits (sample XXIVb level 32) 129

Figure 79 Silica skeleton from a dicotyledonous plant from the EN I deposits (sample XIV level 30) 129

Figure 710 Millet-type silica skeleton from the EN II deposits (sample Xa level 16) 129

Figure 81 Percentages of the osseous parts of cattle long bones 139

Figure 82 Percentages of the osseous parts belonging to the long bones of middle-sized mammals (goats sheep and pigs) 139

Figure 83 Skeletal fragments of long bones of OvisCapra from level 14 all with dog-gnawing marks 140

Figure 84 Fragments of proximal epiphyses of femur and tibia of Bos taurus with fracture marks caused by impacts from the extraction of marrow level 24 151

Figure 85 Animal bones from level 3 (a) radius diaphysis (b) radius proximal part (c) scapula (d) fragment of femur (e) phalanx I of Capra aegagrus 153

Figure 86 Distal metacarpus of Capra aegagrus and Ovis aries 153

Figure 87 Animal bones (a) ulna in lateral view probably belonging to a wild boar (level 23) (b) Sus scrofa ferus canine fragment (level 10) (c) Sus scrofa domesticus ulna in lateral view (level 14) 154

Figure 88 Diaphysis width range (SD) of Sus scrofa domesticus and Sus scrofa ferus from Zambujal (Portugal) Cerro de la Virgen (Spain) Argissa-Magula and Knossos 154

Figure 89 Meles meles (a) left mandible in lateral view (level 14) (b) lower canine (level 14) (c) left ulna in medial views the proximal epiphysis is not fused (level 3) Martes (d) distal part of humerus in cranial view (level 9) 155

Figure 810 Age classes of the mandibles of Ovis and Capra 157

Figure 811 Age classes of the mandibles of Bos taurus 158

Figure 812 Distal part of metacarpus belonging to a male (possibly ox) of Bos taurus with osseous deformations on the articular surfaces 158

Figure 813 Age classes of Sus scrofa domesticus maxillae and mandibles 159

Figure 814 Correlation of the measurements of phalanx I belonging to Bos taurus 160

Figure 91 Map showing location of sites mentioned in the text 1 Ashkelon 2 lsquoAin Ghazal 3 Atlit Yam 4 Hagoshrim and Tel Ali 5 Ras Shamra 6 Cap Andreas Kastros 7 Khirokitia 8 Tenta 9 Asikli Houmlyuumlk 10 Mersin 11 Can Hasan III 12 Ccedilatalhoumlyuumlk 13 Suberde 14 Haccedililar 15 Nea Nikomedeia 16 Argissa-Magula 17 Sesklo 18 Achilleion 19 Franchthi Cave 20 Sidari Corfu 21 Cave of the Cyclops Youra 22 Melos 23 Santorini 24 Knossos Crete 25 Tel Aray 2 26 Umm el Tlel 27 Qdeir 28 El Kowm 2 173

Figure 92 Adult male agrimi (Capra aegagrus cretica) showing phenotypic resemblance to the wild bezoar goat 177

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

xvLIST OF FIGURES IN THE TEXT

Figure 44 Morphoscopy of sands without acid treatment 57

Figure 45 Morphoscopy of sands after the elimination of calcareous grains (subsequent to acid treatment) 57

Figure 51 Drawing of Triticum turgidum LT aestivum from the 1997 excavations at Knossos 69

Figure 52 Graphs of measurements and measurement ratios of Triticum turgidumaestivum from Neolithic Knossos 72

Figure 53 Measurements of Triticum turgidumaestivum from Aceramic and Early Neolithic Knossos compared with average values for Erbaba Ramad and Bouqras in the Near East 73

Figure 54 Graphs of measurements and measurement ratios of Lens culinaris from Neolithic Knossos 74

Figure 55 Early sites including those from mainland Greece where Triticum turgidumaestivum is reported 1 Tell Abu Hureyra 2 Tell Halula 3 Tell Aswad 4 Tell Ghoraife 5 Tell Sabi Abyad 6 Servia 7 Cafer Houmlyuumlk 8 Dhali Agridhi 9 Otzaki 10 Sesklo 11 Sitagroi 12 Haccedililar 13 Aşkli Houmlyuumlk 14 Ccedilatal Houmlyuumlk 15 Can Hasan 16 Cayoumlnuuml 17 El Kown 18 Bouqras 19 Tell Ramad 89

Figure 56 Summary of the distribution of all categories of archaeobotanical remains at Neolithic Knossos 90

Figure 61 Climate and topography of Knossos (a) mean annual precipitation in Crete (b) topographic map of the area around Knossos (c) westndasheast topographic section (d) southwestndashnortheast topographic section 96

Figure 62 View of the Knossos valley from Mt Juktas showing present-day vegetation 98

Figure 63 Panoramic view of the site of Knossos showing present-day vegetation 98

Figure 64 Present-day phrygana vegetation on the hills in the study area 98

Figure 65 Present-day vegetation on deep soils in the study area 98

Figure 66 Anatomy of plant taxa identified in wood charcoal assemblages from Neolithic Knossos 106

Figure 67 Wood charcoal diagram from Neolithic Knossos showing relative frequencies of taxa in successive excavation levels 108

Figure 71 Bar chart of phytolith per centage frequencies from the south profile 122

Figure 72 Bar chart of C3 and C

4 phytolith percentage frequencies from the south profile 122

Figure 73 West profile stratigraphy and sampling 123

Figure 74 Bar chart of phytolith percentage frequencies from the west profile 124

Figure 75 Bar chart of C3 and C

4 phytolith percentage frequencies from the west profile 125

Figure 76 Trench section (southwest corner to west face) with phases identified according to phytolith composition and frequencies 127

xvi THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 77 Silica skeleton from grass leaf (long cells and a stoma) from the EN I deposits (sample XXa level 32) 129

Figure 78 Wheat-type silica skeleton from the EN I deposits (sample XXIVb level 32) 129

Figure 79 Silica skeleton from a dicotyledonous plant from the EN I deposits (sample XIV level 30) 129

Figure 710 Millet-type silica skeleton from the EN II deposits (sample Xa level 16) 129

Figure 81 Percentages of the osseous parts of cattle long bones 139

Figure 82 Percentages of the osseous parts belonging to the long bones of middle-sized mammals (goats sheep and pigs) 139

Figure 83 Skeletal fragments of long bones of OvisCapra from level 14 all with dog-gnawing marks 140

Figure 84 Fragments of proximal epiphyses of femur and tibia of Bos taurus with fracture marks caused by impacts from the extraction of marrow level 24 151

Figure 85 Animal bones from level 3 (a) radius diaphysis (b) radius proximal part (c) scapula (d) fragment of femur (e) phalanx I of Capra aegagrus 153

Figure 86 Distal metacarpus of Capra aegagrus and Ovis aries 153

Figure 87 Animal bones (a) ulna in lateral view probably belonging to a wild boar (level 23) (b) Sus scrofa ferus canine fragment (level 10) (c) Sus scrofa domesticus ulna in lateral view (level 14) 154

Figure 88 Diaphysis width range (SD) of Sus scrofa domesticus and Sus scrofa ferus from Zambujal (Portugal) Cerro de la Virgen (Spain) Argissa-Magula and Knossos 154

Figure 89 Meles meles (a) left mandible in lateral view (level 14) (b) lower canine (level 14) (c) left ulna in medial views the proximal epiphysis is not fused (level 3) Martes (d) distal part of humerus in cranial view (level 9) 155

Figure 810 Age classes of the mandibles of Ovis and Capra 157

Figure 811 Age classes of the mandibles of Bos taurus 158

Figure 812 Distal part of metacarpus belonging to a male (possibly ox) of Bos taurus with osseous deformations on the articular surfaces 158

Figure 813 Age classes of Sus scrofa domesticus maxillae and mandibles 159

Figure 814 Correlation of the measurements of phalanx I belonging to Bos taurus 160

Figure 91 Map showing location of sites mentioned in the text 1 Ashkelon 2 lsquoAin Ghazal 3 Atlit Yam 4 Hagoshrim and Tel Ali 5 Ras Shamra 6 Cap Andreas Kastros 7 Khirokitia 8 Tenta 9 Asikli Houmlyuumlk 10 Mersin 11 Can Hasan III 12 Ccedilatalhoumlyuumlk 13 Suberde 14 Haccedililar 15 Nea Nikomedeia 16 Argissa-Magula 17 Sesklo 18 Achilleion 19 Franchthi Cave 20 Sidari Corfu 21 Cave of the Cyclops Youra 22 Melos 23 Santorini 24 Knossos Crete 25 Tel Aray 2 26 Umm el Tlel 27 Qdeir 28 El Kowm 2 173

Figure 92 Adult male agrimi (Capra aegagrus cretica) showing phenotypic resemblance to the wild bezoar goat 177

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

xvi THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Figure 77 Silica skeleton from grass leaf (long cells and a stoma) from the EN I deposits (sample XXa level 32) 129

Figure 78 Wheat-type silica skeleton from the EN I deposits (sample XXIVb level 32) 129

Figure 79 Silica skeleton from a dicotyledonous plant from the EN I deposits (sample XIV level 30) 129

Figure 710 Millet-type silica skeleton from the EN II deposits (sample Xa level 16) 129

Figure 81 Percentages of the osseous parts of cattle long bones 139

Figure 82 Percentages of the osseous parts belonging to the long bones of middle-sized mammals (goats sheep and pigs) 139

Figure 83 Skeletal fragments of long bones of OvisCapra from level 14 all with dog-gnawing marks 140

Figure 84 Fragments of proximal epiphyses of femur and tibia of Bos taurus with fracture marks caused by impacts from the extraction of marrow level 24 151

Figure 85 Animal bones from level 3 (a) radius diaphysis (b) radius proximal part (c) scapula (d) fragment of femur (e) phalanx I of Capra aegagrus 153

Figure 86 Distal metacarpus of Capra aegagrus and Ovis aries 153

Figure 87 Animal bones (a) ulna in lateral view probably belonging to a wild boar (level 23) (b) Sus scrofa ferus canine fragment (level 10) (c) Sus scrofa domesticus ulna in lateral view (level 14) 154

Figure 88 Diaphysis width range (SD) of Sus scrofa domesticus and Sus scrofa ferus from Zambujal (Portugal) Cerro de la Virgen (Spain) Argissa-Magula and Knossos 154

Figure 89 Meles meles (a) left mandible in lateral view (level 14) (b) lower canine (level 14) (c) left ulna in medial views the proximal epiphysis is not fused (level 3) Martes (d) distal part of humerus in cranial view (level 9) 155

Figure 810 Age classes of the mandibles of Ovis and Capra 157

Figure 811 Age classes of the mandibles of Bos taurus 158

Figure 812 Distal part of metacarpus belonging to a male (possibly ox) of Bos taurus with osseous deformations on the articular surfaces 158

Figure 813 Age classes of Sus scrofa domesticus maxillae and mandibles 159

Figure 814 Correlation of the measurements of phalanx I belonging to Bos taurus 160

Figure 91 Map showing location of sites mentioned in the text 1 Ashkelon 2 lsquoAin Ghazal 3 Atlit Yam 4 Hagoshrim and Tel Ali 5 Ras Shamra 6 Cap Andreas Kastros 7 Khirokitia 8 Tenta 9 Asikli Houmlyuumlk 10 Mersin 11 Can Hasan III 12 Ccedilatalhoumlyuumlk 13 Suberde 14 Haccedililar 15 Nea Nikomedeia 16 Argissa-Magula 17 Sesklo 18 Achilleion 19 Franchthi Cave 20 Sidari Corfu 21 Cave of the Cyclops Youra 22 Melos 23 Santorini 24 Knossos Crete 25 Tel Aray 2 26 Umm el Tlel 27 Qdeir 28 El Kowm 2 173

Figure 92 Adult male agrimi (Capra aegagrus cretica) showing phenotypic resemblance to the wild bezoar goat 177

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

xviiLIST OF FIGURES IN THE TEXT

Figure 101 Distribution of calibrated dates sorted by stratum of the samples from the excavations of JD Evans 198

Figure 102 Distribution of calibrated dates sorted by depth of the samples from the 1997 archaeological campaign 198

Figure 103 Calibrated radiocarbon dates from the 1997 excavation at Knossos plotted against the depth of the samples in order to determine the accumulation rate of the habitation deposits 198

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

Preface

The site of Knossos on the Kephala hill in Crete is of great archaeological and historical importance for Greece and Europe Dating back to 7000 bc it is the home of one of the earliest farming societies in southeastern Europe In later Bronze Age periods it developed into a remarkable center of econom-ic and social organization within the island enjoy-ing extensive relations with the Aegean the Greek mainland the Near East and Egypt Arthur Evans excavated the site at the beginning of the 20th cen-tury and through his extensive and spectacular res-toration and reconstruction efforts he transformed Knossos into one of the most popular archaeological sites in the Old World (Evans 1901 1921ndash1935 1927 1928) Knossos is now best known among both specialists and the wider public for its unique cen-tral building conventionally called a palace which is one of the earliest archaeological monuments to have been restored on such a scale

What was not apparent during the early ar-chaeological research at the site was the impres-sive extent and depth of the earlier habitation that lies under the imposing palace even though

the laborious work of Arthur Evans and Duncan Mackenzie in the early 20th century had re-vealed considerable amounts of Neolithic mate-rial (Mackenzie 1903) In 1953 Audrey Furness studied and published the Neolithic pottery from Evansrsquos test soundings with the aim of testing the three ldquoStone Agerdquo periods discussed by Mackenzie (Furness 1953) The successful work of Furness led the British School at Athens to launch a se-ries of systematic investigations at Knossos di-rected by Sinclair Hood and John D Evans from 1956 to 1971 (Evans 1964 1971 1994 Warren et al 1968) The well-known Trenches A to C which were opened in the area of the Central Court of the palace together with the peripheral soundings X and ZE confirmed a chronological sequence of 10 strata representing at least 4000 years of Neolithic occupation including the still-disputed Aceramic phase Looking back at the announcement by JD Evans (1971) of the first and very early radiocar-bon dates for the founding of Knossos (7000 bc) I cannot forget the welcome surprise with which these dates were received and I am very happy to

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

xx THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

see that our recent radiocarbon dates published in this volume confirm Evansrsquos early chronology that was attained without the benefit of our mod-ern technology

Other contributions to our knowledge of the Neolithic of Crete include the work of Richard M Dawkins at Magasa in eastern Crete in 1905 (Dawkins 1905) the investigations of Angelo Mosso and Doro Levi at Phaistos (Mosso 1908) the publication of the Phaistos material by Lucia Vagnetti (Vagnetti 1972ndash1973) and the pioneering research at Katsambas by Stylianos Alexiou (1953 1954) The forthcoming publication of Katsambas by Nena Galanidou and her associates (Galanidou ed forthcoming) and the study of the materi-al from older fieldwork at Gerani and Pelekita in the Zakros area carried out by Yiannis Tzedakis and Costis Davaras respectively (Tzedakis 1970 Davaras 1979) are expected to offer more data re-garding the early occupational horizon of Crete The recent publication by Valasia Isaakidou and Peter Tomkins of The Cretan Neolithic in Context (Isaakidou and Tomkins eds 2008) the latest rescue excavations carried out by the Ephorate of Central Crete in the vicinity of Katsambas and most importantly the announced presence of Mesolithic material on the islands of Crete and Gavdos show that early prehistoric research in Crete and its immediate environs is a dynamic field of investigation

A series of archaeological test soundings was opened in February 1997 in conjunction with the planning of the course of the main and secondary visitorsrsquo routes through the palace a process that involved widening the existing paths establishing new ones and examining the state of the build-ingrsquos foundations The south and east slopes of the Kephala hill were the main focus of investigation (Karetsou 2004 Ioannidou-Karetsou 2006) This research was prompted by the architect Clairy Palyvoursquos suggestion to double the width of the modern narrow stone stair leading from this part of the Central Court to the first level of the Grand Staircase where A Evans made his last attempt to restore the Medallion Pithoi The investigation which lasted five weeks was carried out under dif-ficult weather conditions and according to a very strict timetable

We were all happily surprised that in an area often disturbed for conservation work in the 1950s

and 1960s including the opening of rainwater channels deep pre-Minoan deposits remained in-tact just a few centimeters under the visitorsrsquo feet I took this to be a sign of good fortune since after three decades of personal systematic involvement with Minoan archaeology the dream of my youth to look down to the ldquoNeolithic Cretan timerdquo was becoming a reality

A collaboration with colleagues familiar with the excavation of Neolithic sites and modern data col-lection and analysis methods was my next immedi-ate concern The chance to reexamine the succession of Neolithic occupation strata on the Kephala hill-top some 50 years after the first such investigation at Knossos presented me with great expectations and challenges Professor Nikos Efstratiou of the Aristotle University of Thessaloniki contributed greatly to the success of the project and I would like to take this opportunity to thank him He was responsible both for the selection of the research-ers who gathered at Knossos with very short notice that February and for the coordination of the proj-ect In addition Professor Giorgos Hourmouziadis also of the Aristotle Uni versity of Thessaloniki was very helpful

Many thanks are due as well to my colleague Dr Eleni Banou who participated in the excavation on behalf of the Ephorate to Nikos Daskalakis the skilled foreman of the Knossos project and to the late Andreas Klinis also a Knossos foreman and a man of rare excavation experience

The general aims of the investigation in the Central Court of Knossos in 1997 were (1) to read-dress questions related to the old material and con-clusions reached many years ago and (2) to obtain new data which considering the nature of the ar-chaeological site with the palace standing on top of the Neolithic tell would have been otherwise impossible More specific objectives included the careful study of the stratigraphy for the confirma-tion or revision of the already established Neolithic sequence the determination of whether the alleged Aceramic phase was represented the collection of new evidence for the Neolithic ceramic sequence and the recovery of new archaeozoological and ar-chaeobotanical data and the analysis of their strati-graphic distribution (Efstratiou et al 2004) Most importantly the archaeological information was to be gathered and studied using methodologies that were not available in the pastmdashsedimentological

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

PREFACE xxi

analyses which might clarify the occupational gaps in the impressive Neolithic palimpsest phy-tolith analyses ceramic technological analyses paleoenvironmental observations and most signif-icantly new radiocarbon analyses for the establish-ment of a reliable sequence of dates

The many archaeological questions relating to the long Neolithic habitation of the Knossos tell had always intrigued me especially during my 12 years of service (1992ndash2004) as head of the Knossos Conservation Project I was impressed by the extent of the Neolithic settlement and the density of the scattered material especially that of

the Late and Final Neolithic periods (Fig i) I was enormously pleased by the opportunity we had to investigate this early Cretan farming communi-ty buried deep under the glorious Minoan palace and to contribute to its understanding There is no doubt that the Knossos Neolithic settlementmdashwhether or not it was the first and only one in Cretemdashconstitutes one of the earliest agricultur-al communities in Greece and it is also surely the earliest in the Aegean islands

Alexandra KaretsouHonorary Ephor of Antiquities

a b c d e f g h i j k l m n o p q r s

A

B

C

D

E

F

G

H

I

J

K

L

M

N

O

P

Q

5280

5270

5260

5250

5240

5230

5220

5210

5200

5190

5180

5170

5160

5150

5140

5130

1765

0

1766

0

1767

0

1768

0

1769

0

1770

0

1771

0

1772

0

1773

0

1774

0

1775

0

1776

0

1777

0

1778

0

1779

0

1780

0

1781

0

1782

0

1783

0

1784

0

Figure i The Minoan palace and its Neolithic past areas where Neolithic deposits and ceramics are found are indicated with black dots (1997 ndash2004)

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxii

ReferencesAlexiou S 1953 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 108 [1956] pp 299ndash308

mdashmdashmdash 1954 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo Prakt 109 [1957] pp 369ndash374

Davaras C 1979 ldquoΣπήλαιο Πελεκητών Ζάκρουrdquo ArchDelt 34 (B 2 Chronika) pp 402ndash404

Dawkins RM 1905 ldquoExcavations at Palaikastro IV2 Neolithic Settlement at Magasaacuterdquo BSA 11 pp 260ndash268

Efstratiou N A Karetsou E Banou and D Margomenou 2004 ldquoThe Neolithic Settlement of Knossos New Light on an Old Picturerdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 39ndash51

Evans AJ 1901 ldquoThe Neolithic Settlement at Knossos and Its Place in the History of Early Aegean Culturerdquo Man 1 pp 184ndash186

mdashmdashmdash 1921ndash1935 The Palace of Minos at Knossos IndashIV London

mdashmdashmdash 1927 ldquoWork of Reconstruction in the Palace of Knossosrdquo AntJ 7 pp 258ndash266

mdashmdashmdash 1928 ldquoThe Palace of Knossos and Its Dependencies in the Light of Recent Discoveries and Reconstructionsrdquo Journal of the Royal Institute of British Architects 36 pp 90ndash102

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Galanidou N ed Forthcoming The Neolithic Settlement by the River Kairatos The Alexiou Excavations at Katsamba

Ioannidou-Karetsou A 2006 ldquoΑπό την Κνωσό μέχρι τη Ζάκρο H περιπέτεια της προστασίας των ιστορι-κών αρχαιολογικών χώρων στην κεντρική και ανα-τολική Κρήτηrdquo in Conservation and Preservation of the Cultural and Natural Heritage of the Large Islands of the Mediterranean V Karageorghis and A Giannikouri eds Athens pp 61ndash76

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

Karetsou A 2004 ldquoKnossos after Evans Past In ter-ventions Present State and Future Solutionsrdquo in Knossos Palace City State Proceedings of the Conference in Herakleion Organised by the British School of Athens and the 23rd Ephoreia of Prehistoric and Classical Antiquities in November 2000 for the Centenary of Sir Arthur Evansrsquos Excavations at Knossos (BSA Studies 12) G Cadogan E Hatzaki and A Vasilakis eds London pp 547ndash555

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

Mosso A 1908 ldquoCeramica neolitica di Phaestos e vasi dellrsquoepoca minoica primitivardquo MonAnt 19 pp 142ndash228

Tzedakis I 1970 ldquoἈρχαιλογική Ἐρεύvε Ἀνασκαφί Σπήλαιο Γερανίουrdquo ArchDelt 25 (B 2 Chronika) pp 474ndash476

Vagnetti L 1972ndash1973 ldquoLrsquoinsediamento neolitico di Festogravesrdquo ASAtene 34ndash35 pp 7ndash138

Warren P MR Jarman HM Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

Acknowledgments

The excavators are grateful to all the people who made this publication possible First and foremost we would like to express our thanks to Dr Iordanis Dimakopoulos former Director of the Conservation and Restoration of Monuments Service of the Greek Ministry of Culture and Tourism He fully understood the need for the rescue excavation to be carried out at a time when the visitorsrsquo walkway project at the pal-ace of Knossos was already under way with a tight deadline

Our warmest thanks are also due to the technical staff of the 23rd Ephorate of Prehistoric and Classical Antiquities and the Knossos Conservation Office who worked through the particularly cold February of 2011 We would especially like to thank Nikos Daskalakis Stavros Mavrakis and Michalis Tzobanakis who construct-ed a small shelter to protect both staff and trenches from the rain since the excava-tion ran from sunrise to sunset The late Andreas Klinis a foreman of special skill and astuteness was the person whom we entrusted with the stratigraphy he was the only one to work at a depth of four to eight meters We must also thank Konstantinos Ktistakis for his accurate plans elevations and sections of the trenches Dr Don Evely former Knossos curator for the British School at Athens for his help during the study of the material and archaeologist Maria Kelaidi who spent an entire sum-mer in the courtyard of the Villa Ariadne meticulously sieving the huge amount of soil from the excavation Vital assistance was also provided by the head guard of the palace of Knossos Manolis Apostolakis and the rest of the guard staff The Ephorate accountants Evangelia Fotaki and Litsa Kafousi also provided their services without which the project would not have been possible

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

xxiv

Finally we would like to express our particular thanks for the generosity of Professor Philip Betancourt who enthusiastically supported the publication of this volume by the INSTAP Academic Press and to the Director of Publications Susan Ferrence for all her efforts in ensuring that the resulting publication was of the high-est possible standard Our warmest gratitude is also of course due to all the contrib-utors to the volume

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETE

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

Introduction

Nikos Efstratiou

The construction of a staircase extension in the northeastern part of the Central Court of the Palace of Minos at Knossos prompted the open-ing of a new excavation trench in 1997 After the systematic excavation of the deep Neolithic occu-pation levels by JD Evans in the late 1950s (1964 132) and later more limited investigations of the Prepalatial deposits undertaken primarily dur-ing restoration work no thorough exploration of the earliest occupation of the mound had been at-tempted Although our operation was to be swift and limited in extent we knew that the opening of a trench destined to reach the basal layers of the settlement offered us the opportunity to address many old and new research questions concerning the chronological socioeconomic and spatial as-pects of Cretan Neolithic society (Evans 1994 1)

Since the time of Evansrsquos research excava-tion techniques and field methods have devel-oped rapidly and a new more complex picture of late Pleistocene and early Holocene developments in the Aegean and the eastern Mediterranean has emerged The chance to reexamine the important

but inconspicuous Neolithic deposits of the Knossos tell afforded both an appealing and a demanding challenge

While the Bronze Age palace dominates the his-toriography of the site and its archaeological image the Neolithic settlement at Knossos does not hold the position it deserves in discussions of the early prehistory of the eastern Mediterranean in part be-cause of the limited research directed toward the early prehistory of Crete and the other Aegean is-lands Moreover the publication of the Neolithic settlement has been confined to a few preliminary though excellent field reports and short studies pro-duced by Professor JD Evans and his collabora-tors (1964 132 1971 95 Warren et al 1968 239) When attempted previous syntheses of this ma-terial have been either very cautious analyses of the limited data (Evans 1994 1) or provocative in-terpretations containing attractive but ill-founded speculations (Broodbank 1992 39 Whitelaw 1992 225) Additional Neolithic material recovered from later small field investigations focusing on Bronze Age deposits has been welcome but because such

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxvi

information is scarce it cannot provide the an-swers to many open questions (Manteli and Evely 1995 1)

It is fortunate that certain categories of the ar-chaeological material from Evansrsquos investigations have recently undergone detailed reexamination with respect to issues of spatial organization ce-ramic typology and technology lithics and fau-nal remains (Isaakidou and Tomkins eds 2008)

Despite these new and interesting studies howev-er the need for a better understanding of the founda-tion and development of Neolithic Knossos continues This impressive and long-lived settlementmdashone of the very few tells in Greecemdashis of paramount im-portance to the history of the eastern Mediterranean and the Near East (Berger and Guilaine 2009) Recent developments in the archaeology of Cyprus and the Aegean islands make the reevaluation of long-held concepts about this region and time peri-od all the more urgent as discussed in Chapter 11

Although a number of rigorous surface recon-naissance projects have been undertaken in Crete in the past decades Knossos remains the only early settlement known on the island (Manning 1999 469) The methodology employed in these all- period surveys was not specifically designed to lo-cate early sites however In the last few years field researchers have become increasingly critical of older methods used to identify traces of early habi-tation sites especially in view of the geomorpholog-ical complexity of coastal and island areas (Runnels 2003 121 Ammerman et al 2006 1) Until spe-cially designed surface reconnaissance projects are carried out in various coastal areas the presence of other early occupation sites in Crete remains an open possibility Thus the recently reported re-sults of the Plakias Mesolithic Survey in Crete in which a number of pre-Neolithic sites rich in lith-ic scatters were identified along the southern coast of the island do not come as a surprise (Strasser et al 2010) Indeed current research in Cyprus indi-cates that we may encounter more new and unex-pected late Pleistocene and early Holocene finds in the eastern Mediterranean (Ammerman 2011) Many older views of early habitation patterns in the Aegean islands should now be treated with skepti-cism (Cherry 1990 145)

The newly found Mesolithic habitation remains along the south coast of the island may ultimately support claims of a missing Early Neolithic (EN)

horizon in Crete In the meantime the apparent uniqueness of Knossos within the island is hard to accept in cultural terms and as we shall see in later chapters such a perception is undermined albeit indirectly by the material remains (pottery subsistence) from Knossos along with other ev-idence The key importance of Knossos howev-er for documenting the beginning of farming in the Aegean and mainland Greece whether as a distinctive stage within a westward mobility pat-tern of human groups or as a well-planned coloni-zation episode involving specific Aegean islands remains undiminished At present the notion of a local transition to farming in Crete undertaken by a dynamic Mesolithic population seems improbable as is the case in continental Greece where the ar-chaeological evidence for the arrival of new farm-ing groups seems overwhelming (Perlegraves 2001)

Neolithic Knossos is also important as sug-gested above in the wider geographic context of the early island prehistory of the eastern Mediterranean Recent discoveries on the island of Cyprus have revealed the presence of a num-ber of pre-Neolithic inland and coastal sites trig-gering an interesting debate about a possibly early date for the occupation of the largest east-ern Mediterranean islands and the interpreta-tion of this phenomenon as a historical process with its own distinctive cultural technological and ideological characteristics (Broodbank 2006 Ammerman 2010) Mounting archaeological evi-dence from the Aegean either supports or at least allows us to entertain a new picture of early is-land settlement (Sampson 2006) In this context the founding of the early seventh millennium bc farming village of Knossos on the Kephala hill may still be viewed either as the result of a long pre-Neolithic process of development on the island or as the start of an intrusive occupation by farm-ers from the east Archaeological evidence from the long stratigraphic sequence of the Knossos tell may be called upon to interpret this ambigu-ous cultural process Indeed in relation to main-land Greece specific material evidence from Knossos such as the EN sequence of pottery (fab-ric types surface treatment) attests to idiosyn-cratic elements of a local island development (see Dimitriadis this vol Ch 3) It is still too early to argue whether these characteristics should be in-terpreted as the outcome of island isolationism

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

INTRODUCTION xxvii

and endogenous developments in Crete or as the manifestation of a more generalized and long-standing Aegean island cultural tradition The for-mer would undoubtedly have resulted in a number of other distinctive material features and perhaps oddities that we may search for in the archaeolog-ical record

Both in terms of a pre-ldquohistoricalrdquo reconstruc-tion and as far as the archaeology of the site it-self is concerned our endeavor entails a constant shift between different scales (ldquomacrordquo ldquomicrordquo) and genres of field inquiry (eg use of space ra-diocarbon dating abandonment phases faunal changes pottery changes) The small size of our 1997 dig admittedly limits the overall representa-tional validity of our findings at the site but this does not deter us from addressing some of the broader issues mentioned above We are particu-larly hopeful that the new studies presented heremdashsedimentology phytoliths anthracology ceramic technologymdashtogether with the critical reevaluation of the other categories of material remains such as the fauna and archaeobotany will provide new and meaningful insights into the cultural sequence of

the Knossos settlement The documentation of the tellrsquos stratigraphic sequence which has a depth of more than 8 m along with its comparison to the old and well-established succession of Evansrsquos stra-ta (Efstratiou this vol Ch 2) also contributes to these insights as does the the newly obtained se-ries of radiocarbon dates from accelerator mass spectrometry (AMS) which seems to corroborate the existing chronological framework (Facorellis and Maniatis this vol Ch 10)

All of the categories of material remains with the exception of the pottery are analyzed and pre-sented in the following chapters of the mono-graph The detailed study of the ceramics is still in progress and will appear in a separate volume The contributors wish to underline the contingent nature of their results and syntheses which are constrained by the limited area of the field inves-tigation Nevertheless we hope that the rigor em-ployed in the data collection the meticulous study of the finds the constant cross-checking with JD Evansrsquos record and our final synthesis will bal-ance this unavoidable difficulty

ReferencesAmmerman AJ 2010 ldquoThe First Argonauts Towards the

Study of the Earliest Seafaring in the Mediterraneanrdquo in Global Origins (and Development) of Seafaring A Anderson J Barrett and K Boyle eds Cambridge pp 81ndash92

mdashmdashmdash 2011 ldquoThe Paradox of Early Voyaging in the Mediterranean and the Slowness of the Neolithic Transition between Cyprus and Italyrdquo in The Seascape in Aegean Prehistory (Monograph of the Danish Institute at Athens 14) G Vavouranakis ed Athens

Ammerman AJ P Flourentzos C McCartney J Noller and D Sorabji 2006 ldquoTwo New Early Sites on Cyprusrdquo RDAC 2006 pp 1ndash22

Berger J-F and J Guilaine 2009 ldquoThe 8200 cal bp Abrupt Environmental Change and the Neolithic Transition A Mediterranean Perspectiverdquo Quaternary International 200 pp 31ndash49

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

mdashmdashmdash 2006 ldquoThe Origins and Early Development of Mediterranean Maritime Activityrdquo JMA 19 pp 199ndash230

Cherry JF 1990 ldquoThe First Colonization of the Med-i terranean Islands A Review of Recent Researchrdquo JMA 3 pp 145ndash221

Evans JD 1964 ldquoExcavations in the Neolithic Settlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Settlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A Labyrinth of History Papers in Honour of S Hood D Evely H Hughes-Brock and N Momigliano eds Oxford pp 1ndash20

Isaakidou V and P Tomkins eds 2008 Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) Oxford

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

THE NEOLITHIC SETTLEMENT OF KNOSSOS IN CRETExxviii

Manning SW 1999 ldquoKnossos and the Limits of Settlement Growthrdquo in Meletemata Studies in Ae-gean Archaeology Presented to Malcolm H Wiener on the Occasion of His 65th Birthday (Aegeum 20) P Betancourt V Karageorghis R Laffineur and W-D Niemeier eds Liegravege pp 469ndash482

Manteli K and D Evely 1995 ldquoThe Neolithic Levels from the Throne Room System Knossosrdquo BSA 90 pp 1ndash16

Perlegraves C 2001 The Early Neolithic in Greece The First Farming Communities in Europe Cambridge

Runnels C 2003 ldquoThe Origins of the Greek Neolithic A Personal Viewrdquo in The Widening Harvest The Neo -lithic Transition in Europe Looking Back Looking Forward (Colloquia and Conference Papers 6) AJ Ammerman and P Biagi eds Boston pp 121ndash133

Sampson A 2006 Προϊστορία του Αιγαίου Athens

Strasser TF E Panagopoulou CN Runnels PM Murray N Thompson P Karkanas FW McCoy and KW Wegmann 2010 ldquoStone Age Seafaring in the Mediterranean Evidence from the Plakias Region for Lower Palaeolithic and Mesolithic Habitation of Creterdquo Hesperia 79 pp 145ndash190

Warren P MR Jarman HN Jarman NJ Shackleton and JD Evans 1968 ldquoKnossos Neolithic Part IIrdquo BSA 63 pp 239ndash276

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

1

The Excavation

Nikos Efstratiou Alexandra Karetsou and Eleni Banou

The Archaeology of the LevelsTwo trenches were opened in the northeastern

corner of the Central Court of the palace in Febshyruary of 1997 in a salvage operation that lasted for five weeks (Figs 11ndash13) Trench I a 20 x 20 m sloping area was the first to be opened but its exshycavation was stopped as soon as we realized that the area was heavily disturbed and had been used by previous investigators of the site as a dumping place for archaeological materials such as stones and broshyken pottery We then shifted our efforts to Trench II (Fig 13) The rescue character of the dig entailed a strict timetable for the completion of the work which ultimately dictated many of the methodologshyical decisions taken in the course of the excavation Moreover it was not only time that was limited but also space since the area available for investigation

was extremely small in size squeezed between the quarters of the palace (Room of Medallion Pithoi) and modern constructional remains of the Central Court staircase (Fig 12a b) Nevertheless in view of the importance of the excavation all precautions were taken to safeguard the quality of the fieldwork

At the outset Trench II covered an area of 30 by 20 m and it reached a depth of 80 m (Figs 12c 13) Due to pressure for the conclusion of the dig the excavation area below the depth of 45 m was restricted to an area of 15 x 15 m in size (Fig 14) This was an undesirable situation for many reasons including the practical difficulties of carshyrying out a dig in such a narrow and deep shaft where there was a lack of light which complicatshyed the tasks of making stratigraphic observations

All photos by N Efstratiou all drawings prepared by K Kondogiannis M Ntinou drew Fig 18 Abbreviations used in this chapter are

Ch(s) Chapter(s)EN Early Neolithic

FN Final NeolithicLN Late Neolithicm metersMN Middle Neolithicno(s) number(s)

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

nikos efstratiou alexandra karetsou and eleni banou2

Figure 11 Plan of the Palace of Knossos showing the Central Court and the location of the excavation

recording and sampling These problems became even more acute when the important lower Early Neolithic (EN) I and Aceramic stratigraphic levels (30ndash39) at 45ndash80 m of depth had to be documentshyed with the utmost accuracy Their full reconstrucshytion was attempted with particular emphasis on architectural remains and other minor but meanshyingful stratigraphic features (Fig 14) Despite all the difficulties every effort was made to preshypare a comprehensive study of the data through the systematic collection of samples from alshymost all levels for archaeobotanical faunal physhytolith and anthracological (charcoal) analyses as well as for radiocarbon dating Flotation and dry sieving were employed to retrieve archaeobotanshyical and anthracological remains These efforts proved particularly useful because they provided

material for comparative studies with the Middle (MN) and Late Neolithic (LN) upper levels of the deposit (levels 1ndash13) Moreover the sampling of all the levels of the western and southern proshyfiles of Trench II for sedimentological analyses was meticulously carried out by the late Mariashy Pilar Fumanal Garciacutea and Ernestina Badal of the University of Valencia

The deposit of Trench II covers the Neolithic settlementrsquos entire occupation from the Aceramic to the LN period A general trend observed withshyin the whole of the excavated area is the paucity of finds such as tools lithics and other portable obshyjects The Final Neolithic (FN) phase identifiable mainly by certain ceramic traits is not represented in the trench The deposit was excavated by 39 arshybitrarily defined spits hereafter called excavation

m

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

the excavation 3

Figure 12 Trench II (a) view of the Central Court of the palace looking northeast (b) view of the area of the rescue dig looking northeast (c) view looking northeast of the stratigraphy of the upper part of the trench in the southern profile the positions of the samples taken for sedimentological analysis are also shown

Figure 13 Plan of the excavation trenches next to the staircase

TRENCH I

TRENCH II

a

b

c

-000

-207 -229

5

4

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

nikos efstratiou alexandra karetsou and eleni banou4

Figure 14 South and west stratigraphic profiles of the trench with indications of depths excavation levels soil characteristics cultural periods and architectural features

1 m

2 m

LN

MN

3 m

EN II

4 m

5 m

EN I(late)

6 m

7 mEN I

8 m

ACERAMIC

floor

floor

NOTEXCAVATED

floor

1

2

3

4

910121314a

wall 1(EN II)

16

20212326 + 2728 + 29

30

31

32

33

34

35

3637

38

39

SOUTH WEST

wall 3(EN II)

Black soil with pieces of charcoal and pottery

Soft yellow limestone or kouskouras

Burned layer

Red soil

Dark brown soil

Light brown soil

Yellow limestone (washed kouskouras)

BEDROCK

LEVEL

DEPTH

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

the excavation 5

levels which varied in thickness depending on their various microstratigraphic features (soil sloping surfaces structural and spatial charactershyistics) and especially in the trenchrsquos lower section where visibility was low the digging conditions

A general description of the different excashyvation levels (1ndash39) including their main stratishygraphic occupational architectural and cultural features will follow below For intrinsic and pracshytical reasons the grouping of a number of levels has been attempted Often the individual levels were not very thick and it was difficult to excashyvate them in the deeper parts of the deposit (levels

31ndash36) Some levels have been subdivided in order to divide the excavation area into limited and wellshy defined excavation units or loci with designations such as 9A or 10B The allocation of all 39 levshyels and their features to specific chronological and cultural periods and subphases (Aceramic EN I EN II MN LN) is based on ceramic criteria and follows the sitersquos wellshyestablished typological seshyquence (Mackenzie 1903 Furness 1953 Evans 1964) A full account of the different periods will be presented in Chapter 2 where the overall seshyquence of Knossos is discussed

Excavation Stratigraphy

LeveLs 1ndash3 (LN)

No architectural remains were found in the upper part of the deposit (Fig 14) The archaeshyological deposits of levels 1 and 2 were not unishyform across the excavated area A thick layer of soft limestone (kouskouras) which was probashybly used as building material appeared in level 3 in the central area and along the southern part of the trench A row of stones began to appear in the north and west part of the excavated area These surface levels were mostly mixed with deshybris probably deriving from leveling activities in this part of the Central Court The deposit in the north part of the trench was clayish and sandy with scarce finds while the soil in its southern part was black and contained a considerable number of bones and small pieces of pottery The Neolithic pottery increased considerably in level 3 however along with food refuse (animal bones seashells) The pottery was mainly burnished and is diagnosshytic of the LN period

LeveLs 4ndash8 (wMN)

The general view of Trench II shows that severshyal successive layers of lightshycolored building mashyterial (kouskouras) and very dark soil containing archaeological remains were encountered during the excavation (Fig 12c) This material which was also observed in the upper part of the trench continued through level 4 (Fig 15) The west to east sloping pattern of these layers is clearshycut and

is apparently related to the formation of the upper part of the Neolithic tell It is a characteristic that is present in all of the stratigraphic sections from the Central Court which were published by JD Evans (1964 fig 4)

Levels 4ndash8 were marked by the appearance of the first substantial architectural remains consistshying of stones dissolved kouskouras and a number of clay ldquostructuresrdquo (Fig 15) Most of the archishytectural features appeared to continue to a conshysiderable depth reaching level 7 Unfortunately the character of these apparent structures could not be determined because of their diffused outshyline Nevertheless a hearth was detected at the northwestern end of the trench It was formed by four mediumshysized stones symmetrically arranged around a small empty space and it measured 040

Figure 15 View of excavation level 4 showing hearth in northwest corner of the trench

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

nikos efstratiou alexandra karetsou and eleni banou6

x 060 m At a slightly lower depth ranging from 198 to 218 m two postholes were also found They were situated 020 m south of the hearth They measured 050 x 053 m and were covered by a very reddish soil (Munsell 5YR 68) The postshyhole was coated with clay that was burned in placshyes probably because it came in contact with the burned pole The area marked by the hearth and the two postholes could be considered the first habitation level of the deposit One is tempted to suggest that these features were part of the inteshyrior space of a house with walls that extended beshyyond the limits of the trench They were removed during the excavation of levels 5ndash8 in the middle of the trench and they therefore are not shown in the stratigraphic section (Fig 14) The pottery that belonged to levels 4ndash8 although not particularshyly diagnostic is ascribed to the Middle Neolithic time period

LeveLs 9ndash11 (MN)

Levels 9 and 10 looked quite different from the preceding layers they appeared denser in texshyture and richer in charcoal and animal bones but not seashells (Fig 14) There was also a noticeshyable improvement in the quality of the ceramics which had intense burnishing Decorated piecesmdashmainly incised and a few rippledmdashbecame more abundant The pottery found in these levels beshylongs to the MN period

A concentration of ashes was located under a small projection that was left unexcavated in the southeastern corner of the trench (Fig 16) Its outline resembled that of a pit and its contentsmdashmainly ashesmdashwere removed for flotation (level 9A) A deposit of kouskouras excavated in the southwestern corner of the trench was designatshyed as level 11

LeveL 12 (MN)

This level covered the entire area of the trench and it was fairly thick (Figs 14 17) Some of the typical characteristics of the previous levels 9 and 10 a black layer uniform in color and texture and rich in finds remained unchanged for a considershyable depth After the complete removal of level 12 large white surfaces appeared and they were probshyably the remains of structures or floors (Fig 18)

The apparent shape of these surfaces or structures is misleading because it is the result of different degrees of preservation of the now decayed kousk-ouras that was used as building material in various places Of course this does not exclude the possishybility that some surfaces might have been part of a structure such as the round platformshylike feature made of kouskouras that was excavated as level 12A (Fig 18)

A large quantity of pottery with typical MN shapes and decoration was found

LeveL 13 (MN)

At this level the thick layer of kouskouras which extended over the entire extent of the trench was removed It was the first time that kouskouras was discovered in large lumps that did not contain any finds it was most probably a pure dissolved buildshying material The deposit varied in consistency across the trench however (Fig 19) In certain places the kouskouras had a more anthropogenic character and a brownish color due to the presence of many pieces of charcoal The interpretation of the nature and function of these surfaces is probshylematic given the general sloping of the layers and their limited exposure

The ceramic evidence suggests that level 13 marks the stratigraphic transition between the MN and EN II periods

LeveL 14 (eN II)

The thickness of Level 14 was quite substantial and it exhibited patches of heavily burned animal bones and charcoal (Fig 14) The deposit located in the western side of the trench was characterized by layers that alternated between charcoal ashes and kouskouras The gradual removal of the darkshycolored layer in the western area revealed a rathshyer solid layer of kouskouras with stones aligned in a northwest direction A hearth in the form of a round hollow space dug inside the kouskouras layer with reddish traces of burning was discovshyered close to the west profile of the trench (Figs 110 111)

The first substantial architectural remains in the trench which seem to continue down to level 27 were first noted in level 14 Their full descripshytion follows below The pottery from the levels

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

the excavation 7

Figure 16 Plans of excavation levels 9 and 10

LEVEL 9

-247-234

-209

-240

LEVEL 10

-229

-218

-234

-209

0 05 1 2 m

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

nikos efstratiou alexandra karetsou and eleni banou8

Figure 18 View of excavation level 12 showing kouskouras deposit and feature (12A) in northwest corner

Figure 17 Plan of excavation level 12 showing the round kouskouras feature (12A) in southwest corner

associated with these features including level 14 belongs to the final stages of the EN II period The only previous noticeable architectural featuresmdasha fireplace and postholesmdashbelonged to the MN peshyriod (levels 4ndash8) as discussed above

LeveLs 15ndash16 (eN II)

Level 15 comprises the soil deposit removed from the center of the trench (Fig 112) The level is situated between the burned patches of kouskour as of level 14 in the western section of the trench and a row of solid lumps of kouskourasmdashlocally called kouskouropetresmdashand hard gray stones (sideropetra) which run parallel north to south (Fig 113) The deposit of level 15 was very rich in charcoal remains as well as small stones that may have constituted a pavement

Level 16 refers to the removal of the fireplace already found in the previous level (14) along the western section of the trench together with the dissolved kouskouras material from the same area (Figs 114 115) Finds were almost nonexistent suggesting that the kouskouras was used solely as

a building material The pottery belongs to the EN II period

LeveLs 17 18 aNd 18a (eN II)

After the removal of level 16 the uncovering of the remains of the two almost parallel rows of stones running north to south in the western and

LEVEL 12

-260

-245

-286

-277

-295

050 1 2 m

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

the excavation 9

Figure 19 Plan of excavation level 13

Figure 110 Plan of excavation level 14 showing hearth in northwest corner

LEVEL 13

-294-255

-267 -302

-285

050 1

050 1 2 mLEVEL 14

-269

-289

-303-285

-331

-328

-325

2 m

hearth

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

nikos efstratiou alexandra karetsou and eleni banou10

Figure 111 View of excavation level 14 showing hearth in northwest corner (bottom left of photo)

eastern areas of the trench was completed (Fig 116) These were defined as walls 1 and 2 (030 m and 020 m wide respectively) The distance beshytween the two walls was 065 m It is difficult to know whether they were parts of one larger solid structure or if they constituted two independent stone features (Fig 117) The latter seems more plausible since wall 2 proved to have been founded deeper than wall 1 The construction material with which the upper part of wall 2 was made either piseacute or mudbrick was found dissolved all over the main central area of the trench and it had already been removed in level 15 (see above) The archishytecture in the trench thus is complex It includes a concentration of stones adjoining the west proshyfile and shown in the stratigraphic section of the trench (Fig 14) walls 1 and 2 and a long narrow corridorshylike area between the walls (Fig 118)

The pottery (largely nondiagnostic) from level 17 remained unchanged from level 14 but its volshyume was reduced significantly Level 17 seems to have been a rather pure architectural stratum It contained building material and was strikingshyly different in composition from the preceding rich black layers Wall 1 in the western part of the trench was made of large solid lumps of kouskou-ras with stones and clay (kouskourohoma) used as binding material The much narrower wall 2 differed in construction with more stones less kouskouras and the use of clay and straw piseacute as a binding material The investigation of wall 2 ended with the removal of a yellowish stershyile loose deposit with few finds in the area of the trenchrsquos eastern profile

At the base of level 17 the architectural picture of the area became even more interesting though not less enigmatic The upper part of a massive elshyliptical stone structure attached to wall 2 began to appear along with the first traces of another wall positioned south of it (Fig 116 walls 3 and 4) This massive stone structure eventually became the trenchrsquos dominant architectural feature and it is described in detail in the next section

The clayish sterile layer from the corridor beshytween walls 1 and 2 constitutes levels 18 and 18A It probably represents the remains of the upper structures of both these walls The pottery beshylongs to the EN period but more precise distincshytions cannot be made (EN II)

LeveLs 19ndash21 (eN II)

The removal of level 18 in the corridorshylike area revealed a dark blackish deposit (level 19) with pieces of charcoal and some hard finished surshyfaces probably the remains of a floor (Fig 119) In order to facilitate our work in such a restricted trench we had to remove walls 1 and 2 (level 20) after they had been recorded drawn and photoshygraphed Their construction was characterized by the use of solid lumps of kouskouras (kouskouro-petres) and hard gray stones (sideropetres) with piseacute and pure kouskouras as binding material

The exposure of level 21 revealed a densely built area in most parts of the trench (Figs 120 121) A row of stones that could have been part of a roughly made wall running eastndashwest continshyued underneath wall 1 this was designated wall 5 (Fig 120) Similarly below wall 2 a stone feashyture (wall 6) ran in a northndashsouth orientation (Fig 121) Nondiagnostic pottery occurred in small quantities

The full profile of the elliptical stone structure (wall 3) already present in level 18 (if not earlier as the upper stones were first spotted near wall 1 in level 15) was gradually revealed It included some fallen stones along the eastern section of the trench that were noted above (level 21 Fig 121) Its total exposure involved the removal of a number of levshyels (18ndash27) The elliptical structure is considered at length in the discussion of architectural remains at the end of this chapter but two points should be given special attention at present its long life and

hearth

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

the excavation 11

Figure 112 Plans of excavation levels 15 and 16 showing appearance of walls 1 and 2 running from north to south

LEVEL 15

LEVEL 16

050 1 2 m

050 1 2 m

-286

-293

-326

-323

-310-320

-316

-333

-330

-309

WALL

1 2

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

nikos efstratiou alexandra karetsou and eleni banou12

Figure 113 View of excavation level 15 from above Figure 114 View of excavation level 16 from above

Figure 115 View of excavation level 16 facing west section

Figure 116 Plan of excavation level 17 showing walls 1 and 2 and the first appearance of walls 3 and 4

careful construction It should also be stressed that the elliptical structure does not seem to have been functionally linked to the nearby wall 2

LeveLs 22ndash24 (eN II)

It was decided that the excavation should conshytinue along the western half of the trench leaving the eastern section and the elliptical structure for future investigation (Figs 122ndash124) Excavation levels 22 23 and 24 were black often burned layshyers that contained many pieces of charcoal and

050 1LEVEL 17

-345-309

-359

-371

-360

-359

-379

-375

WALLWALL

1 2

34

A

Aacute

-328

2 m

-330

-357

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

the excavation 13

Figure 117 View of level 17 facing west section

Figure 118 View of excavation level 18 from above

Figure 119 View of excavation level 19 from above

Figure 120 Plan of excavation level 21 showing walls 3 4 5 and 6

LEVEL 21050 1

WALL

WALL

5

4-366

-370

-378

3

2 m

WALL6

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

nikos efstratiou alexandra karetsou and eleni banou14

were rich in finds (Fig 122) A flimsy wall (wall 7) was discovered in the southern part of the trench together with two discarded grinding stones (Figs 123 124) The pottery was not diagnostic for the most part except for the material from level 24 which was typical of the EN II period

LeveLs 25ndash27 (eN II)

The excavation in the western part of the trench stopped at level 24 Our efforts then shifted to the eastern section where levels 25 26 and 27 were removed (Fig 125) The deposit was rich in charshycoal remains animal bones and pottery The sucshycession of levels 25ndash27 along the western profile suggested that they were hard surfaces possibly successive floors of clayish texture with dissolved piseacute and kouskouras The pottery dates to the EN II period

LeveLs 28ndash30 (eN II aNd eN I)

Level 28 comprised the deposit from the main part of the trench west of the massive stone wall (Figs 126 127) The deposit consisted of black soil rich in charcoal with large amounts of anshyimal bones but little pottery It represents a pure habitation deposit and more particularly the rich remains of a hearth made of stones (Fig 128 no 1) The contents of the hearth were carefully colshylected for flotation

The excavation of level 29 continued in the same area and revealed more such features (Figs 128 129) A second hearth (no 2) surrounded by a thick and intensely burned deposit lay in close proximity to the first one and probably represented

Figure 121 View of excavation level 21 from above

its earlier phase Its contents were collected as level 29A The remains of a third hearth (no 3) were found under the foundation of the elliptical wall All three hearths were made of stones and covered by black burned earth

Another layer of deposit from the central area of the trench excluding its eastern part where the

Figure 122 View of excavation level 23

Figure 123 View of excavation level 24 showing wall 7 and grinding stones

wall 6

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

the excavation 15

Figure 124 Plans of excavation levels 22 and 24

WALL

WALL

3 4

6

050 1 2 m

-392

-392

-395

LEVEL 22

LEVEL 24

WALL

WALL

050 1 2 m

-410

-412

-396

3 4

7

grinding stones

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

nikos efstratiou alexandra karetsou and eleni banou16

elliptical wall is situated was designated level 30 In this level a fourth large hearth (Figs 128 no 4 130) appeared directly beneath the area where nos 1 and 2 were situated obviously marking an earlier habitation phase Moreover a group of stones in a rectangular formation appeared along the southern part of the trench (see below Fig 134 these will be discussed in conjunction with the level 31) The pottery of level 30 was of exshytremely good quality with many pieces bearing inshycised and plastic decoration (EN I)

Figure 125 Plan of excavation level 27

Figure 126 View of excavation level 28 from above

Figure 127 View of excavation level 28 facing west profile

050 1 2 mLEVEL 27

WALL

3 4-420 -420

-409 -418

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

the excavation 17

Figure 128 Plans of excavation levels 29ndash29a and 30ndash30a showing walls and hearths

050 1 2 mLEVEL 29ndash29a

Hearth

HearthWALL

Hearth

3

1

2

3 4

-451

-450

29a-444

LEVEL 30ndash30a

Hearth

WALL

30a43 4

-458

050 1 2 m

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

nikos efstratiou alexandra karetsou and eleni banou18

Figure 129 View of level 29A showing hearths 1 2 and 3

The overall functional relationship between these four adjoining and successive hearths and the elliptical structure is not clear (see below for disshycussion) It should be noted that by the end of levels 28 and 29 the full profile of the elliptical structure with a total surviving height of 070 m was exshyposed (Figs 131 132)

LeveLs 31ndash36 (eN I)

After the removal of level 30 the excavation took a different course because of various difficulshyties including the presence of the massive elliptishycal wall which was occupying most of the eastern half of the trench the problematic situation along the western section where the excavation had alshyready stopped at level 18 and the pressure to comshyplete the rescue excavation in a timely manner It was decided that the excavation should focus on a more limited area in the western half of the trench

The new trench measured 150 x 150 m (Fig 14) This unfortunate decrease in the excavashytion area had serious consequences In theory the

Figure 130 View of excavation level 30 showing hearth 4

representational validity of our findings was limshyited even more especially given the already fragshymentary nature of the architectural and spatial evidence Practically we were confronted with difshyficulties in the actual digging recording and samshypling of the deeper archaeological deposits with limited light and visibility some levels ended up being quite thick (050 m) because of the space reshystrictions Nevertheless we are content that most of these constraints were dealt with in the most satisshyfactory way possible The total thickness of levels 31ndash36 reached almost 3 m A series of three hearths (nos 5 6 and 7) was revealed (Fig 133 134) Two of the fireplaces were found in level 31 aligned along the north baulk of the trench (Fig 134) the third belonged to level 32 Two of them were found along the north profile and the third (no 7) was atshytached to the elliptical wall (Fig 133)

These levels were not uniform in color or conshytent (Fig 14) Grayish soil with burned material and scattered stones pieces of charcoal pottery and the refuse of habitation floors alternated with successive thin bands of yellowish soil probably

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

the excavation 19

Figure 131 View of excavation level 30 showing hearth 4 and the elliptical structure

Figure 132 View of the elliptical stone wall from levels 24ndash27

dissolved kouskouras from decayed walls Some examples of sunshydried bricks with imprints of thatch were found unfortunately the limited area exposed did not allow us to define any of the strucshytures with which they might have been associated It is fair to suggest however that although no speshycific architectural remains were recorded the deshyposit overall probably represents the refuse of a house or structure and its contents

Certain levels were extremely rich in ceramics most of them diagnostic of EN I Levels 33 and 34 for example yielded sherds with typical EN I offshyset rims and rippled decoration

LeveL 37 (eN I)

Level 37 is considered the earliest EN I occupashytion phase of Knossos and despite its limited exshytent it revealed interesting new features two pits covered by reddish soil and many stones perhaps filling debris (Fig 135) Pit 1 was found in the northwest corner of the trench and it was comshypletely devoid of pottery Pit 2 was set close to the elliptical wall of the eastern part of the trench its deposit produced some pottery similar to that of level 36 together with a few animal bones The deshyposit of level 37 was scattered with pieces of mudshybricks bearing imprints of straw

LeveLs 38ndash39 (aceraMIc)

Levels 38 and 39 represent the basal layers of the Neolithic occupation (Fig 14) The deposit was poor in finds except for some pieces of obsidishyan and dissolved unbaked mudbrick It was similar in content to level 37 except that it contained no pottery Levels 38 and 39 have thus been assigned to the Aceramic phase

Figure 133 View of excavation level 31

Notes on the Architectural RemainsThe criteria for identifying domestic strucshy

tures as ldquohousesrdquo are the presence of wellshydefined ground plans the remnants of building materials

and habitation debris associated with inside spacshyes The presence of hearths and pits or rich deposshyits of pottery and other finds related to domestic

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

nikos efstratiou alexandra karetsou and eleni banou20

Figure 134 Plans of excavation levels 31 32 and 34 showing hearths 5 6 and 7

LEVEL 31

WALL

Hearth Hearth

43

65

-486

-454-474

-473 -476

LEVEL 32

LEVEL 34

Hearth WALL73 4

-489

-454

-509

050 1 2 m

050 1 2 m

050 1 2 m

-805 -804

-814-818 -808

Bacute

B

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

the excavation 21

Figure 135 Plan of excavation level 37 showing pits 1 and 2

structures may be interpreted with less certainshyty since these features and finds could easily have been associated with outdoor spaces and activities as well Openshyair spaces such as courtyards street lanes or corridors were most likely lively and mulshytifunctional areas for a Neolithic community

The identification of interior and exterior spaces at Knossos was hampered by the limited scale of our excavation and the lack of contextual evidence Nevertheless in the long stratigraphic sequence of the Neolithic deposits investigated we were occashysionally able to isolate structures and features releshyvant to the criteria noted above Unfortunately any definitive interpretation of our finds is problematshyic although the architectural features documentshyed by JD Evans in his campaigns were useful for comparison (1964 132 1994 1)

Reconstruction of the spatial development of the Neolithic settlement on the Kephala mound has been of paramount interest to archaeologists (Evans 1971 95 Broodbank 1992 39 Katsianis 2002) What we know so far regarding the estabshylishment and growth of Neolithic Knossos comes primarily from the area beneath the Central Court of the palace however (Evans 1971 fig 1) The

trenches opened along the periphery of the Kephala mound hardly offer concrete evidence for a sound spatial reconstruction and can only inspire some very general remarks Attempts to document the pace of growth using the evidence from specifshyic sections of the settlement and then projecting it to the tell as a whole are undermined by the absence of solid architectural evidence (Whitelaw 1992 225)

Here we present some new spatial features that might be considered characteristic of the orgashynizational structure of Neolithic Knossos along its eastern periphery during the sitersquos long and different cultural periods (Tomkins 2008 21) Although their fragmentary nature and contextushyal deficiency must be acknowledged they lead us to the important conclusion that the eastern edge of the Kephala mound was an integral part of the Neolithic tell and its occupational sequence from the very beginning of the settlementrsquos life (see disshycussion of the Aceramic in Efstratiou this vol Ch 2) A brief summary of the architectural findshyings from Trench II is offered below

All of the architectural remains encountered in our excavation have been dated to the EN II period

LEVEL 37050 1 2 m

PIT

PIT

1

2

-741-753

-733

-454

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

nikos efstratiou alexandra karetsou and eleni banou22

(levels 14ndash29) In levels 16ndash19 we encountered two parallel walls (nos 1 and 2) that most probashybly formed part of a house (Fig 118) The area dishyrectly below the two walls revealed a row of stones that formed a stretch of wall (no 5) running eastndashwest Beneath wall 2 an earlier wall (no 6) with a similar orientation was also found (Figs 120 121 124) The east end of wall 5 seemed to apshypear again in level 22 where it was visible against the west side of wall 6 (Fig 124) Whether walls 5 and 6 which were set at an angle defined the corshyner of an earlier house within the EN II horizon is difficult to say Levels 23ndash26 revealed the remains of a narrow wall (no 7) running in an eastndashwest direction its function is unclear (Figs 123 124)

The massive elliptical wall (no 3) first appeared in level 18 It had a northndashsouth orientation with a curve inclining toward the east at its southern end (Figs 116 118) Measuring 030 m in width and preserved to an impressive height of nearly 070 m it constituted the dominant architectural feature of the entire deposit spanning levels 18ndash27 (Fig 132) It was probably constructed in level 27 (Fig 125) a yellowishshycolored mudshy or clayshytextured deposit The elliptical wall was made of mediumshysized stones of kouskouras bound together with mortar A thick band of yellow soil without finds found next to the wall (see description of levels 17 18 and 18A) may represent a superstructure made of piseacute (clay and straw)

Due to the small extent of the excavated area it is impossible to estimate the total length or the course of the elliptical wall in relation to other spashytial features of the settlement Its use as a defensive or retaining feature would have been problematic since the bend of the curve faced the Central Court (west) and did not follow the natural contour of the Kephala hill so as to protect or retain it (Fig 125 cf Evans 1971 pl VI) Moreover another wall (no 4) of the same width (030 m) was found set at a right angle to the elliptical structure and running eastward (Figs 125 133) It most probshyably served as a supporting element or a partition wall The presence of a perpendicular adjoined wall indicates that we may be dealing with an imshyportant largeshyscale and composite building This interpretation is suggested by the structurersquos careshyful construction and its long life span which exshytended through most of the EN II period Elliptical

walls are rare in the Neolithic levels of Knossos although a very flimsy elliptical wall made of clay also of EN II date was excavated in Stratum IV by JD Evans (1964 149 fig 14) Further attempts at interpretation are hampered by the restricted size of the excavation

No structural features were found in either the LN levels (1ndash3) or the MN levels (4ndash13) of the exshycavation The MN deposits however contained remains of possible hearths postholes and habshyitation refuse Similarly no traces of architecshyture were found in levels 30ndash37 assigned to the EN I period although four hearths (nos 4ndash7) and two pits (nos 1 and 2) were encountered in varishyous levels

Given the limited area exposed it is difficult to discern the spatial association of features such as the hearths and pits with settlement architecture It is not clear whether the EN II hearths 1ndash3 of levels 29 and 29A (Figs 128 129) hearth 4 from levels 30 and 30A (Figs 128 130) hearths 5 and 6 from level 31 and hearth 7 from level 32 (Fig 134) were situated within the interior of a house or in an open area such as a courtyard or a lane between houses of EN II date The same applies to the two EN I pits found in level 37 (Fig 135)

The soshycalled corridor area located between walls 1 and 2 (Fig 118) was kept clear from strucshytures for over two millennia or most of the long life of the Neolithic settlement Whether this area belonged to an openshyair space or the interior of a structure the fact that it remained open could not have been fortuitous It probably means that a general plan of the settlement was agreed upon and kept unchanged at least for this part of the site Other long spaces or corridorshylike arrangeshyments are known from the EN II house in Stratum V (Area C) of the Central Court a few meters to the west of our trench (Evans 1964 146 156ndash157 fig 12) The most striking example of a corshyridor flanked by narrow walls however is the MN house at Katsambas (Alexiou 1955 fig 2 Zois 1973 fig 19) Alexiou (1955) regarded this long space as a courtyard with a surrounding wall (peribolos) an interpretation that probably does not apply to our corridor Nevertheless it appears that similar spaces possibly of diverse function occurred elsewhere in Crete during the Neolithic

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

the excavation 23

ConclusionsDespite the small size of our trench we can

match the new data recovered with a number of suggestions and hypotheses previously advanced by JD Evans in his discussion of the spatial feashytures of Neolithic Knossos (1971 95) We can now confirm his observation regarding the common northndashsouth orientation of the walls in all parts of Neolithic Knossos (Evans 1964 138) Indeed most of the walls uncovered in our sounding exshyemplify this orientation This probably indicates a long duration of spatial organization patterns and social relations within the community Moreover the fact that the only solid architectural remains from the trench date to the EN II period corrobshyorates Evansrsquos suggestion that the Neolithic setshytlement acquired its largest size at that time Our sounding strongly indicates that the EN II settleshyment reached the northeastern fringe of the natushyral hill of Kephala

The interpretation of the massive elliptical strucshyture which is unlike anything found by Evans reshymains problematic It is possible that it served specific collective organizational purposes pershyhaps defensive although this would imply a more complex political scene in Neolithic Crete than is supported by the available archaeological evshyidence Knossos appears to have been a solitary

settlement on the island throughout the seventh and sixth millennia bc although this picture may soon change

Finally the habitation remains from the soshycalled Aceramic levels are undeniably important not so much because they substantiate the exisshytence of a potteryshyfree phase in the Aegean (a longshydebated issue) but rather because they docushyment archaeologically the arrival of the first farmshyers in Crete (see Efstratiou this vol Chs 2 and 11) This lower part of the long Knossos sequence consists of occupation refuse such as burned mashyterial from hearths animal bones and pieces of sunshydried mudbricks The latter may represent a different building technique from that associated with the piseacute in use subsequently as Evans also noshyticed He concluded that the first settlement of the EN I periodmdashldquoAceramicrdquo or notmdashmight have had a temporary character (Evans 1964 136) In later publications although adhering to his first suggesshytion of a smallshysized site and community (Evans 1994 2) he seemed reluctant to comment on the relative permanence of the first settlement acshyknowledging the existence of more than one early phase (1971 103) Only extensive future research can elucidate the scale and characteristics of the earliest settlement of Crete

ReferencesAlexiou S 1955 ldquoἈνασκαφὴ Κατσαμπᾶ Kρήτηςrdquo

Prakt 110 [1960] pp 311ndash320

Broodbank C 1992 ldquoThe Neolithic Labyrinth Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 39ndash75

Evans JD 1964 ldquoExcavations in the Neolithic Setshytlement of Knossos 1957ndash60 Part Irdquo BSA 59 pp 132ndash240

mdashmdashmdash 1971 ldquoNeolithic Knossos The Growth of a Set tlementrdquo PPS 37 pp 95ndash117

mdashmdashmdash 1994 ldquoThe Early Millennia Continuity and Change in a Farming Settlementrdquo in Knossos A

Labyrinth of History Papers in Honour of S Hood D Evely H HughesshyBrock and N Momigliano eds Oxford pp 1ndash20

Furness A 1953 ldquoThe Neolithic Pottery of Knossosrdquo BSA 48 pp 94ndash134

Katsianis M 2002 Detecting the Growth of Neolithic and Early Bronze Age Knossos through the Modelling of the Depositional Evidence A GIS Application MA diss University College London

Mackenzie D 1903 ldquoThe Pottery of Knossosrdquo JHS 23 pp 157ndash205

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1

nikos efstratiou alexandra karetsou and eleni banou24

Tomkins PD 2008 ldquoTime Space and the Reinvention of the Cretan Neolithicrdquo in Escaping the Labyrinth The Cretan Neolithic in Context (Sheffield Studies in Aegean Archaeology 8) V Isaakidou and P Tomkins eds Oxford pp 21ndash48

Whitelaw TM 1992 ldquoLost in the Labyrinth Comments on Broodbankrsquos Social Change at Knossos before the Bronze Agerdquo JMA 5 pp 225ndash238

Zois A 1973 ΚρήτηmdashἘποχή τοῦ Λίθου Athens

  • NK_Frontmatter
  • NK_Ch1