lable at ScienceDirect

Quaternary International 411 (2016) 216e232

Contents lists avai

Quaternary International

journal homepage: www.elsevier .com/locate/quaint

Levallois points and triangular flakes during the Middle Palaeolithic innorthwestern Europe: Considerations on the status of these pieces inthe Neanderthal hunting toolkit in northern France

Emilie Goval a, b, *, David H�erisson b, Jean-Luc Locht a, c, Aude Coudenneau d

a Inrap Nord-Picardie, Amiens, Franceb UMR7194 du CNRS, Paris, Francec UMR8591 du CNRS, Meudon, Franced UMR 7269 du CNRS, Aix-en-Provence, France

a r t i c l e i n f o

Article history:Available online 19 May 2015

Keywords:NeanderthalPointsUse-wear analysisButchery knivesHuntingMiddle Palaeolithic

* Corresponding author. Inrap Nord-Picardie, AmieE-mail address: emilie.goval@inrap.fr (E. Goval).

http://dx.doi.org/10.1016/j.quaint.2015.04.0391040-6182/© 2015 Elsevier Ltd and INQUA. All rights

a b s t r a c t

Over the past twenty years, during the course of programmed and particularly rescue archaeologicaloperations, prehistorians in northern France have identified numerous sites attributed to the MiddlePalaeolithic. Over twenty of these have been excavated and are part of one of the most reliable chro-nostratigraphic frameworks known in Western Europe. Northern France thus acts as a sort of test lab-oratory where the fusion of abundant archaeological and palaeoenvironmental data can now culminatein a general synthesis. The point is omnipresent in small numbers in these lithic assemblages, and is auniversal element of the Neanderthal toolkit. Levallois points are often associated with triangular flakeproduction. In this article we use various lithic assemblages come from 26 excavations represented 35human occupation levels located in a geographic area covering the northeast quarter of France, fromNormandy to the Flemish plains. Levallois points and triangular flakes are compared according todifferent disciplines: technology and use wear analysis. Chronology and environmental reconstructionallowed a better understanding of the Neanderthal subsistence behaviour. This article aims to providenew and original insights into point and triangular flake production, their functions and their mobilityduring the Middle Palaeolithic.

© 2015 Elsevier Ltd and INQUA. All rights reserved.

1. Introduction

Today, in everyday language, a point is defined as the sharp andpointed end of an object. In daily life, the conception and use of thistype of object has now become banal (the knife, for example), butthe situation was undoubtedly very different for Middle Palae-olithic human groups. During that time, Neanderthals producedflakes, blades, points and at times bifaces, following differentconcepts and operative schemas. Points were produced accordingto different methods, resulting in the production of objects withvery different shapes, dimensions and angles.

Earlier research carried out in northern France to MiddlePalaeolithic focused mainly on the original character of thisgeographic area. It is common that in lithic assemblages, produc-tion of Levallois flakes, blades and point is attested. Thus,

ns, France.

reserved.

knowledge about the different techniques of knapping identified innorthern France was largely developed in the last twenty years. Theresearch of A. Tuffreau and S. Revillion at the excavation of Seclinidentified and characterized the Laminar knapping (Tuffreau andR�evillion, 1984; R�evillion, 1994; R�evillion and Tuffreau, 1994;Locht dir., 2002; Locht et al., 2010).

Nevertheless, despite the large number of occupation levelsuncovered for Middle Palaeolithic, studies of point productionremain rare and poorly documented. The point was considered for along time as an anecdotal element of the Neanderthal toolkit inEurope, but recent studies show that this object had special statusamong these populations (Knecht, 1997; Soressi and Locht, 2010;Goval, 2012; Lazuen, 2012). Numerous studies, most of whichwere conducted in the Near East and in Africa, demonstrate the useof points and triangular flakes as projectile points (Shea, 1988,2003, 2006; Plisson and Beyries, 1998; Villa and Lenoir, 2006;Wilkins et al., 2012), but in Europe, points and triangular flakeshave been poorly studied. It is now widely acknowledged thatNeanderthals were capable of hunting and consuming large

E. Goval et al. / Quaternary International 411 (2016) 216e232 217

herbivores. Although we knowwhich species were hunted by thesegroups, the methods and tools used to hunt them are still oftenunidentified. It is difficult to accurately determine which tools werepart of the hunting equipment, but it appears certain that the flintpoint was one of these components, as shown by studies con-firming the hafting of these tools and the presence of impact marks(Bo€eda et al., 1996; Plisson and Beyries, 1998; Bonilauri, 2010).Conversely, other use-wear studies show that the point washandheld, like a knife, and that the sharp end was used as a tip(Beyries in Tuffreau, 1993; Moncel et al., 2009; Coudenneau, 2013).The status of the point during the Middle Palaeolithic remainsenigmatic, as regards both its role and function, as it is produced invariable quantities and in different ways at the various sites.

If points are produced by prehistoric in the Middle Palae-olithic, we can assume that they had a special function in lithic

Fig. 1. Location of all the sites from MIS 7 to 3 in northern France with Levallois p

assemblages. Compared to other products, the first characteristicof this object is to have an apex. It is therefore essential to identifywithin lithic assemblages, objects whose production objective isto obtain a point and those obtained by flake production but havean apex. In this article, the latter are referred to as triangularflakes.

This article focuses on the point and triangular flakes as anobject in the toolkit, but also on their production, role, mobilityand use of these tools, or in other terms, its status. It aims togain a better understanding of the Neanderthal hunting toolkitbut also to enhance our knowledge of the function of Levalloispoints and triangular flakes during the Middle Palaeolithic inEurope. The question of the use of these objects as huntingtools needs to be addressed for the whole of the Europeancontinent.

oints and triangular flakes (one site may include several occupations levels).

Table 1Summary of all the occupation levels with Levallois points and/or triangular flakes during the Middle Palaeolithic in northern France.

Site Level Relative dating Absolute dating Numberof artefact

Number ofLevalloispoint

Number oftriangularflake

Type of triangular flake Presenceof fauna

Use-wearanalysis

Main references

Saint-Illiers-la-Ville S2N2 Isotopic stage 7 232 ± 23 Ka(TL heated flint)

364 2 4 pseudo-Levallois point 3),convergent scraper (1)

Blaser, 2013

Biache-Saint-Vaast IIA Isotopic stage 7 230 ± 24 Ka (ESR/os) 47,155 2 124 convergent scraper X X Tuffreau and Somm�e, 1988;H�erisson, 2012

Therdonne N3 End of isotopicstage 7/6

178 ± 11 Ka(TL heated flints)

46,163 138 27 pseudo-Levallois point (18),pseudo-Levalloispoint retouched (7),convergentscraper (2)

X X Locht et al., 2010;H�erisson, 2012

Bapaume-les-Osiers B Beginning isotopicstage 6

1724 2 20 No information Tuffreau, 1987; Koehler, 2008

Havrincourt N0 Beginning isotopicstage 6

3 1 0 X Goval et al., 2013;Antoine et al., 2014

Le Pucheuil B Beginning isotopicstage 6

4111 10 0 Delagnes, 1993; Delagnesand Ropars, 1996

Champvoisy Beginning isotopicstage 6

2058 4 69 pseudo-Levallois point (4),convergent scraper (51),mousterianpoint (9), scraper“d�ejet�e” (5)

Tuffreau, 1989

Caours NC Isotopic stage 5e 116 ± 6 Ka(TL heated flints)

1374 0 25 pseudo-Levallois points X Antoine et al., 2006;Locht et al., 2009

Caours N1 Isotopic stage 5e 221 (in progress) 0 4 pseudo-Levallois points X Antoine et al., 2006;Locht et al., 2009

Caours N2 Isotopic stage 5e 132 (in progress) 0 2 pseudo-Levallois points X Antoine et al., 2006;Locht et al., 2009

Caours N3 Isotopic stage 5e 319 (in progress) 0 5 pseudo-Levallois points X Antoine et al., 2006;Locht et al., 2009

Caours N4 Isotopic stage 5e 122.8 ± 8.6 Ka(TL heated flints)

1100 (in progress) 0 9 pseudo-Levallois points X Antoine et al., 2006;Locht et al., 2009

Bettencourt-Saint-Ouen N3b Isotopic stage 5d 1298 1 11 pseudo-Levallois point Locht dir., 2002Villiers-Adam sector 1 Isotopic stage 5c 110 ± 11 Ka

(TL heated flints)2190 56 29 pseudo-Levallois point (21),

convergent scraper (4),mousterian point (4)

Locht et al., 2003

Fresnoy-au-Val N2 Isotopic stage 5c 106.8 ± 7.5 Ka(TL heated flint)

1270 1 2 pseudo-Levallois point X Goval and Locht, 2009

Seclin D7 Isotopic stage 5c 999 0 3 convergent scraper R�evillion, 1994Saint-Illiers-la-Ville S2N1 Isotopic stage 5c 105 ± 7.8 Ka

(TL heated flint)803 1 5 pseudo-Levallois point Blaser, 2013

Gouy-Saint-Andr�e Isotopic stage 5a 491 18 1 convergent scraper Depaepe and Deschodt, 2001Bettencourt-Saint-Ouen N2b Isotopic stage 5a 6466 128 12 pseudo-Levallois point (10),

convergent scraper (2)X X Locht dir., 2002

Riencourt-l�es-Bapaume

CA Isotopic stage 5a 5000 3 27 pseudo-Levallois point (13),convergent scraper (14)

X Ameloot-van-der-Heijden, 1993;Tuffreau, 1993

Blangy-Tronville Isotopic stage 5a 337 10 0 Depaepe et al., 1999Fresnoy-au-Val N1 Isotopic stage 5a 4284 13 0 X Goval and Locht, 2009Revelles “les Terres Sellier” Isotopic stage 5a 368 5 0 Guerlin, 2002Villers-Bretonneux SHS Isotopic stage 5a 102 1 2 mousterian point Depaepe et al., 1997Mauquenchy WAII Isotopic stage 5a 83 ± 7.6 Ka

(TL heated flint)200 4 13 pseudo-Levallois point Locht et al., 2013b

E.Goval

etal./

Quaternary

International411

(2016)216

e232

218

Chav

ignon

SGF

Isotop

icstag

e5a

1164

46

conve

rgen

tscraper

(4),

mou

sterianpoint(1),

scraper

“d� ejet� e”

(1)

Sellier,sou

spresse

Auteuil

Isotop

icstag

e5a

1550

4719

pseudo-Le

vallo

ispoint

Swinnen

etal.,19

96;

Koe

hler,20

12Cou

rmelles

Isotop

icstag

e5a

572

0Se

llier

andCou

tard,2

007

Villen

euve

l'Archev

� eque

BIsotop

icstag

e5a

212

111

pseudo-Le

vallo

ispoint

Dep

aepe,

2007

Lailly-Bea

urega

rdB

Isotop

icstag

e5a

1018

101

pseudo-Le

vallo

ispoint

Dep

aepe,

2007

Molinon

sA

Isotop

icstag

e5a

710

71

pseudo-Le

vallo

ispoint

Dep

aepe,

2007

Bettenco

urt-Saint-Ouen

N1

Endof

isotop

icstag

e5a

438

40

XLo

chtdir.,20

02

Rienco

urt-l� es-

Bap

aume

C[C12

]En

dof

isotop

icstag

e5a

8813

6727

pseudo-Le

vallo

ispoint(13),

conve

rgen

tscraper

(12),

scraper

“d� ejet� e”(2)

Tuffreau

,199

3

Bea

uva

is1

Isotop

icstag

e4

55.6

±4Ka

(TLhea

tedflint)

1947

034

8pseudo-Le

vallo

ispoint

XLo

cht,20

04

Attilly

N2

Isotop

icstag

e3

740

8co

nve

rgen

tscraper

Loch

tan

dGuerlin

,199

7

E. Goval et al. / Quaternary International 411 (2016) 216e232 219

2. Levallois points and triangular flakes: definitions andmethods of production

The Levallois point is easily identifiable in a lithic assemblage,but the same is not necessarily true of the wide diversity of trian-gular flakes often present in Middle Palaeolithic sites. In otherterms, must we only consider the pointed end of an object or thepointed object as a whole when assessing the notion of points inlithic industries? Prehistoric archaeology generally targets thelatter option and considers the pointed object as a whole (Plissonand Beyries, 1998; Coudenneau, 2013). A point must thus firstand foremost be defined as the association of two convergent edgesand a pointed end, but the technological and morphofunctionalcharacteristics of the piece must also be taken into account. Themorphological and technological diversity of Levallois points andtriangular flakes during the Middle Palaeolithic in Europe, andparticularly in northern France, is considerable.

Point is defined as two converging edges and pointed end. Apoint can be obtained by a Levallois's production, in this case is aLevallois point. A point can also be obtained by a flake's production.In this case, it is the triangular morphology of the object that givesit the status of point. The triangular flake can be retouched, forminga Mousterian point or convergent scraper for example.

Bordes' typological list included numerous types of points: theLevallois point, elongated Levallois point, pseudo-Levallois point,mousterian point, convergent side-scraper, offset convergent side-scraper (Bordes, 1961). Therefore, the typological spectrum ofpoints and triangular flakes is varied. One of the questions is: arethe Levallois points instead of triangular flakes are produced forspecific uses?

Note that in this article, we only use the term point to refer tounifacial points, as opposed to the numerous bifacial points inSouth African assemblages, for example (Villa and Lenoir, 2006).Apart from the case of the Keilmessergruppen and certain leaf-shaped points (Rubens, 2007; Rubens and Di Modica, 2011), bifa-cial points are absent from Western European assemblages duringthe Middle Palaeolithic.

In addition, and unlike in African and Near Eastern assemblages,points and triangular flakes are only present in small proportions inMiddle Palaeolithic European assemblages. The presence of Leval-lois points and triangular flakes in lithic industry production isoften ubiquitous, but never exclusive. Varied methods are used toproduce these pieces, and consequently, they are more or lessstandardized. The Levallois point is considered to be a sought-after,standardized object, produced using a controlled technique (Bo€eda,1994). It is obtained by convergent unipolar or bipolar debitage. Thepseudo-Levallois point in lithic assemblages in northern France isgenerally obtained using discoidal debitage where the sought-afterproduct is preferentially a short and thick flake (Locht, 2004). Atsome rare sites, the pseudo-Levallois point is produced byconvergent unipolar debitage, and the percussion axis is offset inrelation to the elongation axis of the piece (Locht et al., 2003). Inmany other cases, point production is based on non-Levalloisconvergent unipolar debitage. The knapper uses the original nat-ural convexities of the block and prepares the striking platforms ina rudimentary way. As at Villiers-Adam, where point and triangularflake production is important, different debitage modes wereapplied (unipolar, simple convergent with one surface; unipolar,convergent with two productive surfaces knapped successively).Recourse to one or the other of these modes depends on the orig-inal shape of the block and the degree of exploitation variesdepending on raw material quality (Locht et al., 2003). The trian-gular flakes in the lithic assemblages in northern France are mainlyon Levallois flakes. In level IIA from Biache-Saint-Vaast, 124convergent side-scrapers were produced on elongated flakes, then

Fig. 2. Distribution of the sites with Levallois points and/or triangular flakes by MIS (same size of the symbols means same proportion; a bigger symbol means more numerousartefacts of this type).

E. Goval et al. / Quaternary International 411 (2016) 216e232220

retouched to obtain the convergence of the edges at the distal endof the flake (H�erisson, 2012), an almost perfect blank/tool adequacyis sought after (Bo€eda, 1988). “This adequacy is recorded by the lowlevel of investment required to obtain the final tool” (H�erisson,2012: 169).

As well as technological and typological diversity, the compar-ison of Levallois points and triangular flakes in northern France alsoreveals significant morphological and dimensional variety. Pointscan by definition be assimilated to triangular or possibly sub-triangular flakes and are thus made up of the association of a

E. Goval et al. / Quaternary International 411 (2016) 216e232 221

base opposed to a pointed end. Points present diverse morphol-ogies, depending on the raw material and production method usedand the function the object is destined for. Some are large, sym-metrical and balanced; others on the other hand are small, offset,with thickset bases. This raises the question of a relationshipbetween production, shape, and function.

3. Sites and results

Over the past twenty years, prehistorians in northern Francehave detected numerous sites attributed to the Middle Palaeolithicthrough programmed archaeology, and especially through rescueoperations. Over twenty of these have been excavated and inte-grated into one of the most reliable chronostratigraphic frame-works in Western Europe (Antoine et al., 2003, 2007; amongothers). Northern France thus acts as a test laboratory where thecombination of abundant archaeological and palaeoenvironmentaldata can now lead to a general synthesis.

3.1. Site corpus presentations

This study includes 26 sites representing 35 human occupationlevels located in a geographic area covering the northwest quarterof France, from Normandy to the Flemish plains (Fig. 1). These sitesare mainly situated in slope or plateau environments, apart fromseveral exceptions. All of the occupation levels are correlated to theMiddle Palaeolithic, from the beginning ofMIS 7 toMIS 4/beginningof MIS 3. They all include Levallois points and/or triangular flakes invarying proportions. The Levallois points and triangular flakes fromsix lithic series were analysed for microwear traces by differentresearchers [Riencourt-les-Bapaume: S. Beyries; Bettencourt-Saint-Ouen: J.-P. Caspar, V. Rots; Therdonne: A. Coudenneau; Beauvais: A.Coudenneau; Biache-Saint-Vaast: S. Beyries, V. Rots and Fresnoy-au-Val: A. Coudenneau].

In order to conduct a diachronic and synchronic analysis ofhuman occupation levels, excavations are grouped by MIS within

Fig. 3. Therdonne e level 3. Large Levallois points obtained by converg

paleoenvironmental and climatic context. We have also systemat-ically indicated if faunal remains were discovered in associationwith lithic industries containing “points” where use wear werehighlighted. If fauna is preserved in excavations in northern France,it is first and foremost due to exceptional preservation conditionson the site. The aim of specifying this indication is not to drawparallels between the quantity of side-scrapers and the quantity offauna. An absence of fauna on the site is usually due to taphonomicfactors and not to palethnographic ones. It is important to note if“points” were used for butchery activities. Key data are referencedin Table 1.

3.2. Sites with “points” from MIS 7 and 6

3.2.1. Palaeoenvironmental background, site corpus and typo-technological data

Our analysis begins at MIS 7, which acts as an inevitable chro-nological barrier as no sites pre-dating this period have yieldedLevallois points in northwest France. In northern France, MIS 7 to 6are characterized by cyclic climatic and environmental variationswhere human settlements and mammalian faunas are only reallypresent when temperate conditions return duringMIS 7 (7a and 7c)or potentially in the extreme beginning of MIS 6 (Auguste, 2009;H�erisson, 2012).

During the course of these temperate phases, a much coolerclimate prevailed than at present, dominated by oceanic influencesand marked by grasslands propitious to human occupations(Auguste, 2009; H�erisson, 2012). Six sites contain Levallois pointsproduced by convergent unipolar reduction: Bapaume-les-Osiers(series B, Tuffreau, 1976, 1987; Koehler, 2008), Havrincourt (seriesN0, Goval et al., 2013; Antoine et al., 2014), Le Pucheuil (series B,Delagnes, 1993; Delagnes and Ropars, 1996), Therdonne (series N3,Locht et al., 2010; H�erisson, 2012), Champvoisy (Tuffreau,1989) andSaint-Illiers-la-Ville (series S2N2, Blaser, 2013) [Table 1 and Fig. 2].The site of Therdonne is exceptional in that it contains 138 largeLevallois points, although they only represent 0.3% of the total

ent unipolar reduction (Drawings: S. Lancelot in H�erisson, 2012).

E. Goval et al. / Quaternary International 411 (2016) 216e232222

series (all types of pieces included) or 2.7% (excluding shards)[Fig. 3]. The analysis of this site is particularly interesting as theabsence of point cores raises the question of in situ production. Twotypologically Levallois points are present in level IIA of Biache-Saint-Vaast but their production is clearly part of a recurringLevallois flake schema. In addition, triangular flakes were discov-ered in the sites of Therdonne and Saint-Illiers-la-Ville, often inhigher proportions than Levallois points (Table 1). In level IIA at thesite of Biache-Saint-Vaast, 124 convergent side-scrapers were

Fig. 4. Biache-Saint-Vaast e level IIA. Double c

recorded (H�erisson, 2012). Elongated Levallois flakes with variedshapes and dimensions were used as blanks for the majority of theconvergent products (Fig. 4). Three tool sub-types were identified.The first is characterized by a distal zone formed by a semi-abruptconvex edge with an abrupt prehensile zone, opposed to a sharpstraight edge for transformative contact (Fig. 4 n�1). The secondsub-type is based on the axial symmetry of the tool (Fig. 4 n� 2 and4). The left and right edges converge into a pointed distal part, andboth sides have sharp cutting edges. The third sub-type is

onvergent side-scrapers (H�erisson, 2012).

E. Goval et al. / Quaternary International 411 (2016) 216e232 223

characterized by pieces with global axial symmetry (Fig. 4 n�3),retouch in the distal area sometimes creates an abrupt borer, visibleon the profile of the pieces (H�erisson, 2012). The presence of manyelongated convergent side-scrapers in level IIA of Biache-Saint-Vaast must be related to the presence of 189,760 faunal remains,including 19,420 identified bones from 18 different species, wherethe aurochs, the bear and the rhinoceros are predominant (Tuffreauand Somm�e, 1988; Auguste, 1995; H�erisson, 2012).

3.2.2. Microwear analysisMicrowear analysis was conducted on two sites: Therdonne by

A. Coudenneau and Biache-Saint-Vaast by V. Rots. The microwearanalysis of the Levallois points from Therdonne included 119 non-retouched Levallois points. Seventeen of these points displaymacroscopic use-wear (Coudenneau, 2013). All of these arerelated to butchery activities but the functional analysis showedthat their use was not strictly identical throughout the operativesequence. Four phases in the butchery operative sequence wereidentified, revealing four different tool groups with differentactive zones. In this way, some of these points may have beenused to perforate skin or remove tendons (schemas 1 and 2),others to cut meat (schema 3) and others to remove limbs(schema 4). When these points were used as butchery knives, aconvex/concave-convex configuration of the cutting edges ispredominant. Two points may also have been used as projectiles(or at least as hunting arms), and thus reveal another potentialfunctional aspect of the Levallois points from Therdonne(Coudenneau, 2013) [Fig. 5].

Microwear analysis was also carried out on level IIA of the site ofBiache-Saint-Vaast by V. Rots (2013). In the same way as for theanalysis of Therdonne, the function of convergent side-scrapers is

Fig. 5. Therdonne e level 3. Schema showing the distri

in most cases related to butchery knives, and in a minority of cases,to a hunting activity (Rots, 2013) [Fig. 6]. In addition, in this series,the prehensile zone has been extensively modified. These piecesbear a proximal zone with backed distal edges either to hold thepiece directly or to facilitate hafting (H�erisson, 2012; Rots, 2013).Moreover, most of these butchery knives are asymmetrical andcorrespond to the first sub-type of convergent side-scrapersdescribed above. Hast points are logically symmetrical, slenderand can be partly assimilated to the second sub-group of conver-gent side-scrapers described above (H�erisson, 2012; Rots, 2013).According to the microwear study by V. Rots (2013), over fiftytriangular flakes bear use-wear.

3.3. Sites with “points” from MIS 5e

3.3.1. Palaeoenvironmental background, site corpus and typo-technological data

Few sites correlated to MIS 5e have been discovered in northernFrance. The site of Caours (currently under excavation, direction J.-L. Locht) is one (Antoine et al., 2006; Locht et al., 2009). It is the onlycurrently-known evidence of the Eemian interglacial in a fluvialcontext in the Somme Basin, and thus cannot be considered asrepresentative of human occupations as a whole at that time. U/Thdates indicate an average age of ±120 ka (Antoine et al., 2006). Theclimate was considerably warmer than the present-day climate,temperate, with oceanic influence and deciduous forest cover. Theforest-covered landscapewas closed, with abundant animal species(rhinoceros, hippopotamus, etc.), but few individuals (Locht et al.,2014).

The typological spectrum of points and triangular flakes isrestricted, and only 45 pseudo-Levallois points were recorded, all

bution of use-wear on points (Coudenneau, 2013).

Fig. 6. Biache-Saint-Vaast e level IIA. Top: Microwear on an elongated mousterian point (E8-513), interpreted as used as a spear point: 1. Pseudo-burin spall on the distal left point(16x); 2. Striation from counter pressure, in the prolongation of the large scar on the ventral proximal surface (100x); 3. Small pseudo-burin spall detached from the ventral distalleft point (50x); 4. Impact striation associated with edge damage on the ventral medial left edge (100x); 5. Scarring evidence on the hafting boundary on the ventral medial left edge(16x); 6. Obliquely initiated, step terminating hafting scarring on the dorsal proximal right edge (8x) [Rots, 2013]. Bottom: Microwear on a convex side-scraper (B76 15Y IIa),interpreted as use as a hafted butchering knife: 1. Small tip fracture with associated use damage on the dorsal distal right edge (16x); 2. Use-wear polish on the ventral distal rightedge (200x); 3. Use-wear polish on the ventral distal right point (200x); 4. Bright spot on the ventral medial left edge associated with dorsal hafting scarring at the haft boundary(200x); 5. Bright spot in striation associated with hafting scarring and hafting polish on the ventral proximal left edge (100x); 6. Start of hafting damage around the haft boundaryon the dorsal medial right edge (10x); 7. Hafting polish associated with rounding on the dorsal medial ridge (200x) [Rots, 2013].

E. Goval et al. / Quaternary International 411 (2016) 216e232224

levels included. No Levallois points were discovered (Table 1 andFig. 2). Conversely, the faunal spectrum is better represented, with892 large mammal remains brought to light between 2003 and2006 (Antoine et al., 2006).

3.3.2. Microwear analysisNo use-wear analysis of the points has, as of yet, been conducted

and thus no link has been established between point productionand potential function.

E. Goval et al. / Quaternary International 411 (2016) 216e232 225

3.4. Sites with “points” from MIS 5d and 5c

3.4.1. Palaeoenvironmental background, site corpus and typo-technological data

TheWeichselian early glacial (MIS 5d to 5a) in northern France ischaracterized by a slow climatic degradation and the progressivecontinentalization of the climate in successive stages (Antoineet al., 2003). Sites are mainly preserved at bottom of the slopeswhere the pedosedimentary record is the most developed (Lochtet al., 2010). At this time, the animal spectrum was less variedthan during the previous MIS 5e and was made up of large herds ofherbivores.

Ten sites attributed to this chronological period have beendiscovered in northern France, including five sites with Levalloispoints and/or triangular flakes: Bettencourt-Saint-Ouen (level N3,Locht, dir., 2002), Villiers-Adam (sector 1, Locht et al., 2003),Fresnoy-au-Val (series 2, Locht, dir., 2008; Goval and Locht, 2009),Saint-Illiers-la-Ville (S2N1, Blaser, 2013), and Seclin (series D7,R�evillion, 1994) [Table 1 and Fig. 2]. Levallois points are generallylimited to one piece per series. They are systematically producedusing a convergent unipolar method. The site of Villiers-Adam is acase apart. The main knapping aim here appears to have been theproduction of typologically Levallois points, of which there are 56(Locht et al., 2003: 89). They were mainly produced by convergentunipolar reduction (Fig. 7). Moreover, 29 retouched pieces areassociated with these points, including 8 triangular flakes(convergent side-scrapers and mousterian points), and 21 pseudo-Levallois points complete the assemblage (Locht et al., 2003). Thestudy highlights the careful retouch of the point, where bothretouched edges converge, regardless of the type of triangular flake.

Fig. 7. Villiers-Adam e sector 1. Levallois points from a convergent unipo

The ventral face of these triangular flakes is sometimes thinned bylarge invasive removals or by several removals on the ventral anddorsal flake surfaces (Locht et al., 2003: 93).

The triangular flakes and pseudo-Levallois points in these as-semblages are better represented but remain marginal (never ac-counting for more than 1% of the whole assemblage). No particularblank morphology appears to have been selected before retouchingthe tool to obtain a tip.

3.4.2. Microwear analysesMicrowear analysis was conducted on series 2 from the site of

Fresnoy-au-Val by A. Coudenneau. This site contained a singleLevallois point and no functional interpretation could be proposed.However, the analysis brought to light the macroscopic splinteringof a morphologically pointed piece and a polish due to contact withrelatively fresh skin, indicating a defleshing action (Goval, 2012;Coudenneau, 2013) [Fig. 8]. The microwear of the site ofBettencourt-Saint-Ouen was studied by J.-P. Caspar, but the pointsfrom level N3, correlated to MIS 5d, could not be included due tothe presence of a thick white patina.

3.5. Sites with “points” from MIS 5a

3.5.1. Palaeoenvironmental background, site corpus and typo-technological data

The climatic and palaeoenvironmental conditions duringMIS 5aare globally similar to those of MIS 5c, with a more marked con-tinentalization of the environment and the climate (Antoine et al.,this volume). MIS 5a is identified by the presence of a grey forestsoil. The end of this same stage is characteristic of the formation of

lar reduction sequence (Drawings: P. Raymond in Locht et al., 2003).

Fig. 8. Fresnoy-au-Val e series 2. Piece interpreted as used for fleshing more or less fresh skin. Details of the macrowear (scale shown: 1 cm) and microwear (scale shown: 1 mm). Onthe edges, the macrowear (lines) and microwear (shaded area) are shown [Analysis: Coudenneau in Goval, 2008].

E. Goval et al. / Quaternary International 411 (2016) 216e232226

steppic soils, leading to the conservation of many occupations innorthern France. No less than 16 occupation levels with Levalloispoints and/or triangular flakes have been recorded. These areoccupation levels of: Gouy-Saint-Andr�e (Depaepe and Deschodt,2001), Bettencourt-Saint-Ouen (levels N2b and N1, Locht dir.,2002), Riencourt-les-Bapaume (levels CA and C12), Tuffreau,1993; Goval and H�erisson, 2006), Blangy-Tronville (Depaepeet al., 1999), Fresnoy-au-Val (series 1, Goval and Locht, 2009),Revelles « les terres Sellier » (Locht et al., 2013a), Villers-Bretonneux (series SHS, Depaepe et al., 1997), Mauquenchy (se-ries WAII, Locht et al., 2013b), Chavignon (Sellier, 2015), Auteuil(Swinnen et al., 1996; Koehler, 2012), and Courmelles (Sellier andCoutard, 2007) [Table 1 and Fig. 2]. Three sites located in theVanne Valley have been added to these sites, for the purposes of

this study, although they are situated a little more to the south:Villeneuve l'Archev�eque (series B, Depaepe, 2007), Lailly-Beauregard (series B, Depaepe, 2007) and Molinons (series A,Depaepe, 2007) [Fig. 1]. Levallois points were discovered in all ofthese sites in variable quantities but their proportions only rarelyattain more than 3% of the whole lithic series. At the sites ofFresnoy-au-Val, Revelles, Blangy-Tronville, the series N1 fromBettencourt-Saint-Ouen and Courmelles, no triangular flakes areassociated with these Levallois points. Varied triangular flakes(pseudo-Levallois points, convergent side-scrapers, mousterianpoints, etc.) are associated with the Levallois points in all the otheroccupation levels, but are often limited in number (Fig. 9).

Level CA of the site of Riencourt-les-Bapaume stands out due tothe presence of more abundant triangular flakes, including

Fig. 9. Levallois points and triangular flakes from different sites correlated with MIS 5a. 1: Molinons (drawing: J.-L. Locht), 2 to 8: Riencourt-l�es-Bapaume, series C, C12 (drawings: G.Leroy), 9 and 10: Auteuil (drawings: J.-L., Locht), 11. Revelles (drawing: N. Sellier-Segard).

E. Goval et al. / Quaternary International 411 (2016) 216e232 227

retouched Levallois points, Mousterian points, double convergentside-scrapers and offset convergent side-scrapers (Ameloot-Vander Heijden, 1993; Tuffreau, 1993) [Fig. 10]. Series C12 from thesite of Riencourt-l�es-Bapaume also contains many triangular flakes.These flakes systematically display a more elongated morphology.The tip angle obtained by retouch entails a more acute convergenceof both flake edges than for the Levallois points from the sameseries (Goval, 2008, 2012).

This systematic association of these different types of points inthe sites raises questions as to the complementarity or functional

supplementarity of the assemblages. Potential faunal analysesprovide little or no further information as only N2b from the site ofBettencourt-Saint-Ouen contained sparse faunal remains.

3.5.2. Microwear analysesMicrowear analyses were conducted on three sites: Fresnoy-au-

Val by A. Coudenneau, Bettencourt-Saint-Ouen by J.-P. Caspar andRiencourt-l�es-Bapaume (CA) by S. Beyries. Among the thirteenLevallois points in series 1 from the site of Fresnoy-au-Val, only twopieces could be studied. The first appears to have been used for

Fig. 10. Riencourt-l�es-Bapaume e level CA. Triangular flakes and Levallois point (drawings: G. Leroy in Tuffreau, 1993).

E. Goval et al. / Quaternary International 411 (2016) 216e232228

butchery activities, the second bears marks equivalent to thoseobtained experimentally with hast weapons, that is to say with aperforating weapon. Nonetheless, this affirmation must be treatedwith caution as the dimensions of this tool seem too large to beused in this way (Coudenneau, 2013) [Fig. 11].

A microwear study was applied to series N2b2, N2b3 and N1 ofthe site of Bettencourt-Saint-Ouen. Seven Levallois points wereunearthed in series N2b2, four of which were studied but none ofthem revealed use-wear traces (Locht et al., 2003). In series N2b3,out of the 117 Levallois points, eight pieces display diverse usemarks. Two were used for wood-working activities, five are asso-ciated with butchery involving bone contact and one was used tocut tender animal tissue (Locht et al., 2003). In the same way as theLevallois point from the site of Fresnoy-au-Val, one point presentsdistinctive marks. It bears short retouch on the distal end of theright edge. The examination of the lower surface shows that the leftedge and apical zone are chipped and denticulated. This damagemay denote violent penetration into an animal carcass. It is similarto certain experimental Levallois points used as hast weapons(Plisson and Beyries, 1998) (Locht et al., 2003).

In the early 1990s, S. Beyries made some microwear observa-tions on the CA series from Riencourt-l�es-Bapaume (Beyries inTuffreau, 1993). The Levallois points and triangular flakes did notundergo systematic use-wear research, but the study showed thatthe flakes with convergent edges were mainly used for wood-working (Beyries in Tuffreau, 1993). The apical part of a Mouste-rian point was also used to perforate relatively soft wood (Beyries inTuffreau, 1993). Recently these results were critically reviewed andrequire clarification (Claud et al., 2013).

3.6. Sites with “points” from MIS 4 and beginning of 3

3.6.1. Palaeoenvironmental background, site corpus and typo-technological data

During the Weichselian Lower Pleniglacial (MIS 4) carried outup until now show that northern France was abandoned by

human groups during this time. The recolonization of thesenorthern regions began at the end of MIS 4 (Locht, 2004; Lochtet al., 2014) and continued during the Middle Pleniglacial (MIS3). The climate, the environment and the animal communities aregenerally characteristic of the “mammoth steppe” (Guthrie, 1982).Ten sites in northern France are currently correlated to MIS 4 and3 and two of these contain triangular flakes [Attilly (Locht et al.,2013a) and Beauvais (Locht et al., 1995; Locht, 2004)]. No Leval-lois points have been recorded (Table 1 and Fig. 2). At the Beauvaissite, pseudo-Levallois points are abundant (nb ¼ 348) and areunquestionably one of the main production aims. In addition,during the excavation of Beauvais site, 465 faunal remains werediscovered (Locht, 2004).

3.6.2. Microwear analysesA functional analysis of the macrotraces on the points from

Beauvais site was conducted by A. Coudenneau. Out of the 348points observed from series 1, only 22 pieces conserved usewear(Coudenneau, 2013). Among these 22 points:

- Six present comparablemarks to experimentally obtained tracesresulting from butchery activities, including one with traces of atransversal action on a hard material on the opposite edge,

- Three points were used in a lengthwise action on semi-hardindeterminate matter (Fig. 12),

- Three points were used on different and indeterminate semi-hard materials with dotted contact,

- Six points bear use traces of a transversal action on hard, bone-type matter,

- Three points were used in a transversal contact action on semi-hard materials; for two of them the observed traces are com-parable to those on experimental wood-working tools and forone of them, to those obtained by scraping dry or tanned skin,

- One point was used with transversal contact on semi-hard tohard indeterminate materials.

Fig. 11. Fresnoy-au-Val e series 1. Point with a projectile type break, photo of the whole piece, upper surface, lower surface and profile, macroscopic detail of the break. Scale of themacroscopic photo: 1 cm (Coudenneau, 2013).

E. Goval et al. / Quaternary International 411 (2016) 216e232 229

These results suggest the use of some of these pseudo-Levalloispoints as butchery knives and not as hast arms or projectile points.Nonetheless, 10 others pseudo-Levallois points present an apicalbreak-type removal with a stepped end characteristic of use as ahunting arm. However, the dimensions and morphological char-acteristics (thickness) of these points do not correspond to theexpected attributes of projectile points (Coudenneau, 2013). Themost accurate interpretation appears to be that these points wereused as hast weapons, as spears to kill wounded or trapped animals(Coudenneau, 2013).

4. Discussion

The diachronic approach adopted in this article was chosen inorder to evaluate the impact of environmental changes on theproduction of points and triangular flakes during the MiddlePalaeolithic. Analyses show that Levallois points were producedusing different production methods fromMIS 7 to the beginning ofMIS 3, and that no distinction can be established in relation to thechoice of the method. Similarly, no quantitative link can be made.Levallois points are ubiquitous in Middle Palaeolithic assemblages

Fig. 12. Beauvais e level 1. Pseudo-Levallois points with wear attributed to lengthwise use on an indeterminate semi-hard material. Photos of the whole pieces, upper surface, lowersurface and profile of the used edge, contours showing the use-wear and macroscopic details of the observed wear. Scales: x10 ¼ 5 mm (Coudenneau, 2013).

E. Goval et al. / Quaternary International 411 (2016) 216e232230

in northern France, their proportion rarely exceeds 3% (not forbeginning of MIS 3 and MIS 4). Environmental changes present inMIS 7 to the beginning of MIS 3 did not impact the amount ofLevallois points and triangular flakes production.

In many sites, the presence of the Levallois point is associatedwith the production of triangular flakes, including pseudo-Levalloispoints, convergent side-scrapers and Mousterian points (Fig. 2).Elongated Levallois flakes with varied dimensions were often used

as blanks for the latter. Nevertheless, although Mousterian pointsare sometimes more abundant than Levallois points, their propor-tion in the overall assemblage remains low. Middle PalaeolithicLevallois points and triangular flakes are very varied, in as far astheir dimensions and morphologies are concerned. One of the aimsof this article was to assess the relationship between production,shape and function. In spite of the discovery of numerous MiddlePalaeolithic sites in northern France (35 occupation levels), very

E. Goval et al. / Quaternary International 411 (2016) 216e232 231

few microwear studies have been applied to Levallois points andtriangular flakes. A similar observation was made recently by V.Rots for Belgian sites (2011). Three different cases emerge:

� For different reasons, the whole series did not undergo micro-wear studies,

� Microwear analysis was applied to the series with no typo-technological distinction, making the correlation betweenmicrowear and points impossible,

� Only a sample of the series was analysed, and as points are onlypresent in small quantities, they are only rarely selected for suchstudies.

However, six sites were studied for microwear traces. FromMIS 7 (Biache-Saint-Vaast, Therdonne) to MIS 4 (Beauvais), theLevallois point and triangular flake were used mainly as butcheryknives, and occasionally as hast points. The presence of the latterin the different assemblages needs to be related to the presumedfunctions of the excavated sites. It appears more likely to find abutchery knife in an area used to butcher an animal brought backto the site, than a hast weapon used during collective hunting.Furthermore, the microwear studies undermine the notion thatthe point was, above all, an object chosen for its pointed part.When the point is used as a butchery knife, the cutting edgeopposite a backed edge is generally the active part of the tool. Theuse of the term 00point00 is thus ambiguous as it is related to thepresumed function of the tool. In the case of the site of Ther-donne, the point was used during butchery activities to inciseskins and flesh. Many Levallois points have been identified asbutchery knives, as have certain triangular flakes, without anyapparent link between form and function. Nonetheless, thestudies conducted by Coudenneau (2013) show that thisconnection is important for the hast point, a slender, symmetricaland well balanced piece. The use of the Levallois point or thetriangular flake by Middle Palaeolithic groups for daily activities isthus a response to complementary and varied activities. Thisreasoning can just as easily be transposed to modern societywhere knives are elongated with a pointed end and, whereas thecutting edge is used systematically, the point is only used on anoccasional basis.

Sometimes retouch transforms the terminus of Levallois pointor triangular flake to obtain a Mousterian point, convergeretouched scraper, etc., but these retouches do not show differentuse-wear. Probably the retouch allowed obtaining the tip itself.

Ten human occupation levels contain faunal remains in associ-ationwith lithic industry (Fig. 2). Only three of them have use-wearstudies on Levallois points and triangular flakes in association withfaunal remains: Biache-Saint-Vaast, Therdonne and Beauvais(Fig. 2). The use-wear analyzes and the zooarchaeological obser-vations are concurrent, but unfortunately putting these data inparallel and drawing reliable conclusions is complicated.

Moreover, recent studies of sites in northern France have shownthe mobile character of Middle Palaeolithic points (Goval, 2008,2012; H�erisson, 2012), as Levallois points were systematicallyintroduced to the site or taken from the site. These observationshighlight the mobile character of this object in the Neanderthalhunting toolkit throughout the Middle Palaeolithic.

The point was omitted from lithic analyses of Middle Palae-olithic assemblages in northern France for a long time, in favour ofthe associated blade and Levallois production. However, they werepresent throughout the Middle Palaeolithic, from MIS 7 onwards,and were decisive and indispensable components of the Neander-thal hunting toolkit. We hope that this article marks the beginningof more systematic analyses of points in northern France, and moregenerally in north-western Europe.

References

Ameloot-Van der Heijden, N., 1993. L'industrie laminaire du niveau C.A. du gise-ment pal�eolithique moyen de Riencourt-l�es-Bapaume (Pas-de-Calais). Bulletinde la Soci�et�e Pr�ehistorique Française 90, 324e327.

Antoine, P., Auguste, P., Bahain, J.-J., Coudret, P., Depaepe, P., Fagnart, J.-P.,Falgu�eres, C., Fontugne, M., Frechen, M., Hatt�e, C., Lamotte, A., Laurent, M.,Limondin-Lozouet, N., Locht, J.-L., Mercier, N., Moigne, A.M., Munaut, A.-V.,Ponel, P., Rousseau, D.-D., 2003. Pal�eoenvironnements pl�eistoc�enes et peuple-ments pr�ehistoriques dans le bassin de la Somme (nord de la France). Bulletinde la Soci�et�e Pr�ehistorique Française 100, 5e28.

Antoine, P., Limondin-Lozouet, N., Auguste, P., Locht, J.-L., Galheb, B., Reyss, J.-L.,Escud�e, E., Carbonel, P., Mercier, N., Bahain, J.-J., Falgu�eres, C., Voinchet, P., 2006.Le tuf de Caours (Somme, Nord de la France) : mise en �evidence d'une s�equencede tufs calcaires eemiens et d'un site pal�eolithique associ�e. Quaternaire 17,281e320.

Antoine, P., Limondin-Lozouet, N., Chauss�e, C., Lautridou, J.P., Pastre, J.-F., Auguste, P.,Bahain, J.-J., Falgu�eres, C., Galehb, B., 2007. Pleistocene fluvial terraces fromnorthern France (Seine, Yonne, Somme): synthesis and new results. QuaternaryScience Reviews 26, 2701e2723.

Antoine, P., Goval, E., Jamet, G., Coutard, S., Moine, O., H�erisson, D., Auguste, P.,Gu�erin, G., Lagroix, F., Schmidt, E., Robert, V., Debenham, N., Meszner, S.,Bahain, J.-J., 2014. Les s�equences lœssiques pl�eistoc�ene sup�erieur d'Havrincourt(Pas-de-Calais, France) : stratgraphie, pal�eoenvironnement, g�eochronologie etoccupations pal�eolithiques. Quaternaire 25 (4), 295e349.

Auguste, P., 1995. Cadres biostratigraphiques et pal�eo�ecologiques du peuplementhumain dans la France septentrionale durant le Pl�eistoc�ene. Apports de l'�etudepal�eontologique des grands mammif�eres du gisement de Biache-Saint-Vaast(Pas-de-Calais). Th�ese de doctorat, Museum national d'Histoire naturelle,Paris, p. 1924.

Auguste, P., 2009. Evolution des peuplements mammaliens en Europe du Nord-Ouest durant le Pl�eistoc�ene moyen et sup�erieur. Le cas de la France septen-trionale. Quaternaire 20 (4), 527e550.

Blaser, F., 2013. Saint-Illiers-la-Ville, La Vall�ee des Pr�es. Du Pl�eistoc�ene moyen auPl�eistoc�ene sup�erieur : les occupations pal�eolithiques de Saint-Illiers-la-Ville.Inrap, SRA Centre-̂Ile-de-France, p. 559.

Bo€eda, E., 1988. Analyse technologique du d�ebitage du niveau IIA. In: Tuffreau, A.,Somm�e, J. (Eds.), Le gisement pal�eolithique moyen de Biache-Saint-Vaast (Pas-de-Calais), Stratigraphie, environnement, �etudes arch�eologiques (1�ere partie),vol. 21, pp. 85e214.

Bo€eda, E., 1994. Le concept Levallois : variabilit�e des m�ethodes, Paris, Monographiedu CRA, vol. 9. CNRS Editions, p. 283.

Bo€eda, E., Connan, J., Dessort, D., Muhesen, S., Mercier, N., Valladas, H., Tisnerat, N.,1996. Bitumen as a hafting material on Middle Palaeolithic artefacts. Nature380, 336e338.

Bonilauri, S., 2010. Les outils du Pal�eolithique moyen : une m�emoire techniqueoubli�ee ? Approche techno-fonctionnelle appliqu�ee �a un assemblage lithique deconception Levallois provenant du site d'Umm el Tlel (Syrie centrale). Th�ese del'Universit�e Paris Ouest Nanterre.

Bordes, F., 1961. Typologie du Pal�eolithique ancien et moyen. Delmas, Bordeaux,p. 101.

Claud, E., Thi�ebaut, C., Coudenneau, A., Deschamps, M., Mourre, V., Colonge, D.,2013. Le travail du bois au Pal�eolithique moyen : nouvelles donn�ees issues del'�etude trac�eologique de plusieurs industries lithiques d'Europe occidentale. Inregards crois�es sur les outils li�es au travail des v�eg�etaux : an interdisciplinaryfocus on plant-working tools. In: Anderson, P.C., Cheval, C., Durand, A. (Eds.),XXXIIIe rencontres internationales d'arch�eologie et d'histoire d'Antibes, Sous ladirection de. �Editions APDCA, Antibes, pp. 367e381.

Coudenneau, A., 2013. �El�ements triangulaires et armes de chasse au Pal�eolithiquemoyen. In: Constats et r�eflexions �a travers l'�etude techno-morpho-fonctionnellede quatre s�eries d'Europe occidentale. Universit�e d'Aix-Marseille, p. 366.

Delagnes, A., 1993. Un mode de production in�edit au Pal�eolithique moyen dansl'industrie du niveau 6e du Pucheuil (Seine-Maritime). Pal�eo 5, 111e120.

Delagnes, A., Ropars, A., 1996. Pal�eolithique moyen en pays de Caux (Haute-Nor-mandie). Le Pucheuil, Ettouteville : deux gisements de plein air en milieuloessique, Paris. Documents d'Arch�eologie Française 56, 243.

Depaepe, P., 2007. Le Pal�eolithique moyen de la vall�ee de la Vanne (Yonne, France) :mati�eres premi�eres, industries lithiques et occupations humaines. M�emoire XLIde la Soci�et�e Pr�ehistorique Française, p. 298.

Depaepe, P., Locht, J.-L., Swinnen, C., 1997. D�ecouverte de sites du Pal�eolithiquemoyen sur le trac�e d'un gazoduc en Picardie (France). Notae Praehistoricae 17,25e38.

Depaepe, P., Antoine, P., Guerlin, O., Swinnen, C., 1999. Le gisement pal�eolithiquemoyen de Blangy-Tronville (Somme). Revue Arch�eologique de Picardie 3e4,3e21.

Depaepe, P., Deschodt, L., 2001. Le site de Gouy-Saint-Andr�e “Le Savigny”.-Pal�eolithique et M�esolithique du Nord de la France : nouvelles recherches.Lille Universit�e des Sciences et Technologies de, Lille-Flandres-Artois,pp. 185e198.

Goval, E., 2008. D�efinitions, analyses et caract�erisation des territoires desN�eandertaliens au Weichselien ancien en France septentrionale. Approchestechnologiques et spatiales des industries lithiques, �elargissement au Nord-Ouest de l'Europe. Th�ese de doctorat. Universit�e des Sciences et Technologiesde Lille, p. 543.

E. Goval et al. / Quaternary International 411 (2016) 216e232232

Goval, E., 2012. Peuplements n�eandertaliens dans le nord de la France. In:Recherches Arch�eologiques, vol. 4. Inrap-CNRS, p. 312.

Goval, E., H�erisson, D., 2006. Coexistence des chaînes op�eratoires Levallois etlaminaires au sein des assemblages C12 et C de Riencourt-l�es-Bapaume (Pas-de-Calais, France). Notae Prehistoricae 26, 25e39.

Goval, E., Locht, J.-L., 2009. Remontages, syst�emes techniques et r�epartitions spa-tiales dans l'analyse du site weichselien ancien de Fresnoy-au-Val (Somme,France). Bulletin de la Soci�et�e Pr�ehistorique Française 106 (4), 653e678.

Goval, E., H�erisson, D., Antoine, P., Auguste, P., Bahain, J.-J., Claud, E., Coutard, S.,Font, C., Guerrin, G., Hulin, G., Jamet, G., Moine, O., Prilaux, G., Schmidt, E.,Valladas, H., 2013. Des derniers N�eandertaliens aux premiers Homo Sapiens :les occupations pal�eolithiques d'Havrincourt « Les Bosquets » (Pas-de-Calais).Rapport de fouille du Canal Seine-Nord Europe. Service R�egional de l'Arch-�eologie Pas-de-Calais, p. 432.

Guerlin, O., 2002. Revelles « Les Terres Sellier » (Somme). rapport d'�evaluation.Service R�egional de l'Arch�eologie de Picardie, p. 28.

Guthrie, R.-D., 1982. Mammals of the Mammoth Steppe as paleoecological in-dicators. In: Hopkins, D.M., et al. (Eds.), Paleoecology of Beringia. Academicpress, pp. 307e326.

H�erisson, D., 2012. �Etude des comportements des premiers N�eandertaliens du Nordde la France. Les occupations saaliennes des gisements de Biache-Saint-Vaast etde Therdonne. Th�ese de doctorat de l'Universit�e des Sciences et Technologies deLille, p. 502.

Knecht, H., 1997. Projectile points of bone, antler and stone: experimental explo-rations of manufacture and use. In: Knecht, H. (Ed.), Projectile Technology.Plenum Press, New York, pp. 191e212.

Koehler, H., 2008. L'apport du gisement des Osiers �a Bapaume (Pas-de-Calais) aud�ebat sur l'�emergence du Pal�eolithique moyen dans le Nord de la France.Bulletin de la Soci�et�e Pr�ehistorique Française 105 (4), 709e736.

Koehler, H., 2012. Comportements et identit�es techniques au Pal�eolithique moyendans le bassin parisien : une question d'�echelle d'analyse ? Universit�e ParisOuest Nanterre La D�efense.

Lazuen, T., 2012. European Neanderthal stone hunting weapons reveal complexbehaviour long before the appearance of modern humans. Journal of Archae-ological Science 39, 2304e2311.

Locht (dir.), J.-L. (Ed.), 2002. Le gisement de Bettencourt-Saint-Ouen (Somme,France) : cinq occupations du Pal�eolithique moyen au d�ebut de la derni�ereglaciation. Documents d'Arch�eologie Française, vol. 90. Maison des Sciences del'Homme, Paris, p. 176.

Locht, J.-L., 2004. Le gisement pal�eolithique moyen de Beauvais (Oise). Contribution�a la connaissance des modalit�es de subsistance des chasseurs de renne duPl�eniglaciaire inf�erieur du Weichselien. Th�ese de doctorat, Universit�e Lille,p. 209.

Locht (dir.), J.-L. (Ed.), 2008. Le gisement pal�eolithique moyen de Fresnoy-au-Val(Somme, France). Service r�egional de l'Arch�eologie de Picardie, p. 130.

Locht, J.-L., Antoine, P., Swinnen, C., 1995. Le gisement pal�eolithique de Plachy-Buyon (Somme). Revue arch�eologique de Picardie 3e33.

Locht, J.-L., Guerlin, O., 1997. Le gisement pal�eolithique moyen du Bois de laBocquill�ere, �a Attilly (Aisne). Rapport d'�evaluation. Service R�egional de l'Arch-�eologie de Picardie, p. 51.

Locht, J.-L., Antoine, P., Bahain, J.-J., Dwrila, G., Raymond, P., Limondin-Lozouet, N.,Gauthier, A., Debenham, N., Frechen, M., Rousseau, D.-D., Hatt�e, C., Haesaerts, P.,Metsdagh, H., 2003. Le gisement pal�eolithique moyen et les s�equencespl�eistoc�enes de Villiers-Adam (Val-d'Oise, France). Chronostratigraphie, envi-ronnement et implantations humaines. Gallia Pr�ehistoire 45, 1e111.

Locht, J.-L., Antoine, P., Auguste, P., Limondin-Lozouet, N., 2009. Caours “Les Pr�es”,Rapport triennal de fouille programm�ee. Service R�egional de l'Arch�eologie dePicardie, p. 140.

Locht, J.-L., Goval, E., Antoine, P., 2010. Reconstructing Middle Palaeolithic Hominidsbehaviour during OIS 5 in Northern France. In: Settlement Dynamics, vol. III.Institut für Ur-und Frügeschichte und Arch€aeologie des Mittelalters. Universit€atTübingen, pp. 329e355.

Locht, J.-L., Coutard, S., Antoine, P., Sellier, N., Ducrocq, T., Paris, C., Guerlin, O.,Kiefer, D., Defaux, F., Deschodt, L., Limondin-Lozouet, N., 2013a. Donn�eesin�edites sur le Quaternaire et le Pal�eolithique du nord de la France. RevueArch�eologique de Picardie 3 (4), 5e70.

Locht, J.-L., Sellier, N., Antoine, P., Koehler, H., Debenham, N., 2013b. Mauquenchy(Seine-Maritime, France) : mise en �evidence de deux niveaux d'occupation

pal�eolithique dans un sol gris forestier date du SIM 5a (D�ebut Glaciaireweichselien). Quaternaire 24 (3), 247e257.

Locht, J.L., Goval, E., Antoine, P., Coutard, S., Auguste, P., Paris, C., H�erisson, D., 2014.Palaeoenvironments and prehistoric interactions in northern France from theEemian Interglacial to the end of the Weichselian Middle Pleniglacial. In: Actesdu colloque “Where wild things are”, Durham, 2012. Oxbow Press, pp. 70e78.

Moncel, M.-H., Chac�on, M.-G., Coudenneau, A., Fernandes, P., 2009. Points andconvergent tools in the European Early Middle Paleolithic site of Payre (SE,France). Journal of Archaeological Science 36, 1892e1909.

Plisson, H., Beyries, S., 1998. Pointes ou outils triangulaires ? Donn�ees fonctionnellesdans le Moust�erien levantin. Pal�eorient 24 (1), 5e24.

R�evillion, S., 1994. Les industries laminaires du Pal�eolithique moyen en Europeseptentrionale : l'exemple des gisements de Saint-Germain-des-Vaux/Port-Racine (Manche), Seclin (Nord), et de Riencourt-l�es-Bapaume (Pas-de-Calais).Publications du CERP 5, 187.

R�evillion, S., Tuffreau, A., 1994. Les industries laminaires au Pal�eolithique moyen,Actes de la table ronde internationale organis�ee par l'ERA 37 du CRA-CNRS.Dossier de Documentation Arch�eologique 18, 193.

Rots, V., 2011. Tool use and Hafting in the Western European Middle Palaeolithic, LePal�eolithique moyen en Belgique, m�elanges Marguerite Ulrix-Closset. In: Dir.Toussaint, M., Di Modica, K., Pirson, S. (Eds.), Eraul 128 (4), 277e287.

Rots, V., 2013. Insights into early Middle Palaeolithic tool use and hafting in WesternEurope. The functional analysis of level IIa of the early Middle Palaeolithic site ofBiache-Saint-Vaast (France). Journal of Archaeological Science 40, 497e506.

Rubens, K., 2007. Bifacial elements in continental northwestern Europe during thelast glacial cycle (MIS5d-3): the relationship between Mousterian, Micoquianand ‘Mixed’ assemblages. Institute of Archaeology 18, 84e100.

Rubens, K., Di Modica, K., 2011. Le Pal�eolithique moyen en Belgique. M�elangesMarguerite Ulrix-Closset. Eraul 128 (4), 239e260.

Sellier, N., 2015. Bilan des recherches sur le Pal�eolithique moyen de l'Aisne : �etudedes assemblages lithiques du D�ebut glaciaire weichselien au Pl�eniglaciaire duWeichselien. Actes de la table ronde d'Amiens. M�emoires de la Soci�et�ePr�ehistorique Française 59, 75e99.

Sellier, N., Coutard, S., 2007. Donn�ees r�ecentes sur le Pal�eolithique moyen de l'Aisne.Une occupation du Weichselien ancien �a Courmelles. Revue Arch�eologique dePicardie 3 (4), 5e16.

Shea, J.-J., 1988. Spear point from middle Palaeolithic of the Levant. Journal of FieldArchaeology 15 (4), 441e450.

Shea, J.-J., 2003. The Middle Paleolithic of the East Mediterranean Levant. Journal ofWorld Prehistory 17 (4), 313e394.

Shea, J.-J., 2006. The origins of lithic projectile point technology: evidence fromAfrica, the Levant, and Europe. Journal of Archaeological Science 33, 823e846.

Soressi, M., Locht, J.-L., 2010. Les armes de chasse de Neandertal. Premi�ere analysedes pointes moust�eriennes d'Ang�e. Arch�eopages 28, 6e11.

Swinnen, C., Locht, J.-L., Antoine, P., 1996. Le gisement moust�erien d'Auteuil (Oise).Bulletin de la Soci�et�e Pr�ehistorique Française 93 (2), 173e182.

Tuffreau, A., 1976. Les fouilles du gisement Acheul�een sup�erieur des Osiers �aBapaume (Pas-de-Calais). Bulletin de la Soci�et�e Pr�ehistorique Française 73,231e243.

Tuffreau, A., 1987. Le Pal�eolithique inf�erieur et moyen du Nord de la France (NordPas-de-Calais, Picardie) dans son cadre stratigraphique. Th�ese de doctorat d'�Etatde l'universit�e des Sciences et Technologies de Lille, p. 609.

Tuffreau, A., 1989. Le gisement moust�erien de Champvoisy (Marne), Pal�eolithique etM�esolithique du Nord de la France : nouvelles recherches 1. Universit�e desSciences et des Techniques de Lille-Flandre-Artois, pp. 69e77.

Tuffreau (dir.), A. (Ed.), 1993. Riencourt-l�es-Bapaume (Pas-de-Calais). Un gisementdu Pal�eolithique moyen, Paris. Document Arch�eologie Française, vol. 37. Edi-tions de la Maison des Sciences de l'Homme, p. 128.

Tuffreau, A., R�evillion, S., 1984. Le gisement pal�eolithique moyen de Seclin (Nord) :premiers r�esultats de la campagne de fouille 1983. Cahier de G�eographiePhysique 5, 31e46.

Tuffreau, A., Somm�e, J., 1988. Le gisement pal�eolithique moyen de Biache-Saint-Vaast (Pas de Calais). M�emoires de la Soci�et�e Pr�ehistorique Française 21, 340.

Villa, P., Lenoir, M., 2006. Hunting weapons of the Middle Stone Age and the MiddlePalaeolithic : spear points from Sibudu, Rose Cottage and Bouheben, SouthernAfrica. Humanities 18, 89e122.

Wilkins, J., Schoville, B.-J., Brown, K.-S., Chazan, M., 2012. Evidence for early haftedhunting technology. Science 338, 942e946.