No Horns on their Helmet? Essays on the Insular Viking Age

Celtic, Anglo-Saxon, and Scandinavian Studies I

NO HORNS ON THEIR HELMETS?ESSAYS ON THE INSULAR VIKING-AGE

CELTIC, ANGLO-SAXON, AND SCANDINAVIAN STUDIES

ISSN 2051-6509

General Editor: David N. Dumville

Editorial Advisory Board

Professor Stefan Brink Professor Colm Ó Baoill(University of Aberdeen) (University of Aberdeen)

Professor Julia Crick Professor Ralph O’Connor(King’s College, London) (University of Aberdeen)

Professor Patrick Crotty Dr Aideen O’Leary(University of Aberdeen) (University of Aberdeen)

Professor Daniel Donoghue Professor Pádraig Ó Riain(Harvard University) (National University of Ireland, Cork)

Professor Catherine McKenna Professor Paul Russell(Harvard University) (University of Cambridge)

Professor Stephen Mitchell Professor Jane Stevenson(Harvard University) (University of Aberdeen)

ii

NO HORNS

ON THEIR HELMETS?

ESSAYS

ON THE

INSULAR VIKING-AGE

CLARE DOWNHAM

THE CENTRE FOR ANGLO-SAXON STUDIES

AND

THE CENTRE FOR CELTIC STUDIES

UNIVERSITY OF ABERDEEN

iii

Celtic, Anglo-Saxon, and Scandinavian Studies

ISSN 2051-6509

No Horns on their Helmets?Essays on the Insular Viking-Age

First published 2013by

The Centre for Anglo-Saxon Studiesand

The Centre for Celtic StudiesUniversity of Aberdeen

ISBN 978-0-9557720-1-6Printed in the E.U. on acid-free paper

The right of Clare Elizabeth Downham to be identified as the author of this work

has been asserted by her in accordance with the Copyright, Designs and Patents

Act 1988.

All rights reserved. No part of this publication may be reproduced, stored in a

retrieval system, or transmitted, in any form or by any means, electronic,

mechanical, photocopying, recording, or otherwise, without prior permission of

the publisher.

Composed by Ruth Fitzsimmons

Designed by Martin Cooper, Uniprint

Printed by The University of Aberdeen

iv

Contents

Previous places of publication of these essays vi

General Editor’s foreword vii

I SMOKE AND MIRRORS

1 An imaginary viking-raid on Skye in 795? 3

2 Annals, armies, and artistry: ‘The Anglo-Saxon Chronicle’, 865–96 9

II INSULAR IDENTITIES IN THE VIKING-AGE

3 ‘Hiberno-Norwegians’ and ‘Anglo-Danes’: anachronistic ethnicities and Viking-Age England 41

III VIKINGS AS SEEN FROM IRELAND

4 Irish chronicles as a source for rivalry between vikings, A.D. 795–1014 75

5 The career of Cerball mac Dúnlainge, overking of the Osraige 91

6 The good, the bad, and the ugly: portrayals of vikings in ‘The Fragmentary Annals of Ireland’ 111

IV SCANDINAVIAN SETTLEMENT IN IRELAND

7 The historical importance of Viking-Age Waterford 129

8 Living on the edge: Scandinavian Dublin in the twelfth century 157

V MYSTERIES OF THE IRISH-SEA ZONE

9 Eric Bloodaxe – axed? The mystery of the last Scandinavian king of York 187

10 Cold feet: King Magnús and the Irish Sea 209

v

Previous places of publication of these essays

1 Scottish Gaelic Studies 20 (2000) [ISSN 0080-8024]

2 Anglo-Saxon 4 (2010) [ISSN 1754-372X]

3 Mediaeval Scandinavia 19 (2009) [ISSN 0076-5864]

4 Northern Scotland 26 (2006) [ISSN 0306-5278]

5 Ossory, Laois and Leinster 1 (2004) [ISSN 1649-4938]

6 The Medieval Chronicle 3 (2004) [published by Editions Rodopi N.V.,

Amsterdam: ISBN 90-420-1834-8]

7 The Journal of Celtic Studies 4 (2004) [ISSN 1781-1406]

8 West over Sea. Studies in Scandinavian Sea-borne Expansion and

Settlement before 1300. A Festschrift in Honour of Dr Barbara E.

Crawford, edd. Beverley Ballin Smith et al. (Leiden: Brill 2007)

[ISBN 978-90-04-15893-1]

9 Mediaeval Scandinavia 14 (2004) [ISSN 0076-5864]

10 The Journal of Celtic Studies 5 (2005) [ISSN 1781-1406]

The author and the publishers of this volume are much indebted to the original

publishers of chapters 5, 6, and 8 for their willingness to allow – indeed, their

friendly encouragement to pursue – republication of those essays.

vi

General Editor’s Foreword

When Clare Downham resigned her lectureship in Celtic and History at The University of Aberdeen with effect from 1 February, 2010, after elevensemesters’ service, to take up a new challenge at another university, her colleagueshere wished to offer her a token of their gratitude for her energetic exercise of herrole in this university, especially in playing a full part both in the rebuilding ofthe Department of Celtic & Gaelic and in the development of The Centre forCeltic Studies and The Centre for Anglo-Saxon Studies.

Eventually a decision was taken to offer Dr Downham a volume of her paperspublished in Aberdonian journals, 2000–2010, to which have been added (at herrequest) three closely related papers from other publications (and we are indebtedto the generosity of the editors and publishers of those volumes for granting usappropriate permission; these papers have been reëdited into our house-style).

We take pride in issuing this as the first volume of a new book-series reflectingthe names of our new degree-courses, the single-honours M.A. in Celtic &Anglo-Saxon Studies, and the joint M.A. in Celtic & Anglo-Saxon withHistory/Scandinavian Studies. In this connexion we are grateful to those colleagues– from Harvard University, King’s College London, University College Cork, andThe University of Cambridge – who have agreed to join an equal number ofAberdonians on the Editorial Advisory Board.

D.N.D.

July 2012

vii

FOR JENNY

Diuina prouidencia labili consuluit mundo scribendisollicitudine litterarum tenaci memoria, ut humanum genus, quodmorte interueniente non potest permanere, scribendi suffragio diuina

uel humana magnalia posteris temporibus recoleret

(Jocelin of Furness, Vita Sancti Patricii)

viii

I

SMOKEAND

MIRRORS

1

An imaginary viking-raid on Skye in 795?

ONE of the earliest deeds attributed to vikings on the west coast of Scotland isthe pillaging of Skye in the year 795. It has been referred to by eminent scholarsin the field of viking-studies.1 Nevertheless, investigation of the textual evidencesuggests that the record of this raid arises from an error made in good faith around1500.

‘The Annals of Ulster’, generally regarded as one of the most reliablechronicle-sources for mediaeval Gaelic history, provide the sole witness to this raidon Skye. I do not doubt that estimate of their value. However, some considerationof textual history shows that the supposed reference to Skye is a misinterpretation.

‘The Annals of Ulster’ have survived principally in two late mediaevalmanuscripts. The earlier copy (Dublin, Trinity College, MS. 1282 [H.1.8], alsoknown as ‘The Book of Seanadh Mac Maghnusa’) was written by Ruaidhrí ÓLuinín as far as the annal for 1489.2 The text was continued by other scribes, butÓ Luinín seems to have completed his section of the work by 1492.3 Thesuccessive scribes of this chronicle (including Ó Luinín) preserved in considerablemeasure the original spellings of their exemplars, and study of the language of‘The Annals of Ulster’ has suggested that much of the material within thatchronicle was recorded contemporaneously with events.4 The second manuscript(Oxford, Bodleian Library, MS. Rawlinson B.489 [S.C. 11836]) was a copy andaugmentation of the first. This too was written by Ruaidhrí Ó Luinín, in this caseas far as the annal for 1528.5

1Barbara E. Crawford, Scandinavian Scotland (Leicester 1987), p. 40; James Graham-Campbell& C.E. Batey, Vikings in Scotland – an Archaeological Survey (Edinburgh 1998), p. 43; Anna Ritchie,Viking Scotland (London 1993), p. 19.

2The Annals of Ulster (to A.D. 1131), I, edd. & transl. Seán Mac Airt & G. Mac Niocaill (Dublin1983), p. ix.

3Aubrey Gwynn (ed. Nollaig Ó Muraíle), Cathal Óg Mac Maghnusa and The Annals of Ulster(Enniskillen 1998), p. 51.

4Tomás Ó Máille, The Language of The Annals of Ulster (Manchester 1910).5For the most thorough discussion, see Brian Ó Cuív, Catalogue of Irish Language Manuscripts

in The Bodleian Library at Oxford and Oxford College Libraries (2 vols, Dublin 2001/3), I.153–63(no. 33).

3

Scottish Gaelic Studies 20 (2000) ISSN 0080 - 8024

Here is the relevant entry for 795.6

AU 795.3:Loscadh Rechrainne ó geinntib, 7 Sc(r)í do <cho>scradh 7 do lomradh.‘The burning of Rechru by heathens, and Sc(r)í was overwhelmed and laid waste.’

The vital word in this entry, printed by editors as Scí (and taken to be Skye), waswritten as scri by Ó Luinín in the Dublin-manuscript. This nonsense-word wasemended by underpointing the r for deletion. Seán Mac Airt and Gearóid MacNiocaill noted this manuscript alteration confusingly in their edition of ‘TheAnnals of Ulster’ and called the corrector H¹ (but it is not clear how the scriberesponsible for a single point can be identified).7 It should be said that W.M.Hennessy had reported the manuscript faithfully and with clarity (as has beenconfirmed by reference to a photostat of the Dublin-manuscript).8 In themanuscript (folio 35va21) we read 7scridoscradh, written all as one unintelligibleword. Thus the idea of unambiguous reference to Skye in the chronicle-record issomewhat tenuous. Something other than Scí may have been intended in theexemplar of the Dublin-manuscript of ‘The Annals of Ulster’.

An alternative is suggested by a similar entry relating to the year 795 in ‘TheAnnals of The Four Masters’. This reads as follows.9

AFM 790[=795].2:Losccadh Rechrainde ó dhíbhearccaibh, agus a sccríne do chosccradh agus do lomradh.‘The burning of Rechru by marauders, and its shrines were broken and plundered.’

The resemblance of this report to that of ‘The Annals of Ulster’ is striking. Themain differences are use of the words díbhearccach (‘marauder’) and sccrín(‘shrine’). Although the scrín / Scí differentiation radically affects the interpretationof events, the entries seem to represent a single source, and record of the sameincident seems intended. (The similarity of both reports was also remarked byHennessy and MacCarthy.)10

6The Annals of Ulster, I, edd. & transl. Mac Airt & Mac Niocaill, p. 250 (795.3).7Ibid., p. 250, n. b.8Annala Uladh, Annals of Ulster, otherwise Annala Senait, Annals of Senat; a Chronicle of Irish

Affairs, edd. & transl. William M. Hennessy & B. MacCarthy (4 vols, Dublin 1887–1901), I.275,n. 10.

9Annala Rioghachta Eireann. Annals of the Kingdom of Ireland by The Four Masters, from theEarliest Period to the Year 1616, ed. & transl. John O’Donovan (2nd edn, 7 vols, Dublin 1856),I.396.

10Annala Uladh, edd. & transl. Hennessy & MacCarthy, I.275, n. 10; Early Sources of ScottishHistory, A.D. 500 to 1286, transl. Alan Orr Anderson (2 vols, Edinburgh 1922; rev. imp., byM.O. Anderson, Stamford 1990), I.255–6.

NO HORNS ON THEIR HELMETS? 4

It is likely that the record in ‘The Annals of The Four Masters’ was derivedfrom ‘The Annals of Ulster’. The manuscript known as ‘The Book of SeanadhMac Maghnusa’ was cited as a source used by The Four Masters in theintroduction to their chronicle;11 Míchél Ó Cléirigh and his colleagues completedthis in 1636 (Dublin, Royal Irish Academy, MS. Stowe C.iii.3 [1220]).12 UnlikeÓ Luinín, Ó Cléirigh and his companions modernised the language of theirsources. In this entry they must have taken scri to mean scrín and adjusted theinflexion accordingly. There is no evidence that either the compilers of ‘The Annalsof The Four Masters’ or the ‘corrector’ of ‘The Annals of Ulster’ had anothersource which would have influenced their respective interpretations. The case forscrín or Scí in the original chronicle-record therefore depends on one letter beingadded to or subtracted from the meaningless form scri.

It is possible that, in the exemplar before Ruaidhrí Ó Luinín or one of hispredecessors, the letter n was originally expressed by a suprascript stroke.Presumably it gave scrı� for scrin. This was the commonest of abbreviations inmediaeval European manuscripts.13 The corrector of the Dublin-manuscript of‘The Annals of Ulster’ sought to make sense of this miscopied word by removalof the r to read sci (for Scí, Skye), and it is likely that he rewrote history in theprocess.

It is curious that John O’Donovan, in quoting ‘The Annals of Ulster’, gave acorrected and somewhat modernised form of the entry.14 He then added aquotation derived from a seventeenth-century Hiberno-English translation of‘The Annals of Ulster’ (now London, British Library, MS. Additional 4795),which follows this corrected form:15

“A.D. 794. Losgad Rachrainne o Gentib ocus a scrine do coscradh ocus do lomrad.”[The burning of Rechrainn by Gentiles, who spoyled and impoverished the shrines. –Cod. Clarend., 49.]

I have not been able to find the origin of this form of the entry, but it providesfurther evidence for a perception that ‘The Annals of Ulster’ recorded theplundering of shrines at Rechru.

11Annala Rioghachta Eireann, ed. & transl. O’Donovan, I.lxiv–lxv; Gwynn, Cathal Óg MacMaghnusa, p. 39; cf. M.A. O’Brien, ‘Miscellanea hibernica: 5. A wrong entry in AU and FM’, Étudesceltiques 3 (1938) 365–6, for a different approach to the relationships of these chronicles in respectof another problematic entry.

12Paul Walsh, The Four Masters and their Work (Dublin 1944), p. 1.13W.M. Lindsay, Notae Latinae. An Account of Abbreviation in Latin MSS. of the Early Minuscule

Period (c. 700–850) (Cambridge 1915), p. 1.14Annala Rioghachta Eireann, ed. & transl. O’Donovan, I.397, n. i.15Ibid., I.xxxiv, and Annala Uladh, edd. & transl. Hennessy & MacCarthy, IV.vi-vii.

VIKING-RAID ON SKYE IN 795? 5

A search for similar chronicle-references for 795 has yielded records of theburning of Rechru in Irish and Welsh texts, but no mention of Skye or shrines.16

AClon 792.2:Rachrynn was burnt by the Danes.

ByS (P) 785 [=795]:Pump mlyned a phedwar ugeint a seithgant oed oed Krist pan doeth y paganyeidYwerdon ac y distrywyd Rechrenn.‘785 was the year of Christ when the heathens first came to Ireland, and Rechra washarried.’

ByS (S) 795:Yd aeth paganyeit gyntaf y Iwerdon, ac y diffeithwyt Rechreyn.‘Heathens first went to Ireland, and Rechra was ravaged.’

It is evident that there was a religious establishment at Rechru which may havehoused shrines to be plundered in 795. Deaths of its ecclesiastics are recorded in‘The Annals of Ulster’ from 739 to 975.17 It is also worth remarking that ‘TheAnnals of The Four Masters’ share with ‘The Annals of Tigernach’ a notice of theplundering of Rechru by Foreigners (Gaill) in 1038.18 Ecclesiastical Rechru’sidentification as Lambay Island (Co. Dublin) was suggested by O’Donovan andhas been convincingly verified by Máire Herbert.19 This Rechru was not RathlinIsland, Co. Antrim, as has been often supposed.20 This church on Lambay was

16The Annals of Clonmacnoise, being Annals of Ireland from the Earliest Period to A.D. 1408, ed.Denis Murphy (Dublin 1896), p. 127; Brenhinoedd y Saeson, ‘The Kings of the English, A.D. 682–1097: Texts P, R, S in Parallel, ed. & transl. David N. Dumville (Aberdeen 2012), pp. 14/15.

17The Annals of Ulster, I, edd. & transl. Mac Airt & Mac Niocaill, pp. 192 (739.6), 196 (743.3),222 (769.3), 226 (773.8), 254 (799.6), 308 (850.1) – holding the abbacy in plurality with that ofanother Columban house, Durrow –, 412 (975.3) – an airchinnech killed by gentiles. This sequenceis shared with ‘The Annals of The Four Masters’, at uncorrected dates 738, 743, 764, 768, 794,848, and 973.

18Annala Rioghachta Eireann, ed. & transl. O’Donovan, II.834 (1038.9); The Annals of Tigernach,ed. & transl. Whitley Stokes (2nd edn, 2 vols, Felinfach 1993), II.269 (377) (1038.1).

19Annala Rioghachta Eireann, ed. & transl. O’Donovan, I.397, n. h; Máire Herbert, Iona, Kells,and Derry. The History and Hagiography of the Monastic Familia of Columba (Oxford 1988), pp. 42–3; cf. Diarmuid Ó Murchadha, The Annals of Tigernach. Index of Names (London 1997), pp. 176–7.

20For example, by Crawford, Scandinavian Scotland, p. 40; L. Mac Mathúna, ‘The vikings inIreland: contemporary reaction and cultural legacy’, in Celts and Vikings. Proceedings of the FourthSymposium of Societas Celtologica Nordica, ed. Folke Josephson (Göteborg 1997), pp. 41–65;Graham-Campbell & Batey, Vikings in Scotland, p. 43; but not misidentified in Adomnán of Iona,Life of St Columba, transl. Richard Sharpe (Harmondsworth 1995), p. 82.


part of the ecclesiastical federation of St Columba, whose leading monastery, Iona,was plundered in 795 according to ‘The Annals of Inisfallen’.21

AI [795].2:Orcain Iæ Coluim Chille ocus Inse Muirethaig ocus Inse Bó Finne.‘The plundering of Í Coluim Chille and of Inis Muiredaig and of Inis Bó Finne.’

The evidence suggests that reference to the plundering of Skye in ‘The Annals ofUlster’ cannot be taken at face-value. It is a late textual alteration which resultedfrom a double scribal error, and the report is not substantiated by other primarysources. Thus there is no historical evidence that Skye was raided in 795. It ismore likely that the exemplar or source of ‘The Annals of Ulster’ recorded thelooting of sacred objects of the church at Rechru (Lambay) in 795 than that itattested a plundering of Skye.22

21The Annals of Inisfallen (MS. Rawlinson B.503), ed. & transl. Seán Mac Airt (Dublin 1951),p. 118; The Annals of Inisfallen reproduced in Facsimile from the Original Manuscript (Rawlinson B.503)in The Bodleian Library, facs. edd. R.I. Best & E. MacNeill (Dublin 1933), folio 13vc16–18.

22I should like to thank David Dumville for encouraging me to write this paper and for hisimproving criticisms and suggestions.

VIKING-RAID ON SKYE IN 795? 7

2

Annals, armies, and artistry:‘The Anglo-Saxon Chronicle’, 865–96

‘THE ANGLO-SAXON CHRONICLE’ from 865 to 896 is an engrossingdescription of affairs in England during the mature years of Alfred the Great, kingof the West Saxons and then overking of the Anglo-Saxons (871–99). Much ofthe narrative is preöccupied with the description of viking-campaigns, and it is amajor source for understanding how vikings first came to conquer and settleEnglish territory. Nevertheless, it is striking that the presentation of informationin ‘The Anglo-Saxon Chronicle’ for those years was influenced by stylistic andpolitical considerations. These can provide important clues to the circumstancesof the composition of annals 865 to 896.

For the years 865–96 there seem to be two distinct phases of chroniclingactivity in ‘The Anglo-Saxon Chronicle’. The first runs from A.D. 864/5 to 891/2(annals 865–92) and belongs to the Chronicle’s ‘Common Stock’ (60 B.C.–A.D.892), while the second constitutes its first continuation, for the four years 893–6. Ruth Waterhouse has discussed the former section.1 She has drawn attentionto the distinct word-order of annals 865–91 and the stylistic features (such as itsverbs of motion) which distinguish it from what precedes and what follows.2 PeterSawyer has argued persuasively that this section properly ends at 892 (not 891),which is therefore where that ‘Common Stock’ of the Chronicle ends.3 It is alsoin this section that the beginning of the year was calculated from September.4

1R. Waterhouse, ‘Stylistic features as a factor in detecting change of source in the ninth-centuryAnglo-Saxon Chronicle’, Parergon 27 (1980) 3–8.

2Cecily Clark pointed out that the opening words, Her for se here, or a minor variant, characteriseall annals from 867 to 887. This seems to be an indicator of unity within this section of the chronicle:C. Clark, ‘The narrative mode of The Anglo-Saxon Chronicle before the Conquest’, in England beforethe Conquest. Studies in Primary Sources presented to Dorothy Whitelock, edd. Peter Clemoes &K. Hughes (Cambridge 1971), pp. 215–35, at p. 219.

3P.H. Sawyer, The Age of the Vikings (London 1962), pp. 13–16, 18–19, 20–1; cf. DavidN. Dumville, Wessex and England from Alfred to Edgar. Six Essays on Political, Cultural, andEcclesiastical Revival (Woodbridge 1992), p. 89, and A. Meaney, ‘The Anglo-Saxon Chronicle c. 892:materials and transmission’, Old English Newsletter 18 (1984/5), no. 2, pp. 26–35.

4M.L.R. Beaven, ‘The beginning of the year in the Alfredian Chronicle (866–87)’, EnglishHistorical Review 33 (1918) 328–42. This needs to be kept in mind when comparing ‘TheAnglo-Saxon Chronicle’ with other sources (such as ‘The Annals of Saint-Vaast’: see below, nn. 71,91, 101, 103–4, 131) in which the chronicle-year begins at Christmas.

Anglo - Saxon 4 (2010) ISSN 1754 - 372X

9

The distinctiveness of the second section, comprising annals 893–6, was notedby Cecily Clark.5 Here the year seems to begin at the end of December or thebeginning of January (although Dorothy Whitelock suggested that it began at theend of the campaigning year).6 In these years the annals are longer, the author’svoice is more apparent (he referred to himself as ic, ‘I’, in annal 894), and a moremarked pro-Alfredian bias is demonstrated in the author’s analysis of motive andin his rhetorical digressions.7

In annals 892 and 896, cross-references to earlier events reïnforce the sense ofthese years as boundaries between sections and demonstrate the internal cohesionof those sections. Despite this division of ‘The Anglo-Saxon Chronicle’ from 865to 896 into two parts, we see in these annals a similar method of presenting pastevents. In that respect, Thomas Shippey’s comment on the annals 893–6 isrelevant for the whole period: the author ‘likes to see history as a chain, not a net’.8

In other words, instead of presenting all the major events of successive years in anunconnected fashion (in ‘chronicle-style’ as defined by Clark), the author selectedand developed a main narrative thread which links affairs from one year to thenext.9 Ultimately the choice of story-line seems to be determined by identifyingthat viking-army which was the biggest and least settled, and which thus posedthe greatest threat to the West Saxons. This identification then determined theprioritisation and organisation of information presented in each annal.

In annals 865 to 880, the main narrative focus is provided by the selectedmovements of an army whose arrival in England is announced in annal 865. Theprogress of this army in conquering three of the tetrarchic kingdoms – of theNorthumbrians, Mercians, and East Angles – is narrated. The great danger whichthe army posed to the remaining kingdom, that of the West Saxons, was stemmedat the battle of Edington (Wiltshire) in 878, but this was followed by a tense yearwhen the defeated viking-army camped at Cirencester (Gloucestershire), in Merciabut near the West-Saxon border, before settling down in East Anglia in 880.10

After that, it ceased to be a threat for some years, and in consequence it was notfurther commented on.

5Clark, ‘The narrative mode’, pp. 221–4.6Two of the Saxon Chronicles Parallel, with supplementary Extracts from the Others, ed. Charles

Plummer (2 vols, Oxford 1892–9; rev. imp., by D. Whitelock, 1952), II.cxli.7Clark, ‘The narrative mode’, p. 223; T.A. Shippey, ‘A missing army: some doubts about the Alfredian

Chronicle’, In Geardagum 4 (1982) 41–55, at p. 46. A revised and expanded second edition of Shippey’sarticle (with the same title) may be found in Anglo-Saxon 1 (2007) 319–38; see pp. 331–2.

8Shippey, ‘A missing army’ (1982), p. 50; (2007), p. 336.9Clark, ‘The narrative mode’, pp. 217–20.10Dorothy Whitelock, ‘The importance of the battle of Edington’, Report of The Society of Friends

of the Priory Church of Edington, Wiltshire (1975–7) 6–15, at p. 13, reprinted in her book From Bedeto Alfred. Studies in Early Anglo-Saxon Literature and History (London 1980), essay XIII, with thesame pagination.


Undoubtedly the army (Old-English here – the term most frequently used in‘The Anglo-Saxon Chronicle’) of annal 880 was rather different in character fromthat whose arrival is reported in annal 865. Reïnforcements arrived, as in 871,and other branches settled down. For example, Hálfdan and his followers settledamong the Northumbrians in 876 but they cease to be mentioned as they nolonger posed an active threat to the West Saxons (we can probably surmise thatHálfdan’s force continued to have a large impact on affairs in Northumbria).11

Inevitably, therefore, we see a degree of selectivity in what is recorded. Thedevelopment of a sustained narrative can also obscure some of the complexities ofaffairs by focusing on one sequence of events.

From 880 the narrative focus has shifted to an army which assembled(gegadrode) at Fulham (Middlesex) in 879.12 This seems to have recruited vikingsfrom England before departing for the Continent. Part of this army returned in892. Either in retrospect (which seems likely, as I argue below)13 or through WestSaxons’ fear of this army’s return, ‘The Anglo-Saxon Chronicle’ reports theover-wintering stations and some other aspects of this army’s activities abroad.14

This is the narrative focus until the army returns from the Continent (and asmaller army under the authority of one Hæsten also arrives on English coast) inthe year 892/3. As Shippey has demonstrated, the narrative spotlight is on thebigger army; Hæsten temporarily disappears from narrative view but reëmergeswhen there is something relevant to report in relation to ‘the main line of action,at Benfleet’.15 This continues until annal 896 where the dispersal of the army isreported.

The narrative in ‘The Anglo-Saxon Chronicle’ gives the impression that it is afull account of the actions of one viking-army from late 865 to 880, and a second

11After the annal for 876 there is silence in ‘The Anglo-Saxon Chronicle’ regardingNorthumbrian affairs until annal 893 (the first annal of the continuation of the Common Stock).It is only broken when the vikings of Northumbria sent ships to raid Wessex. Cross-reference withother Insular records suggests that other activities are not recorded – for example, the departure ofthe viking-leader Ívarr from East Anglia in 870: The Annals of Ulster (to A.D. 1131), I, edd. & transl.Seán Mac Airt & G. Mac Niocaill (Dublin 1983), pp. 326/7 (870.6).

12Annal 879: The Anglo-Saxon Chronicle. A Collaborative Edition, III, MS A, ed. Janet M. Bately(Cambridge 1986), p. 59 (s.a. 879); The Anglo-Saxon Chronicle. A Collaborative Edition, IV,MS B,ed. Simon Taylor (Cambridge 1983), p. 37 (s.a. <880>); The Anglo-Saxon Chronicle. A CollaborativeEdition, V, MS. C, ed. Katherine O’Brien O’Keeffe (Cambridge 2001), p. 62 (s.a. 880); TheAnglo-Saxon Chronicle. A Collaborative Edition, VI, MS D, ed. G.P. Cubbin (Cambridge 1996),p. 27 (s.a. 879); The Anglo-Saxon Chronicle. A Collaborative Edition, VII, MS. E, ed. Susan Irvine(Cambridge 2004), p. 51 (s.a. 879); The Anglo-Saxon Chronicle. A Collaborative Edition, VIII, MSF, ed. Peter S. Baker (Cambridge 2000), p. 72 (s.a. 879).

13See below, pp. 25, 28, 35-6.14Dumville, Wessex, p. 89.15Shippey, ‘A missing army’ (1982), p. 51; (2007), p. 336.

‘THE ANGLO SAXON CHRONICLE’, 865–96 11

viking-army from 879 to 896. In Asser’s version of events, this is made yet morepronounced by his continued reference to praefatus exercitus.16 However, there isenough evidence to enable one to argue that the Chronicle is not comprehensivein its report. Furthermore, the Chronicle occasionally reveals an awareness ofmultiple viking-armies in England; but, as Shippey has shown, priority is givento one line of action while others receive less attention.17

Thus we see from annals 866 to 896 a process of authorial selectivity both inidentifying which thread of viking-activities will constitute ‘the main line of action’and in conscious linking of events from year to year in relation to it. Other eventsand people are included either where they relate to the main narrative thread orwhere they have independent significance for the history of the West-Saxonkingdom (for example, Alfred’s defensive action against small groups ofviking-ships, arrivals at or departures from his court, his relationship to the papacyor the Carolingians, the deaths of worthies, and heavenly portents). This methodhas its benefits in producing a fairly coherent narrative with a sense ofchronological progression from one year to the next (can you almost hearF.M. Stenton issuing a sigh of relief when he reached the year 865 in his account?– pages 246–65 of the third edition of his book Anglo-Saxon England seem a fairlyeasy ride through events as narrated by ‘The Anglo-Saxon Chronicle’).18 However,there are some pitfalls in this approach, namely that the Chronicle-authors hadmade choices about what to include in and exclude from the narrative. It is almostcertainly a highly incomplete record of events, but this is sometimes obscured bythe way in which the narrative was composed.

It can be argued that clarity comes at the cost of exclusion in the Chronicle’saccount. If one wishes to view English history for these years from anon-West-Saxon perspective, it is necessary to probe some points in ‘TheAnglo-Saxon Chronicle’ which hint at, but gloss over, a wider range of events. Inthis analysis, I shall proceed by going through events annal by annal. At the endI shall offer some observations, by way of conclusion, regarding the compositionof this part of ‘The Anglo-Saxon Chronicle’.

Annal 865 (for A.D. 864/5) marks the beginning of the first section underdiscussion. It reports the encampment of a heathen army (hæþen here) on Thanet.19

Neither the chronicler nor Asser provided any clue as to the army’s origin.

16De rebus gestis Ælfredi, §§26, 30, 31, 37, 44, 57, and supra memoratus exercitus, §§32, 45, 46,47, 49, 52: Asser’s Life of King Alfred, ed. William Henry Stevenson (Oxford 1904; rev. imp., byD. Whitelock, 1959), pp. 22–40.

17Shippey, ‘A missing army’ (1982), p. 44; (2007), p. 329.18F.M. Stenton, Anglo-Saxon England (3rd edn, Oxford 1971), pp. 246–65.19Annal 865: ASC.A, p. 46; ASC.B, s.a. <866>5, p. 33; ASC.C, s.a. 866, p. 57; ASC.D, p. 24;

ASC.E, s.a. 865, p. 48.


However, this record does give the first report of tribute being paid to vikings bythe English: the people of Kent promised money for peace.20 Despite this treaty,the viking-army ravaged East Kent; but thereafter its fate is unknown. One maysuppose that it then left England, although it is possible that all or part of it joinedthe army which arrived in the kingdom of East Angles in late 865 or that itremained in England but fades into obscurity as far as the records are concerned.Annal 865 is a reminder that the fleet whose arrival would be reported in thefollowing annal need not have been the only one in English waters at that time.

In 865/6 the infamous micel here came into England and took upwinter-quarters in the kingdom of the East Angles, where the locals made peaceand the viking-army was supplied with horses. Asser called the arrivals magnapaganorum classis (‘a great fleet of heathens’).21 The previous history of this fleetis obscure, although commments by Asser and (a century later) by Æthelweardprovide scope for speculation. Asser wrote that it came de Danubia, but modernhistorians have suggested that he had Denmark in mind.22 The Chronicle ofÆthelweard records for this year that classes tyranni Iguuares ab aquilone in terramAnglorum hiemaueruntque inter Orientales Anglos (‘the fleets of the tyrant Iguuararrived in the land of the English from the north, and they wintered among theEast Angles’).23 As vikings from Ireland were active in Pictland in 866 (under theleadership of Auisle [Ásl] and Amlaíb [Óláfr], companions of Ívarr), identificationwith them is a distinct possibility.24

This was not the first ‘great’ force to reach England: compare for example ‘The

20For discussion of the term ‘Danegeld’ and its distinction from tribute, see S. Keynes, ‘Heregeld(“army-tax”)’, in The Blackwell Encyclopaedia of Anglo-Saxon England, edd. Michael Lapidge et al.(Oxford 1999), p. 235.

21Annal 866: ASC.A, p. 47; ASC.B, s.a. <867>, p. 34; ASC.C, s.a. 867, p. 58; ASC.D, p. 24;ASC.E, p. 48; De rebus gestis Ælfredi, §21 (Asser’s Life, ed. Stevenson, p. 19). For general discussion,see R. Abels, ‘Alfred the Great, the micel hæðen here and the viking threat’, in Alfred the Great. Papersfrom the Eleventh-centenary Conferences, ed. Timothy Reuter (Aldershot 2003), pp. 265–79.

22Alfred the Great. Asser’s Life of King Alfred and Other Contemporary Sources, transl. SimonKeynes & M. Lapidge (Harmondsworth 1983), p. 238, n. 44.

23Chronicon Æthelweardi. The Chronicle of Æthelweard, ed. & transl. A. Campbell (Edinburgh1962), p. 35 (IV.1). As David Dumville has suggested to me, ‘usurper’ would be a better translationof tyranni than Campbell’s ‘tyrant’.

24The Annals of Ulster, I, edd. & transl. Mac Airt & Mac Niocaill, pp. 320/1 (866.1). PatrickWormald suggested further that the renewed interest of vikings in England in the mid-860scorresponded with declining opportunities for vikings in Francia due to the strategies employed byCharles the Bald: C.P. Wormald, ‘Viking studies: whence and whither?’, in The Vikings, ed.R.T. Farrell (Chichester 1982), pp. 128–53, at p. 137.


Anglo-Saxon Chronicle’ for the years 851–60 as well as Asser’s Life of King Alfred,§§4 and 6 (A.D. 851), 10 (A.D. 855), and 18 (A.D. 860).25 The Chronicler’susage seems to be to comment on the size of a large army on its first arrival orfollowing a recent division of troops into a larger and a smaller section. This isconsistent with other references to a ‘big army’ (micel here) in ‘The Anglo-SaxonChronicle’ and by Asser within the period of study.26 Modern historians’ use ofthe term as a designation for an army active for twenty-five years and in particularwith capitalisation and with translation of micel as ‘great’ (‘The Great Army’) doesnot therefore equate with that of contemporary sources.

We do get some clue to the trouble which the ‘big army’ of 865/6 will cause,for its provision with horses suggests intent to pursue a land-campaign. In 866/7(annal 867), the army went north to the kingdom of the Northumbrians. Thiscoïncided with a Northumbrian civil war which allowed these vikings to seizecontrol of York before the local rival parties could come to terms to fight theforeign enemy. The Northumbrian attempts at defence were ineffective. The rivalNorthumbrian kings Osberht and Ælle were killed (apparently in battle),27 andthe survivors made peace with the vikings. This had been a long way for the vikingsfrom East Anglia to travel, but they benefited greatly from the political disunityof the Northumbrians. It seems likely that these vikings were tipped off concerningthese events, which leads one to question where and how they might have receivednews of this: there were, of course, longstanding links between the two kingdoms,especially of an economic nature.28

In annal 868 ‘The Anglo-Saxon Chronicle’ and Asser’s Life of King Alfred tellus that the army then proceeded from Northumbria to the kingdom of theMercians (in winter 867) and in the following year (annal 869) returned to York

25Annals 851–60: ASC.A, pp. 43–6; ASC.B, pp. 30–3; ASC.C, pp. 54–7; ASC.D, pp. 21–3;ASC.E, pp. 46–7; ASC.F, pp. 64–6; Asser, De rebus gestis Ælfredi, §§4, 6, 10, 18 (Asser’s Life, ed.Stevenson, pp. 5, 6, 8, 17).

26For A.D. 875, 885, and 892, see ASC.A, pp. 49–50, 52–3, 55; ASC.B, pp. 36, 38, 40–1;ASC.C, pp. 60–1, 63–4, 66; ASC.D, pp. 26, 28–9, 30; ASC.E, pp. 50, 52, 53; ASC.F, pp. 71, 73,76–7; De rebus gestis Ælfredi, §58 (Asser’s Life, ed. Stevenson, p. 47).

27Annal 867: ASC.A, p. 47: ‘7 �þær was ungemetlic węl geslægen Norþanhymbra, sume binnan,sume butan, 7 �þa cyningas begen ofslægene’. Cf. ASC.B, s.a. <868>, p. 34; ASC.C, s.a. 868, p. 58;ASC.D, p. 24; ASC.E, p. 48; ASC.F, p. 67. Alfred Smyth has argued that Ælle was in fact killedafter the battle, but this opinion is derived from much later texts: Alfred P. Smyth, ScandinavianKings in the British Isles, 850–880 (Oxford 1977), pp. 189–94.

28Smyth has pointed out vikings’ good nose for civil war or political disunity and their abilityto profit from it: ibid., p. 182. It is unclear how long this particular civil war had been raging amongthe Northumbrians. If Æthelweard was correct about the northern origins of this viking-army, itsleaders may have already been aware of this war before they travelled to East Anglia; but then whydid these vikings target East Anglia first? This pattern of activity could be explained by an alliancebetween Dubliners from North Britain and vikings from Kent.


for a year. As York was a temporary base of operations, some measures may havebeen taken to secure its loyalty in their absence, whether by leaving some troopsthere or through the seizure of hostages or both. In Mercia we see the vikingsfollowing tactics similar to those employed at York, namely the seizure of a fortifiedsite which they could use to defend themselves against the natives: in this case,Nottingham was selected as a winter-base. The troops of the Mercian king Burhredand those of King Æthelred and his brother Alfred from Wessex (providing thehelp for which Burhred appealed) failed in their siege of that town, and theMercians were forced to make peace with the enemy. It was presumably inconsequence of that peace that the viking-army left Nottingham and returned toYork (annal 869). In the following annal we are told that the vikings of York rodeacross Mercia to East Anglia (winter 869). One wonders whether part of theagreement at Nottingham was to allow troop-movements by these vikings inMercian territory or whether the army was seeking to demonstrate its authorityby this act.

These vikings now focused their attention on the task of subduing the EastAngles. In 869/70 they wintered at Thetford and killed King Edmund on 20November, 869, following a decisive battle. ‘The Anglo-Saxon�Chronicle’ (annal870) tells us that in consequence these vikings conquered the kingdom (þæt landeall geeodon).29 The death of Edmund as a christian martyr, whose cult developedrapidly, led to elaborate narratives being written of his death. This phenomenonmay account for various additions made in later versions of ‘The Anglo-SaxonChronicle’ for this year.30 Æthelweard and Abbo of Fleury, writing at much thesame time as one another in the late tenth century, both identified Ívarr as theviking-leader responsible for the death of King Edmund. Abbo added that Ubbaremained in Northumbria.31 Dorothy Whitelock, in discussing the evidenceconcerning Edmund’s death, commented positively on the reliability of Abbo’sPassio Sancti Eadmundi despite much literary exaggeration within it.32 She didhowever see it as the very latest source on Edmund’s death which can be usefullyemployed for ninth-century history.

29Annal 870: ASC.A, p. 47; ASC.B, s.a. 871, p. 34; ASC.C, s.a. 871, p. 58; ASC.D, p. 24; ASC.E,p. 48; ASC.F, p. 67.

30ASC.F, s.a. 870, p. 67; ASC.E, s.a. 870, p. 48. Both these versions belong to the first half ofthe twelfth century. The E-text comments on the destruction of churches around Peterborough, andthe F-text names the leaders of the viking-army as Ingware and Ubba.

31Chronicon Æthelweardi, ed. & transl. Campbell, p. 36 (IV.2). Passio Sancti Eadmundi, §5:Memorials of St. Edmund’s Abbey, ed. Thomas Arnold (3 vols, London 1890–6), I.8–10; Three Livesof English Saints, ed. Michael Winterbottom (Toronto 1972), p. 72. Dorothy Whitelock, ‘Fact andfiction in the legend of St. Edmund’, Proceedings of The Suffolk Institute of Archaeology 31 (1967–9)217–33, at p. 219 (reprinted in her book, From Bede to Alfred, essay XI, with the same pagination).

32Ibid.


After the conquest of the East Angles, according to Irish record Ívarr returnedto Ireland via North Britain.33 This is not mentioned in ‘The Anglo-SaxonChronicle’ which reports the action of the main force, which now began tocampaign against the West Saxons.34 The record for 871 given in the Chronicleand by Asser (for whom the Chronicle was a principal source) highlights some ofthe difficulties in relying entirely on these accounts. There are inconsistencies andgaps in the record. ‘The Anglo-Saxon Chronicle’ reports nine battles and describessix of them.35 Asser recorded eight engagements and described four of them.36

No details of location, outcome, or anything else are provided for those battlesmerely enumerated. The Chronicler also qualified his enumeration by saying thatthese are engagements south of the Thames, not therefore including other eventswhich could have taken place north of that river. Other issues arise from closeranalysis of this text.

‘The Anglo-Saxon Chronicle’ for the year 871 relates that the viking-armycame to Reading (Berkshire, mostly north of the Thames but part of theWest-Saxon kingdom). After three days the army divided. Asser stated that a group stayed and built a rampart between the River Thames and the River Kennet(Reading is located on the southern bank of the Thames). Both Asser’s Life andthe Chronicle agree that two earls rode farther inland. According to Asser, theyhad the greater part of the army with them, but this seems incompatible with thesubsequent statement that nine earls fell in battle this year: two earls (eachcommanding his own force) were a small proportion of the armies’ totalcommand. Perhaps Asser’s account is exaggerated to highlight the achievement of Æthelwulf, ealdorman of Berkshire, who defeated them at Englefield(Berkshire). In this battle, ‘The Anglo-Saxon Chronicle’ reports, one of the earls,Sidroc, was killed.

z33 Cf. n. 11, above. Æthelweard stated that Ívarr died in the same year as Edmund’s martyrdom.This conflicts with the Irish chronicles which make clear that Ívarr died in 873: Annala RioghachtaEireann, Annals of the Kingdom of Ireland, by The Four Masters, from the Earliest Times to the Year1616, ed. & transl. John O’Donovan (2nd edn, 7 vols, Dublin 1856), I.518/19 (s.a. 871.10);Chronicum Scotorum. A Chronicle of Irish Affairs, from the Earliest Times to A.D. 1135; with aSupplement containing the Events from 1141 to 1150, ed. & transl. William M. Hennessy (London1866), pp. 164/5 (873.2); The Annals of Ulster, I, edd. & transl. Mac Airt & Mac Niocaill, pp. 328/9(873.3). Patrick Wormald (‘Viking studies’, p. 143) suggested that this could be a hagiographicalcommonplace. It could further be explained by Ívarr’s departure to North Britain in 870, leadingto an assumption that he died, which fits nicely with mediaeval perceptions of divine vengeance.

34One wonders whether there was a strategy in this sequence of conquests – were the vikingsseeking to isolate the West Saxons by conquering or subduing other English kingdoms first?

35Annal 871: ASC.A, pp. 48–9; ASC.B, s.a. <872>, pp. 34–5; ASC.C, s.a. 872, pp. 58–9; ASC.D,pp. 24–5; ASC.E, pp. 48–9; ASC.F, pp. 69–70.

36De rebus gestis Ælfredi, §§35–42 (Asser’s Life, ed. Stevenson, pp. 26–34).


The next battle took place four days later when Æthelred, king of the WestSaxons (865–71), and his brother Alfred challenged the vikings at Reading. Thistime, the vikings had the victory. Four days later, another fight took place, atAshdown (Berkshire), which shows that the viking-army had travelled somedistance to the west. The Chronicle emphasises that the whole viking-army waspresent at Ashdown (ealne þone here).37 This cannot mean the whole of theviking-army which arrived in East Anglia in late 865 – that is, it cannot if we agreethat Ívarr was a leader of the viking-army which after the killing of King Edmundset sail for North Britain where it appeared before Dumbarton in 870). He wouldpresumably have taken a force with him (and no doubt a haul of booty).38

Furthermore, there is the statement by Abbo of Fleury that one of the army’sleaders stayed in Northumbria when Ívarr’s troops travelled to East Anglia in theautumn of 869.39 What the chronicler seems to have meant was the wholeviking-army which had been at Reading, perhaps including the survivors ofEnglefield as well as the men who built the defensive works near the Thames.Again, one may allow for some exaggeration by the Chronicler(s), who might havesought to emphasise the English victory.

At Ashdown the viking-army fought in two divisions, one headed by kings(namely Bagsecg and Hálfdan) and one led by earls. Both were put to flight withheavy losses. Bagsecg and five earls are named among the fallen. At Basing(Hampshire), and more ambiguously at the unidentified Meretun, the vikingswon. At Meretun, the vikings fought in two divisions, although this may have beenin detail a formation different from that at Ashdown. What does seem clear fromannal 871 is that the army divided into parts for tactical reasons (as at Readingone active and one defensive, and likewise in the battles of Ashdown and Meretun);so it is possible at times for the Chronicler to speak of plural armies (þa hergas)perhaps representing part of a greater army (alne þone here).40 That parts of thearmy could be designated as hergas in their own right is important when one isinterpreting the Chronicle’s statements.41 The numerous references to a here inannals 865–92 can mean something less fixed than ‘the big/great army’ (as itappeared in annal 866) without additions or departures of troops.

37Annal 871: ASC.A, p. 48; ASC.B, p. 35 (s.a. <872>); ASC.C, p. 59 (s.a.872); ASC.D, p. 25;ASC.E, p. 49; ASC.F, pp. 69–70.

38The Annals of Ulster, I, edd. & transl. Mac Airt & Mac Niocaill, pp. 326/7 (870.6, 871.2);English captives were included in the prey taken back to Ireland.

39See n. 31, above.40Annals 871 and 878: ASC.A, pp. 48–51; ASC.B, pp. 34–5 (s.a. <872>) and 36–7 (s.a. <879>);

ASC.C, pp. 58–9 (s.a. 872) and 61–2 (s.a. 879); ASC.D, pp. 24–5 and 27; ASC.E, pp. 48–9 and50–1; ASC.F, pp. 69–70 and 71–2.

41Cf. N. Lund, ‘The armies of Swein Forkbeard and Cnut: leding or lið?’, Anglo-Saxon England15 (1986) 105–18, at pp. 110–12, and ‘The Danish perspective’, in The Battle of Maldon, AD 991,ed. Donald Scragg (Oxford 1991), pp. 114–42. For his full-scale study, see Niels Lund, Lið, Ledingog Landeværn. Hær og Samfund i Danmark i Ældre Middelalder (Roskilde 1996).


In annal 871, after the account of the battle of Meretun, another viking-armyappears on the scene. This is a ‘big summer-army’ (micel sumorlida; or anotherheathen army, alius paganorum exercitus, according to Asser).42 It arrived atReading.43 Asser wrote that it came de ultramarinis partibus (‘from overseas’), andit is therefore possible that it came to Reading via the Thames.44 Its leader(s)probably knew of the viking-army which had been based at Reading and set outto join it. Asser is clear that the summer-army joined the one which was alreadythere (societati se adiunxit),45 but the Chronicle does not explicitly say so. Assereither was working from an assumption (not in itself an unlikely state of affairs)or had other evidence to this effect.46 Thus when Alfred, now king of the WestSaxons, fought ‘the whole army’ (alne þone here) a little later in the year at Wilton(Wiltshire), this may indicate something other than alne þone here which he foughtat Ashdown. It could mean the united forces of the two armies mentioned in thisannal (or possibly the full force of one army). At Wilton, the vikings won. Theydo not seem to have followed up their victory with an attempt to conquer theWest Saxons. Instead they travelled east (from Reading) to establishwinter-quarters at London where the Mercians made peace with them.47 Perhapsthese vikings seized control of London at this point, for Alfred won the town backfrom Scandinavian control in the 880s.

According to annal 873, vikings went from London and over-wintered atTorksey in Lindsey. Here the Mercians made peace again. According to some, butnot all, versions of ‘The Anglo-Saxon Chronicle’, the army also went into

42Annal 871: as n. 35, above; Asser, De rebus gestis Ælfredi, §40 (Asser’s Life, ed. Stevenson, p. 31).On this episode, cf. Smyth, Scandinavian Kings, pp. 240–54. On the word and its appearanceelsewhere, see D.N. Dumville, ‘The Chronicle of the Kings of Alba’, in Kings, Clerics and Chroniclesin Scotland, 500–1297. Essays in Honour of Marjorie Ogilvie Anderson on the Occasion of her NinetiethBirthday, ed. Simon Taylor (Dublin 2000), pp. 73–86, at p. 81. On its history in Scottish Gaelic asa personal name, see the significant essay by Hermann Pálsson, ‘The name Somhairle and its clan’,in So Meny People Longages and Tonges. Philological Essays in Scots and Mediaeval English presented toAngus McIntosh, edd. Michael Benskin & M.L. Samuels (Edinburgh 1981), pp. 167–72.

43According to ‘The Anglo-Saxon Chronicle’ (B-, C-, D-, and E-texts), but not the A-text.44De rebus gestis Ælfredi, §40 (Asser’s Life, ed. Stevenson, p. 31).45 Ibid.46Alfred, transl. Keynes & Lapidge, p. 243, n. 73; Smyth, Scandinavian Kings, p. 243. The armies

seem to have united by 875. The viking-leaders at Repton are named as Hálfdan (presumably thesame who fought at Ashdown) and three additional kings (not named in the kings’ contingent atAshdown), Guthrum, Oscetel, and Anwend.

47Cf. Smyth, Scandinavian Kings, p. 241.


Northumbria before going to Torksey.48 Dorothy Whitelock suggested that theincentive for vikings to campaign in Northumbria in late 872 was to suppress arevolt against an English dependent king who had been established there by thevikings.49 This idea is based on the testimony of Historia regum Anglorum, Part II,written in the first third of the twelfth century, as well as other twelfth-centurywritings emanating from Durham and the early thirteenth-century FloresHistoriarum of Roger of Wendover.50 These sources assert that Northumbria (atleast north of the Tyne) was ruled by English kings under viking-suzerainty from867 until the reign of Guthred. In 872, they assert, the first of these kings,Ecgberht, was driven out by the Northumbrians. This would provide a motive forthe vikings’ campaign. However, these texts are unacceptably late, and the originsof their reports are uncertain.51

The motive for these vikings’ Northern campaign may rather have been tosubdue various parts of the Mercian kingdom following the capture of London.If so, they achieved their goal the following year (annal 874), when the armymoved from Lindsey to Repton (Derbyshire); the Mercian king, Burhred, fledoverseas and died soon after.52 The vikings then ‘conquered the whole kingdom’(þæt land eall geeodon)53 and established Ceolwulf, a king’s thegn, as a dependent

48Annal 873: ASC.A, p. 49; ASC.B, s.a. <874>, p. 35; ASC.C, s.a. 874, p. 60. Even though theauthor of text D had access to (a relative of ) text C, ASC.D, s.a. 873, p. 26, does not have this. It isinteresting that this information does not seem to have belonged to a version (‘The NorthernRecension’, the ultimate common source of DEF) written at York within a generation of A.D. 1000;but the implications of the textual history remain unclear.

49The Anglo-Saxon Chronicle. A Revised Translation, transl. Dorothy Whitelock et al. (London1961; rev. imp., 1965), p. 47, n. 14.

50(1) Symeonis Monachi Opera Omnia, ed. Thomas Arnold (2 vols, London 1882/5), I.55 and225, II.110; The Church Historians of England, transl. Joseph Stevenson (5 vols in 9, London 1853–8), III.2, p. 78. (2) Rogeri de Wendover Chronica sive Flores Historiarum, ed. Henry O. Coxe (5 vols,London 1841–4), I.323–4 (s.a. 872); Roger of Wendover’s Flowers of History – comprising the Historyof England from the Descent of the Saxons to A.D. 1235 – formerly ascribed to Matthew Paris, transl.J.A. Giles (2 vols, London 1849), I.206. Cf. David Rollason et al., Sources for York History to A.D.1100 (York 1998), pp. 25–7, 32, and 63. For an exposition of this theory, see Peter Sawyer,Scandinavians and the English in the Viking Age (Cambridge 1995), pp. 8–10.

51David N. Dumville, ‘Textual archaeology and Northumbrian history subsequent to Bede’, inCoinage in Ninth-century Northumbria, ed. D.M. Metcalf (Oxford 1987), pp. 43–55, at p. 45(reprinted in his book, Britons and Anglo-Saxons in the Early Middle Ages [Aldershot 1993], essay X,with the same pagination).

52S. Keynes, ‘Anglo-Saxon entries in the “Liber Vitae” of Brescia’, in Alfred the Wise. Studies inHonour of Janet Bately on the Occasion of her Sixty-fifth Birthday, edd. Jane Roberts et al. (Cambridge1997), pp. 99–119, at p. 110.

53Annal 874: ASC.A, p. 49; ASC.B, s.a. <875>, pp. 35–6; ASC.C, s.a. 875, p. 60; ASC.D, p. 26;ASC.E, pp. 49–50.


king who swore to act under their direction.54 This event fully secured theland-route between Northumbria and East Anglia both of those kingdoms hadalready fallen under viking-rule, and the Mercian campaign may have beendeliberately intended for that purpose.

Much light has been shed on the vikings’ activities at Repton by the excavationssupervised by Martin Biddle and Birthe Kjølbye-Biddle. They have demonstratedthat a D-shaped earthen defensive enclosure was constructed there, incorporatingan earlier church (which was badly damaged by vikings). The site seems to havebeen linked to Mercian royalty and to have been deliberately appropriated as asymbol of the vikings’ newly-won authority. A mass-grave at Repton may be datedto 873 and seems to contain victims of war or plague among the viking-army.55

The Biddles have tentatively identified the high-status central burial in thismass-grave as that of the viking-leader Ívarr, whose death is reported in Irishchronicles for 873.56 They have sought to corroborate this by reference to anaccount in the mid-thirteenth-century Old-Scandinavian text Ragnars sagaloðbrókar which states that Ívarr died (and was buried) somewhere in England.57

The latter may be discounted, as it was written at a time and place far removedfrom events. As to the burial at Repton, it is worth keeping in mind that, apartfrom the kings named as belonging to the viking-army, there were numerous jarls(earls); several of the latter are named among the fallen at Ashdown (871) andlater at the battle of Edington (878).58 Jarls are mentioned often enough innear-contemporary sources to suggest that they were high-status military leaders

54Ceolwulf is not named in A. On his reign, see S. Keynes, ‘King Alfred and the Mercians’, inKings, Currency and Alliances. History and Coinage of Southern England in the Ninth Century, edd.Mark A.S. Blackburn & D.N. Dumville (Woodbridge 1998), pp. 1–45, at 12–19 and 35. For thededuction that Ceolwulf’s Mercians were responsible for the killing of King Rhodri Mawr in 878under these vikings’ direction, see D.N. Dumville, ‘Brittany and “Armes Prydein Vawr’”, Étudesceltiques 20 (1983) 145–58, at p. 157 (reprinted in his book Britons and Anglo-Saxons, essay XVI,with the same pagination), with full reference to the Celtic sources; see also D.N. Dumville, ‘Whatis mediaeval Gaelic poetry?’, in Explorations in Cultural History: Essays for Peter Gabriel McCaffery,edd. David F. Smith & H. Philsooph (Aberdeen 2010), pp. 81–153, at 133–5. Rhodri had previouslykilled Horm, ‘chief ’ (toesech) of the Dubgaill, in 856: The Annals of Ulster, I, edd. & transl. MacAirt & Mac Niocaill, pp. 314/15 (856.6).

55M. Biddle & B. Kjølbye-Biddle, ‘Repton and the “great heathen army”, 873–4’, in Vikingsand the Danelaw. Select Papers from the Proceedings of the Thirteenth Viking Congress, Nottingham andYork, 21–30 August, 1997, edd. James Graham-Campbell et al. (Oxford 2001), pp. 45–96.

56Ibid., pp. 81–4.57Ragnars saga loðbrókar, in Fornaldar sögur Norðurlanda ed. Guðni Jónsson (4 vols, Reykjavík

1954), I.219–85, at I.280; The Saga of the Volsungs, the Saga of Ragnar Lothbrok, together with theLay of Kraka, transl. Margaret Schlauch (2nd edn, London 1949), pp. 251–2.

58Annals 871 and 878: ASC.A, pp. 48–9, 50–1; ASC.B, pp. 34–5 (s.a. <872>), 36–7 (s.a. <879>);ASC.C, pp. 58–9 (s.a. 872), 61–2 (s.a. 879); ASC.D, pp. 24–5, 27; ASC.E, pp. 48–9, 50–1.


who might have been buried with the same honour shown at the central burial inRepton. It is therefore doubtful whether Ívarr’s bones have been recovered atRepton or whether they are those of another leader. Ívarr is last mentioned inIreland, and the Chronicle of Æthelweard implies that he was absent from Englandafter 870.59 Sadly, ‘The Anglo-Saxon Chronicle’ reveals little detail of whathappened at Repton.

Having won Mercia, the viking-army divided, according to the Chronicle’sannal 875. Three kings – Guthrum, Oscetel, and Anwend – went with a big armyto Cambridge (mid micle here).60 Simon Keynes and Michael Lapidge havesuggested that this may represent the micel sumorlida which had arrived at Readingin 871.61 An alternative possibility is that at least one of these rulers had beenpromoted to the status of king following the death of King Bagsecg at Ashdownin 871. Hálfdan travelled with another section of the viking-army (sumum þamhere) to Northumbria, taking winter-quarters on Tyneside, and the army (se here)conquered that kingdom (þæt lond geeode).

Hálfdan soon drops from ‘The Anglo-Saxon Chronicle’, since he would nolonger belong to ‘the main narrative thread’. Annal 875 reports that, afterover-wintering on the Tyne, he made war on the Picts and Strathclyde Britons.62

Hálfdan can be identified as a kinsman of Ívarr (annal 878 refers to a brother ofHálfdan and Ívarr), and his battles in the north can be seen as the continuationof Ívarr’s policies there as recorded in Irish chronicles.63 These northern campaignsallow Hálfdan to be equated with the Hálfdan mentioned in Irish sources. ‘TheAnnals of Ulster’ report a great battle between Picts and Dubgennti in 875.64 Thismay be the battle at Dollar (Clackmannanshire) described in ‘The Chronicle of

59Æthelweard, Chronicon, IV.2 (ed. & transl. Campbell, p. 36).60Annal 875: ASC.A, pp. 49–50; ASC.B, p. 36 (s.a. <876>); ASC.C, pp. 60–1 (s.a. 876); ASC.D,

p. 26; ASC.E, p. 50.61Alfred, transl. Keynes & Lapidge, p. 244(–5), n. 84.62This is the first mention of ‘the Britons of Strathclyde’ (*Strætcledwalas) in English sources.

Numismatic evidence relating to the activities of Hálfdan’s troops has been discussed byJ. Graham-Campbell, ‘The northern hoards from Cuerdale to Bossall/Flaxton’, in Edward the Elder,899–924, edd. N.J. Higham & D.H. Hill (London 2001), pp. 212–29, at p. 226.

63Clare Downham, Viking Kings of Britain and Ireland. The Dynasty of Ívarr to A.D. 1014(Edinburgh 2007; 2nd edn, 2008), chapter 5. See above, p. 11~16.

64The Annals of Ulster, I, edd. & transl. Mac Airt & Mac Niocaill, pp. 330/1 (875.3). OnDubgennti, see A.P. Smyth, ‘The Black Foreigners of York and the White Foreigners of Dublin’,Saga-book 19 (1974–7) 101–17, and David N. Dumville, ‘Old Dubliners and New Dubliners inIreland and Britain: a Viking-Age story’, Medieval Dublin 6 (2004) 79–94, reprinted in his CelticEssays, 2001–2007 (2 vols, Aberdeen 2007), I.103–22. See further, chapter 6, below, and C.Downham, ‘Viking identities in Ireland: it’s not all black and white’, Medieval Dublin 11 (2009)185~201.


–

–

the Kings of Alba’ for the same year.65 According to ‘The Annals of Ulster’ Hálfdanalso deceitfully killed Oistin son of Amlaíb in 875.66 This suggests that rivalryexisted between the lineages of Amlaíb (Óláfr) and Ívarr, who had campaignedtogether in North Britain a decade before.

‘The Anglo-Saxon Chronicle’ states in annal 876 that Hálfdan shared out landin Northumbria among his followers. This may have been intended to settle affairsbefore Hálfdan went to campaign overseas. ‘The Annals of Ulster’ demonstratethat he travelled to Ireland, perhaps seeking to win the position there which Ívarrhad once held. He was however killed in 877 in a battle against Finngennti atStrangford Lough on the northeastern coast of Ireland.67 Hálfdan’s title in Irishrecords is consistently that of chieftain (toísech), not king, which perhaps suggeststhat he never received the same level of recognition among the vikings in Irelandas he had held in England.

Apart from record of the activities of Hálfdan in annal 875, ‘The Anglo-SaxonChronicle’ reports that Alfred defeated seven viking-ships in battle, one of whichhe captured. The origins of this battle and its location are not given. The accountseems intended to glorify Alfred more than to shed light on the events of this year.It could suggest that other viking-groups were at large or that naval scouting partieswere sent from the viking-army under the command of Anwend, Oscetel, andGuthrum whose journey to Cambridge (in the Mercian kingdom) is reported inthe same annal. The Chronicle states that this army stayed at Cambridge for ayear, which indicates that the report was written at least one year retrospectively.

The narrative focus of ‘The Anglo-Saxon Chronicle’ after annal 875 switchesto this army which had travelled to Cambridge. That is because it soon becomesa menace to the West Saxons. In annal 876 this southern viking-army (Asser madequite explicit its origin at Cambridge) went west to Wareham (Dorset).68 AtWareham, Alfred made peace with these vikings, who promised that they wouldrapidly leave his kingdom. However, in that same year ‘the mounted army’ (se

65B.T. Hudson (ed. & transl.), ‘The Scottish Chronicle’, Scottish Historical Review 77 (1998)129–61, at pp. 148, 154. See also the discussions by M. Miller, ‘Amlaíb trahens centum’, ScottishGaelic Studies 19 (1999) 241–5, and Downham, Viking Kings, pp. 142–5.

66The Annals of Ulster, I, edd. & transl. Mac Airt & Mac Niocaill, pp. 330/1 (875.4).67Ibid., I.332/3 (877.5); Chronicum Scotorum, ed. & transl. Hennessy, pp. 166/7 (877.3);

Annala, ed. & transl. O’Donovan, I.520/1 (874.12). (On Finngenti, cf. Dubgennti in n. 64, above.)On Æthelweard’s mistaken attribution to Hálfdan of activity in southwestern England in 878, seeAlfred, transl. Keynes & Lapidge, p. 248, n. 99.

68It is not clear who led this army. Oscytel and Anwend are not mentioned again in the CommonStock of ‘The Anglo-Saxon Chronicle’; whether they remained in East Anglia, went abroad, werekilled, or simply lost their power is uncertain. Guthrum reäppears in the account of the invasion ofthe kingdom of the West Saxons in 878.


gehorsoda here) travelled to Exeter (Devon).69 Record of this event seems to berepeated in annal 877: ‘In this year the enemy-army from Wareham came toExeter; and the naval force (se sciphere) sailed west along the coast’.70 This can beexplained as follows: the horse-troop set out at the end of the annalistic year 876(August or the beginning of autumn) and did not arrive in Exeter until the startof the following annal-year 877, which by our reckoning is autumn 876. However,I think it more likely that branches of the army travelled separately: such a mannerof progressing from one place to another is well attested in Continental reports ofviking-activities. For example, ‘The Annals of Saint-Vaast’ report for 883 that ‘theNorthmen ... approached Leuven with foot-soldiers and mounted soldiers and alltheir chattels. Their ships also entered the River Somme from the sea’ (Nortmanni... Latuerum cum equitibus et peditibus atque omni supellectili ueniunt. Naues quoqueper mare Sumnam fluuium ingressae ...). Other examples are provided in thatchronicle’s annals for 885 and 891.71 Horse-troop, infantry, and fleet may havetravelled separately to Exeter, each at its own pace.72 The Chronicle asserts that120 of the viking-ships travelling to Exeter were destroyed in a storm at Swanage(Dorset). Perhaps because of these losses at sea, and perhaps being short of supplies,the horse-troop, which had already taken the fortress of Exeter, decided to makepeace with Alfred; they gave hostages and swore great oaths. Thereafter they leftWessex for Mercia whose lands were shared between the army and King CeolwulfII.

The following events, of 878, are well known. In early January a viking-army(presumably crossing the border from Mercia)73 launched a surprise-attack on theroyal estate of Chippenham (Wiltshire).74 It proceeded to bring the West-Saxonkingdom under its control (geridon Wesseaxna lond 7 gesæton). The arrival of a

69Annal 876: ASC.A, p. 50; ASC.B, s.a. <877>, p. 36; ASC.C, s.a. 877, p. 61; ASC.D, p. 26;ASC.E, p. 50; ASC.F, p. 71; De rebus gestis Ælfredi, §49 (Asser’s Life, ed. Stevenson, p. 37). Asser’sLife, which is textually corrupt here, may be interpreted as saying that the vikings’ horse-troop wentwest at this point: Alfred, transl. Keynes & Lapidge, p. 246, n. 91.

70Annal 877: ASC.A, p. 50; ASC.B, s.a. <878>, p. 36; ASC.C, s.a. 878, p. 61; ASC.D, pp. 26–7; ASC.E, p. 50.

71Annales Vedastini, s.aa. 883, 885, 891: Monumenta Germaniae Historica, Scriptores, I, ed. GeorgHeinrich Pertz (Hannover 1826), pp. 516–31 (annals 877–900 only), at 521, 522–3, 526–7; LesAnnales de Saint-Bertin et de Saint-Vaast suivies de fragments d’une chronique inédite, ed. C. Dehaisnes(Paris 1871), pp. 293–360, at 316–17, 320–3, 337–41; Annales Xantenses et Annales Vedastini, ed.B. von Simson (Hannover 1909), pp. 53–4, 56–9, 69–70.

72Stenton, Anglo-Saxon England, p. 251.73Asser, lacking annal 877 of ‘The Anglo-Saxon Chronicle’, brought the army which had been

at Exeter in 876 directly thence to the attack on Chippenham in January 878: for thoroughdiscussion, see Alfred, transl. Keynes & Lapidge, pp. 246–7.

74Charles Plummer suggested that the attack was designed to capture Alfred who might havebeen staying there over the winter: Two of the Saxon Chronicles, II.92; cf. Smyth, Scandinavian Kings,p. 247.


23-ship fleet in Devon in the same winter, led by a brother of Ívarr and Hálfdan,seems to have been deliberately coördinated to help the Chippenham-army.75 Asserwrote that the fleet had overwintered in Dyfed; he (as a native of southwesternWales) is to be taken as a good authority on this matter. This indicates a continuinglink of the family of Hálfdan and Ívarr with the viking-armies in England. Asserprovided a detailed account of the defeat of this fleet and the death of its leader atCountisbury, a fort in north Devon. Asser stated that 1,200 fell there, but ‘TheAnglo-Saxon Chronicle’ provides a more conservative estimate of 840 or 860,depending on which version one reads.76

Alfred’s great victory at Edington (Wiltshire) which followed at Easter 878seems to have successfully stemmed the tide of viking-conquests.77 Thepeace-agreement which followed included the baptism of King Guthrum andthirty of his leading men. Good relations may have endured between the twoparties. This is suggested by the Chronicle’s respectful manner in its record ofGuthrum’s death in annal 890 (A.D. 889/90): ‘He was Alfred’s godson, and helived in East Anglia and was the first [viking] to settle that kingdom’ (wæs �Ęlfredescyninges godsunu, 7 he bude on Eastenglum 7 �þæt lond ærest gesæt).78 Guthrum’sarmy travelled to Cirencester (Gloucestershire) in Mercia and stayed there for ayear (annal 879) before moving to East Anglia where it settled (annal 880).79 Fromthis point the narrative’s focus switches to a new force of vikings which hadassembled at Fulham in 879 (gegadrode an hloþ�wicenga 7 gesæt æt Fullanhammebe Temese) and which later came to threaten Wessex. This change of focus mayalso be indicative of retrospective rather than contemporaneous composition.

75Geoffrey Gaimar, writing in the twelfth century, identified this brother as Ubba. However, hissource is unknown, and he may have jumped to this conclusion on reading of Ubba’s associationwith Ívarr in the legends of St Edmund’s martyrdom. This identification therefore cannot beregarded as reliable. For the text of Gaimar’s verse chronicle, see Lestorie des Engles solum la translacionMaistre Geffrei Gaimar, edd. & transl. Thomas Duffus Hardy & C.T. Martin (2 vols, London1888/9), I.132, II.101 (lines 3147–9); L’Estoire des Engleis, by Geffrei Gaimar, ed. Alexander Bell(Oxford 1960), pp. 91–2 (lines 2838–92). See further Alfred, transl. Keynes & Lapidge, pp. 238–9(n. 44).

76Annal 878: De rebus gestis Ælfredi, §54 (Asser’s Life, ed. Stevenson, pp. 43–4; Alfred, transl.Keynes & Lapidge, p. 84); ASC.A, pp. 50–1; ASC.B, s.a. <879>, pp. 36–7; ASC.C, s.a. 879, pp. 61–2; ASC.D, p. 27; ASC.E, pp. 50–1; ASC.F, pp. 71–2.

77Whitelock, ‘The importance’.78Annal 890: ASC.A, p. 54. Cf. ASC.B, s.a. <891>, p. 40; ASC.C, s.a. 891, p. 65; ASC.D, p. 30;

ASC.E, p. 53; ASC.F, p. 75.79For discussion, see Dumville, Wesssex, pp. 1–23. On annal 880, see n. 83, below.


Ominously, the Chronicle reports an eclipse in the same annal (879) in whichthe assembly of the army at Fulham is described.80 Asser stated that this ‘bigheathen army’ (magnus paganorum exercitus) came from overseas and united withthe vikings farther up the River Thames (adiunctus est superiori exercitus).81 Theythen overwintered at Fulham. As the army at Cirencester was located by a tributaryof the Thames (the River Churn), it may be that this latter was the other armyreferred to; yet it is clear that the Cirencester-army moved to East Anglia thefollowing year, while the Fulham-army went elsewhere. It may be either that onlypart of the army returning from their defeat at Edington decided to try its luckwith the newcomers or that any link between the two armies was short-lived.82

According to annal 880, the army from Fulham left for the Continent (for ...ofer sę�... on Froncland to Gend).83 Their travels for the next twelve years arerecorded in a skeletal fashion in ‘The Anglo-Saxon Chronicle’. These can bechecked against ‘The Annals of Saint-Vaast’ which show the English account tobe largely accurate. ‘The Anglo-Saxon Chronicle’ merely records their winter-basesand major events such as their defeat at Saucourt in 881. David Dumville hassuggested that this account was written in retrospect, and there is some evidencefor this in annal 887.84 This records the bases of the Continental army for twowinters, 886/7 and 887/8. As the Chronicle-year begins and ends in September,part of this record properly belongs to annal 888 (in which, however, there is norecord of these vikings’ activities).

However, some difficulties arise from a comparison of ‘The Anglo-SaxonChronicle’ and ‘The Annals of Saint-Vaast’. The first is in determining the dateat which the army arrived on the Continent from England. ‘The Annals ofSaint-Vaast’ report the devastation of Thérouanne by these vikings in July 879,but ‘The Anglo-Saxon Chronicle’ does not report their departure from England

80Annal 879: ASC.A, p. 51; ASC.B, s.a. <880>, p. 37; ASC.C, s.a. 880, p. 62; ASC.D, p. 27;ASC.E, p. 51; ASC.F, p. 72. A.P. Smyth, ‘The solar eclipse of Wednesday 29 October AD 878:ninth-century historical records and the findings of modern astronomy’, in Alfred the Wise, edd.Roberts et al., pp. 187–210, has provided a very full consideration of the significance of the recordof the eclipse in this annal.

81De rebus gestis Ælfredi, §58 (Asser’s Life, ed. Stevenson, p. 47; Alfred, transl. Keynes & Lapidge, p. 85).82Dumville, Wessex, p. 6, n. 32, has cast some doubt on the reliability of Asser’s report of the

events of 879. Asser was almost certainly not a first-hand witness to the events of these years, andDumville has suggested that he obtained or deduced this information when he was using ‘TheAnglo-Saxon Chronicle’ as a source for his own work.

83Annal 880: ASC.A, p. 51. Cf. ASC.B, s.a. <881>, p. 37; ASC.C, s.a. 881, p. 62; ASC.D, p. 27;ASC.E, p. 51; ASC.F, p. 72.

84Dumville, Wessex, p. 89. The Chronicle (annal 887) reads up andlang Sigene oþ� Mæterne oþ�Cariei, 7 �þa sæton �þara 7 innan Iona tu winter on þam twam stedum: ASC.A, p. 53. Cf. ASC.B,s.a. <888>, p. 39; ASC.C, s.a. 888, p. 64; ASC.D, p. 29; ASC.E, p. 52; ASC.F, p. 74. These were thewinters of 886/7 and 887/8. The story of the army’s travels resumes in annal 890 (starting in theautumn of 889).


until annal 880 (whose scope is from September 879). Another inconsistency isthat ‘The Anglo-Saxon Chronicle’ (annal 883) says that the vikings stayed oneyear at Condé, whereas ‘The Annals of Saint-Vaast’ report only a half-year stay,from October 882 until Spring 883. ‘The Annals of Saint-Vaast’ were written atSaint-Vaast in Arras from 873 to 900 and may be regarded as the more reliablewitness to events.85 These differences are therefore not easy to reconcile. Mostproblematic is the record of when these vikings left England, as the apparent errorin ‘The Anglo-Saxon Chronicle’ cannot be explained by geographical distancefrom the event being reported. This discrepancy may provide another clue toretrospective composition in ‘The Anglo-Saxon Chronicle’.

While the Continental activities of these vikings are described in ‘TheAnglo-Saxon Chronicle’, annals 881–91, some reference is also made to vikingsin England. According to annal 882, Alfred waged war against four viking-shipsof which two were captured and the others surrendered. As with the record ofAlfred’s earlier naval engagement, in annal 875, no details are given of location ofbattle or provenance of the viking-ships. A somewhat dramatised version of thelater skirmish was given by Asser.86 The record seems in both cases to have beenincluded to enhance the reputation of King Alfred.

In annal 883 an English victory is reported against a viking-army encampedat London. This is found in versions B, C, D, and E of ‘The Anglo-SaxonChronicle’, but not in the A-text, Asser’s Life of King Alfred, or the Chronicle ofÆthelweard.87 The restoration of the city by Alfred is not reported until annal886,88 and this has caused some confusion among historians as to whether therecord in annal 883 is misplaced.89 David Dumville and Simon Keynes haveconclusively demonstrated that there is no reason to mistrust ‘The Anglo-SaxonChronicle’ at this juncture.90 It is possible, for various reasons, that the restorationof the city followed this English victory by a few years.

Towards the end of 884 the Fulham-army of vikings on the Continent cameback into contact with England. The main army divided at Boulogne, with onesection travelling to Leuven (seemingly under the leadership of one Guðrøðr who

85Janet L. Nelson, Charles the Bald (Harlow 1992), p. 270.86De rebus gestis Ælfredi, §64: Asser’s Life, ed. Stevenson, p. 49; Alfred, transl. Keynes & Lapidge,

p. 86.87Annal 883: ASC.B, s.a. <884>, p. 38; ASC.C, s.a. 884, p. 63; ASC.D, p. 28; ASC.E, p. 51. It

is clear from ASC.FLat, s.a. 883, p. 73, that its author had the Old English to hand, even though itis not reproduced in his text; but he remodelled its content in his Latin version.

88Annal 886: ASC.A, p. 53; ASC.B, s.a. <887>, p. 39; ASC.C, s.a. 887, p. 64; ASC.D, p. 29;ASC.E, p. 52; ASC.F, pp. 73–4; Asser, De rebus gestis Ælfredi, §83 (Alfred’s Life, ed. Stevenson, p. 69),provided a context of conflict, as did Æthelweard, Chronicon, IV.3 (ed. & transl. Campbell, p. 46).

89Alfred, transl. Keynes & Lapidge, p. 266, n. 198. Cf. Keynes, ‘King Alfred and the Mercians’,especially p. 22 and n. 96.

90Dumville, Wessex, pp. 6–7; Keynes, ‘King Alfred and the Mercians’, pp. 12–23.


was later killed), the other crossing to England.91 ‘The Anglo-Saxon Chronicle’(annal 885) reports the arrival of this army at Rochester (West Kent) where itbesieged the city.92 When Alfred arrived with a relief-force, the vikings departed.It seems that there are then some sentences missing from the Chronicle which arefound in Æthelweard’s Chronicle. This was originally argued by Stenton, and hisviews have been followed by Dorothy Whitelock, Alistair Campbell, and others.93

The hypothesis is based on the repetition of a series of words at the end of twoconsecutive sentences in Æthelweard’s Chronicle, where all surviving copies of‘The Anglo-Saxon Chronicle’ and Asser’s Life of King Alfred preserve theinformation of the first sentence only. The so-called ‘missing sentence’ has thevirtue of explaining Alfred’s attack on East Anglia. According to Æthelweard, onlypart of the viking-force left Rochester and went overseas; those who remainedrenewed (renouant) their exchange of hostages with the English but neverthelessundertook two raids south of the Thames.94 The vikings of East Anglia sent a forceto Benfleet (Essex) in support; but these allied forces quarrelled, and a troopdecided to go overseas. The word renouant needs to be probed. Perhaps more thanone agreement was made in the course of the year, or perhaps the renewal refersback to an agreement made with the vikings who left Fulham in 879.

After this event, ‘The Anglo-Saxon Chronicle’ and Asser’s Life of King Alfredagree that Alfred took a fleet to the mouth of the River Stour (the border betweenEssex – in Alfred’s kingdom – and East Anglia)95 where he defeated a small, butnot insignificant, force of viking-ships (sixteen according to the Chronicle andÆthelweard, thirteen according to Asser). However, a big naval force of pirates(mycelne sciphere wicenga) encountered the English on return from their victoryand defeated them. ‘The Anglo-Saxon Chronicle’ says that the second sea-battlewas fought on the same day, but it does not identify the provenance of the fleet.Asser wrote that ‘the East Angles’ assembled a fleet from all areas: the implication

91Annales Vedastini, s.a. 884: ed. Pertz, pp. 521–2; ed. Dehaisnes, pp. 318–20; ed. von Simson,pp. 54–6.

92Annal 885: ASC.A, pp. 52–3; ASC.B, s.a. <886>, pp. 38–9; ASC.C, s.a. 886, pp. 63–4; ASC.D,pp. 28–9; ASC.E, p. 52; cf. ASC.F, p. 73. Æthelweard, Chronicon, IV.3 (ed. & transl. Campbell,p. 45); De rebus gestis Ælfredi, §67 (Asser’s Life, ed. Stevenson, p. 51; Alfred, transl. Keynes & Lapidge,p. 87). Asser’s number may be explained by minim-confusion, .xiii. and .xui..

93F.M. Stenton, ‘The south-western element in the Old English Chronicle’, in Essays in MedievalHistory presented to Thomas Frederick Tout, edd. A.G. Little & F.M. Powicke (Manchester 1925),pp. 15–24, at 20–1, reprinted in his book Preparatory to ‘Anglo-Saxon England’ (Oxford 1970),pp. 106–15, at 111–12. For comment, see Chronicon Æthelweardi, ed. & transl. Campbell, p. xxvii;English Historical Documents, c. 500–1042, transl. Dorothy Whitelock (2nd edn, London 1979),p. 198, n. 1; Alfred, transl. Keynes & Lapidge, p. 252, n. 125.

94Æthelweard, Chronicon, IV.3 (ed. & transl. Campbell, p. 44).95Dumville, Wessex, pp. xiv and 8; cf. A. Williams, ‘The vikings in Essex, 871–917’, Essex

Archaeology and History 27 (1996) 92–101.


is that they acted in reäction to the first battle.96 If this fleet was indeed assembledwithin a day, it suggests that fleets were on standby at different ports, ready toassemble in times of need. Asser must have had access to another source (beyondthe Chronicle) for these events, perhaps consequent on his presence at Alfred’scourt from 885. His work provides much additional detail on these events. ‘TheAnglo-Saxon Chronicle’ goes on to say that in the same year the army in EastAnglia violated its peace with Alfred. If we accept the evidence of Æthelweard andAsser, this peace must have been made after the last-mentioned encounter. If werely on ‘The Anglo-Saxon Chronicle’ alone, it could be referring back to the treatymade between Alfred and Guthrum in 878.97

In 886, Alfred occupied/set in order (gesette) London, and all the English whowere not subject to Scandinavians submitted to him.98 Alfred entrusted the townto his brother-in-law, ‘Ealdorman’ Æthelred, as someone who could rally thesupport of the Mercians and to create an ally who could maintain the historicallyMercian city as a bulwark against the vikings to the east.99 Asser referred to arestoration of the town by Alfred who made it ‘habitable again’, presumably afterthe ravages by the English in its recapture in 883.100

In 892 the army whose Continental travels the Chronicler had traced sinceannal 880 came to England. It is possible that the preceding accounts were writtenretrospectively in consequence of this event. Indeed, it is tempting to suggest thatall entries from 865 to 892 were written up at the same time. This is because ofthe stylistic unity of this section and some hints of retrospective composition –notably the significance given to Alfred (868, 871) before his reign began, andthe non-contemporaneous record in annal 887, which refers to viking-activitiesin the following year.

The departure of this fleet from the Continent is described in ‘TheAnglo-Saxon Chronicle’ and ‘The Annals of Saint-Vaast’.101 According to the

96De rebus gestis Ælfredi, §67: Asser’s Life, ed. Stevenson, p. 51; Alfred, transl. Keynes & Lapidge,p. 87.

97Dumville, Wessex, pp. 1–23; Keynes, ‘King Alfred and the Mercians’, pp. 31–3.98Annal 886: ASC.A, p. 53; ASC.B, s.a. <887>, p. 39; ASC.C, s.a. 887, p. 64; ASC.D, p. 29;

ASC.E, p. 52; ASC.F, pp. 73–4. On gesettan, see Two of the Saxon Chronicles, ed. Plummer, II.99(cf. I.348), and Dumville, Wessex, p. 7, n. 35.

99Alfred, transl. Keynes & Lapidge, p. 266, n. 198; Keynes, ‘King Alfred and the Mercians’,p. 24. On Æthelred’s title, see Dumville, Wessex, pp. 1–27, passim.

100De rebus gestis Ælfredi, §83: Asser’s Life, ed. Stevenson, p. 69; Alfred, transl. Keynes & Lapidge,p. 98.



English source (annal 892),102 it proceeded from Boulogne to Lympne (East Kent)‘with horses and all’, and the size of the fleet was 200 ships (or 250 according toText A). On its arrival the army built a fortress at Appledore near Lympne. Soonafter, another and smaller fleet – of eighty ships led by Hæsten – came into theThames. He established a fortress at Milton Regis (also in Kent, but just south ofthe Isle of Sheppey). The Chronicle gives no clue as to Hæsten’s provenance.However, he is mentioned in ‘The Annals of Saint-Vaast’ as based in the SommeValley in 890 and at Amiens in the following year.103 ‘The Annals of Saint-Vaast’also provide some explanation for the departure of the viking-fleets from Franciain this year – a famine was raging there.104

Annal 892 marks the end of the Common Stock of ‘The Anglo-SaxonChronicle’. The annals 893 to 896 form another section, a first continuation,characterised by a different date for the beginning of the year and a change in thestyle of record. This section is also characterised by enumeration of years fromwhen the fleet arrived in England. In annal 896, an overview is given of the plagues which killed many in the preceding three years. Stylistically this sectionhas features in common with Asser’s ‘Life of King Alfred’, as I argue below. Perhaps the Life was available as a model to the author of this section of theChronicle. Perhaps the texts which inspired Asser’s own method of portrayingevents were circulating at the court of Alfred (for example, Einhard’s ‘Life ofCharlemagne’).105 Some of the stylistic features of these annals were discussed byCecily Clark: ‘Not only are they longer and fuller of detail, but their syntax ...shows free use of subordination ... a wider range of connectives ... rhetoricalpatterning’. Perhaps more significantly for the historian, ‘[apparent] objectivity is partly discarded ... gives insight ... into purpose and motivation ... the annalist,instead of effacing himself, now presents himself ’, and there are common themessuch as ‘interest in tactics, expressed in concern with topography, with conditionsof military service and availability of supplies, and also with the king’s own ideas

102Annal 892: ASC.A, p. 55; ASC.B, s.a. <893>, pp. 40–1; ASC.C, s.a. 893, p. 66; ASC.D,s.a. 893, p. 30; ASC.E, p. 53; ASC.F, pp. 76–7. With this annal’s account of fortress-building (cf.nn. 115~20, below), one may compare accounts of vikings’ camps in another theatre of theiractivities: C. Downham, ‘Viking camps in ninth-century Ireland: sources, locations and interactions’,Medieval Dublin 10 (2008) 93~125.

103Annales Vedastini, s.aa. 890 and 891 (ed. Pertz, pp. 526–7; ed. Dehaisnes, pp. 336–41; ed.von Simson, pp. 68–70). Whitelock suggested that he might be the same Hæsten who was active inthe Loire Valley in 866: English Historical Documents, transl. Whitelock, p. 201, n. 12.


105I am not suggesting that Asser was the author of this section of ‘The Anglo-Saxon Chronicle’.On the acute problems of dating Asser’s De rebus gestis Ælfredi Regis, see T.M. Kalmar, ‘Thechronological scaffolding of Asser’s Vita Ælfredi regis’, Anglo-Saxon forthcoming.


and methods’.106 Such things could also be said of the ‘Life of King Alfred’, whenit is compared with the Chronicle’s annals 865–92.

The portrayal of Alfred in the Chronicle as strategist, innovator, and piousruler seems to fit so well with Asser’s picture that the annals 893–6 could be takento represent a continuation of the Chronicle, a continuation which received royalsupport or sponsorship.107 Examples of Alfred as inventor and strategist includehis successful decision to camp between the viking-armies of Appledore andMilton in 893 and his decision to keep half his force constantly on alert byalternating day- and night-shifts to keep the vikings in their camps. Alfred iscredited with choosing where the River Lea could be obstructed to curb vikings’fleet-movements in 895: this caused vikings to flee the region. Perhaps mostfamously, in annal 896 Alfred is credited with designing special longships to defeatthe enemy. This compares with Asser’s portrayal of Alfred as an innovativearchitect and patron of the arts. He wrote of Alfred’s plan for a system offortifications to provide defence against vikings; and he credited Alfred with theinvention of a candle-clock by which he could measure half of each day so that itcould be devoted to service of God.108

Alfred’s piety and clemency are also themes in both these texts. In theChronicle, annal 893, Alfred is shown as a model of godfatherliness to Hæsten.Even when Hæsten rebelled and his wife and two sons were captured, Alfredhanded them back unscathed.109 This compares with Asser’s account of Alfred’scompassion towards defeated vikings and of his sponsorship of Guthrum’sbaptism.110

In annals 893–6, the Chronicle’s first continuation offers the same pitfalls asin its Common Stock ending in 892. This new chronicler coped no better withevents happening concurrently in different locations, as Thomas Shippey hasobserved.111 This renders its account somewhat incomplete when the Chroniclerhas had to deal with some elements of the narrative, and prone to digression inorder to explain other lines of action. In annal 893 the viking-armies fromAppledore and Milton Regis slipped past the English army in Kent. Whathappened next is not particularly clear. As Shippey has stated, the next reference

106Clark, ‘The narrative mode’, pp. 221–4.107Dumville, Wessex, pp. 69–70, has shown that there is a more court-centred approach to ‘The

Anglo-Saxon Chronicle’ 893–920; that is less marked from 865 to 892.108De rebus gestis Ælfredi, §§91, 104 (Asser’s Life, ed. Stevenson, pp. 76–9, 90; Alfred, transl.

Keynes & Lapidge, pp. 101, 108–9).109Annal 893: ASC.A, pp. 56–7. Cf. ASC.B, s.a. <894>, p. 42; ASC.C, s.a. 894, p. 67; ASC.D,

s.a. 894, p. 32.110De rebus gestis Ælfredi, §56 (Asser’s Life, ed. Stevenson, pp. 46–7; Alfred, transl. Keynes &

Lapidge, p. 85).111Shippey, ‘A missing army’ (1982), p. 50; (2007), p. 336.


is to a single army. It could be that the two armies had united and thus wererepresented as a single force by the Chronicler (as the discussion above has shown,this is not without precedent).112 However, given the Chronicle’s comment on thearmy’s reünification, a little later in the same annal, Shippey was probably correctto surmise that the Chronicler had restricted his focus to the deeds of the largertroop from Appledore.113 This was defeated at Farnham (Surrey), and the survivorswere besieged on an island in the River Colne. Their immediate fate is notrecorded; however, it is later indicated that they escaped to a viking-fort at Benfleet(Essex). The Chronicler justified their being allowed to escape: the English armywas forced to march west to defend Wessex against a large fleet which arrived therefrom East Anglia and Northumbria in the same year.114 A hundred of these shipsbesieged Exeter, and another forty went west to besiege a fortress on the northcoast of Devon.

While the main English army travelled west, a small English force went eastto face the vikings at Benfleet. The Chronicle-author had to digress from thechronological portrayal of events in order to explain what had happened. AtBenfleet, Hæsten and the remnants of the army from Milton Regis and Appledorehad set up camp before the English had arrived. The Chronicler added that‘Hæsten had previously built that fort at Benfleet’ (hæfde Hæsten ær geworht þætgeweorc æt Beamfleote).115 This could mean that the fort was built in 893 beforethe defeated troops from the Colne arrived there, as Shippey has suggested.116

There is also subsequent reference to a peace-agreement between Alfred andHæsten in which the viking’s sons had been baptised and money was paid to him:Shippey has suggested that this treaty also belongs to 893. He has concluded thatAlfred may have needed to come to terms with Hæsten and the Milton-armyearlier in 893.117 If that was so, the Chronicler has glossed over these events, eitherthrough his focus on the larger viking-army from Appledore or through a desireto ‘shield Alfred from charges of bribery, and failed bribery at that’.118

If one were to wish to perceive events in 893 in another way, one could referto Æthelweard’s comments on the year 885. He stated that vikings from EastAnglia and Rochester met together at Benfleet in that year.119 If Hæsten had builtthe fort at Benfleet (rather than just refortifying it), this would suggest that he

112See above, p.17, for example.113Shippey, ‘A missing army’, p. 44; (2007), pp. 329–30.114Ibid. (1982), p. 46; (2007), pp. 331–2.115Annal 893: ASC.A, pp. 56–7. Cf. ASC.B, s.a. <894>, p. 42; ASC.C, s.a. 894, p. 67; ASC.D,

s.a. 894, p. 32.116Shippey, ‘A missing army’ (1982), p. 44; (2007), pp. 329–30.117Ibid. (1982), p. 45; (2007), pp. 330–1.118Ibid. (1982), p. 46; (2007), p. 331.119Æthelweard, Chronicon, IV.3 (ed. & transl. Campbell, p. 44).


was in one of those contingents in 885 before going to Francia. Indeed, as DorothyWhitelock pointed out, the peace-agreement could have also dated from 885, but(she noted) there is no evidence that Hæsten was leading troops in England atthat date.120 Whichever way the Chronicle is interpreted at this point, the recordis incomplete, and the narrative focus is restricted either for stylistic or for politicalreasons.

While Hæsten was away on a raid in 893, the English stormed his fortress; allthe goods, ships, and people were seized. The Chronicle describes the return ofHæsten’s family by Alfred and Hæsten’s seeming ingratitude in then raidingMercia. Certainly the partisanship of the Chronicler in favour of Alfred is clear inthis episode.

The narrative then switches back to the viking-force besieging Exeter, but noexplanation is given of the fate of the forty ships raiding North Devon. AlfredSmyth has suggested that these ships were led by Sigeferth, a viking fromNorthumbria, who was mentioned by Æthelweard in relation to 893.121 Smythhas identified him with a jarl of the same name mentioned as active in Dublinthat year.122 Nevertheless, the Chronicler restricted his narrative to the actions ofKing Alfred, who arrived at Exeter in 893 and successfully drove the vikings backto their ships.

The narrative now turns eastwards again to describe the assemblage of the twoviking-armies (þa hergas ... begen), presumably meaning those from Milton andAppledore, at Shoebury in Essex.123 They were provided with reïnforcements fromEast Anglia and Northumbria. This army then travelled via the Thames and Severnto Buttington (Montgomeryshire/Powys).124 Here the viking-army was besiegedby an English force, and the Chronicle states that the vikings were forced to eattheir horses.125 In their attempt to break this siege, the vikings were defeated. Manywere killed, but others escaped and returned to Shoebury. Reference to this battleat Buttington in an Irish chronicle not only shows its significance but could suggest

120English Historical Documents, transl. Whitelock, p. 203, n. 2.121Æthelweard, Chronicon, IV.3 (ed. & transl. Campbell, p. 50); cf. Alfred P. Smyth,

Scandinavian York and Dublin. The History and Archaeology of Two related Viking Kingdoms (2 vols,Dublin 1975/9), I.33–7.

122Ibid.; Downham, Viking Kings, pp. 268–9. Cf. discussion in Alfred, transl. Keynes & Lapidge,p. 337, n. 32.

123Annal 893: ASC.A, p. 57; ASC.B, s.a. <894>, p. 42; ASC.C, s.a. 894, p. 68; ASC.D, s.a. 894,p. 32.

124Alfred, transl. Keynes & Lapidge, p. 287, n. 16.125While eating horseflesh may have been a habit among Scandinavians at this time, this act

may show a level of desperation in an army which might need horses for fighting and transport.


that vikings from Ireland were involved.126 Reïnforcements again came from EastAnglia and Northumbria. This new army travelled north and occupied the city ofChester which – the Chronicler asserted – was deserted.127 It may be that thesetwo westward expeditions had a common motivation, perhaps linked to thedissensions in Dublin around this time or representing an attempt to secure powerin areas to the north of Wessex.

The English army employed a scorched-earth policy in the Chester-region,forcing these vikings to move into Wales the following year. After a plunderingcampaign there, they evaded English armies by travelling overland acrossNorthumbria and East Anglia to reach Mersea Island in Essex.128 We next hear ofthe viking-army which was in the South-west returning eastwards from Exeterand ravaging the coast of Sussex where it was defeated by a local force. However,the fate of the survivors and of the ships which escaped is not revealed: either theChronicler did not know or he was merely concerned to report the English victory.

A survey of the later part of the account for 893 and 894 shows that theChronicle’s focus switches frequently between events in Wessex and those takingplace elsewhere. In some ways this is a clumsy method of reporting, as Shippeyhas noted,129 but it has the virtue of keeping Alfred and Wessex in the picturewhile maintaining a ‘national’ outlook on events. It also creates a sense of the fastpace of activity for the reader, whose imagination has to weave back and forthbetween different locations. It is one of the stylistic elements which make theChronicle for these years particularly engaging.

From the beginning of annal 894 to the end of annal 895 we read of the travelsof the army from Mersea Island to the River Lea and then to Bridgnorth(Shropshire) on the River Severn. In 896 this army split into three parts. Somevikings travelled to East Anglia, some to Northumbria, and those lacking money

126The Annals of Ulster, I, edd. & transl. Mac Airt & Mac Niocaill, pp. 346/7 (893.3). Again,‘The Anglo-Saxon Chronicle’ can be shown to have a rather incomplete version of events at thispoint. It contains no previous record of viking-armies in western Mercia or Wales. However, thereis charter-evidence that vikings were active in the Wrekin-district of Shropshire in 855: P.H. Sawyer,Anglo-Saxon Charters. An Annotated List and Bibliography (London 1968), p. 123, no. 206; EnglishHistorical Documents, transl. Whitelock, pp. 526–7 (no. 90). Irish chronicles report viking-activityin Gwynedd in 877: Annala, ed. & transl. O’Donovan, I.520/1 (874.11); Chronicum Scotorum, ed.& transl. Hennessy, pp. 166/7 (877.2); The Annals of Ulster, I, edd. & transl. Mac Airt & MacNiocaill, pp. 332/3 (877.3).

127David Griffiths has stated that there ‘was arguably’ an ecclesiastical presence in the city at thisdate. However, archaeological evidence suggests that much of the city had an appearance of desertionthen. Remaining inhabitants may have left on hearing news of the vikings’ approach. SeeD. Griffiths, ‘The north-west frontier’, in Edward, edd. Higham & Hill, pp. 167–87, at p. 169.

128Northumbria stretched across to western England: ASC.A, p. 69 (s.a. 919), refers toManchester as being on Norþhymbrum.

129Shippey, ‘A missing army’ (1982), p. 50; (2007), pp. 335–6.


left England for the Seine.130 ‘The Annals of Saint-Vaast’ record the arrival of thesevikings on the Seine, led by Hundeus.131 Finally, in annal 896, the Chroniclereports that the Northumbrian and East Anglian armies harried the West Saxons.Alfred’s limited victory against six ships off the Isle of Wight is reported: it isasserted that twenty viking-ships perished along the south coast that summer, butthe circumstances are not explained.

Minimal attention is paid to events in Northumbria from 893 to 896. Forthese years the Chronicle of Æthelweard offers some revealing statements. Itinforms us that in 893 ‘the pirate Sigeferth’ came from Northumbria and raidedtwice along the south coast before returning to his own land.132 Perhaps inconsequence of this and other activities undertaken by ‘Northumbrians’ in 893,Ealdorman Æthelnoth was sent from Wessex to negotiate peace with them thefollowing year.133 Æthelnoth is perhaps to be identified with the leader of the WestSaxon troops named by Æthelweard in 893 and the ealdorman of Somerset whowas prominent in the West-Saxon resistance to viking-invaders in 878.134 Thedespatch of such a senior figure could indicate Alfred’s eagerness (perhaps evendesperation) for peace at this point. ‘The Anglo-Saxon Chronicle’ has likewise noreport of the death of Guthfrid (Old-Scandinavian Guðrøðr), king of York, andhis burial at ‘the principal church’ (basilica summa) there, mentioned byÆthelweard in relation to 895.135 From the Chronicler’s perspective, these eventswould not serve to add to the glory of King Alfred, nor would they have beenhelpful in maintaining a sustained narrative account of the majortroop-movements of the years from 893 to 896. Such omissions of informationmay provide a guide to the nature of authorial selectivity in this section of ‘TheAnglo-Saxon Chronicle’ (but there is no certainty that Æthelweard’s source wasan early version of the vernacular text). These omissions highlight some potentiallymisleading aspects of the Chronicle.

The accounts in ‘The Anglo-Saxon Chronicle’ raise many questions. Forexample, the Chronicle in these years reveals a high level of coöperation andcoördination between various viking-armies; but one must ask to what extent this

130Sawyer, The Age of the Vikings, pp. 98–9, has provided interesting commentary on thisstatement.

131Annales Vedastini, s.aa. 896 and 897: ed. Pertz, p. 530; ed. Dehaisnes, pp. 352–3; ed. vonSimson, pp. 77–9.

132Æthelweard, Chronicon, IV.3 (ed. & transl. Campbell, p. 50).133Ibid., p. 51.134Ibid., pp. 42, 50. Cf. Alfred, transl. Keynes & Lapidge, p. 337, n. 34.135Æthelweard, Chronicon, IV.3 (ed. & transl. Campbell, p. 51). For recent comment, see

Dumville, ‘Old Dubliners’, pp. 86–9, reprinted in his Celtic Essays, 2001–2007, I.113–17, andG. Williams, ‘A new coin type (and a new king?) from viking Northumbria’, Yorkshire Numismatist4 (2012) 261–75.


is a streamlined account. For example, how many small groups of freebooters arenot mentioned? Their existence may be implied by some of Alfred’s minorsea-battles. Also, when the terms ‘Northumbrians’ and ‘East Angles’ areencountered in this later part of the Chronicle, should we interpret this to meana majority or minority of Scandinavians from these locations, and were thereEnglishmen as well as Scandinavians within their ranks?136

It is possible to see the record provided by ‘The Anglo-Saxon Chronicle’ from 865to 896 as restricted and oversimplified to allow the development of a sustainednarrative, but to say that is not to belittle the achievements of its authors. Thefocus on a main line of action in the Chronicle in these years, and other stylisticfeatures, can provide evidence of how the text was created. It can be argued thatannals 865 to 892 were composed as a single block. This is not only because of itsstylistic unity (as analysed by Cecily Clark and Ruth Waterhouse)137 but alsobecause some of the annals seem to have been written with the benefit ofhindsight. In particular, there is the evidence that the notices of the vikings’Continental travels in 879–92 were written retrospectively – perhaps, as DavidDumville has argued, in the context of that army’s return to England in 892.138

The process of authorial selectivity which we see in ‘The Anglo-Saxon Chronicle’in these years may have been influenced by the outcome of events. Our onlyevidence of the authorship of the Common Stock is the use of the first personplural ‘we’ in annal 892, which may denote a self-identified group of scholars.

The ink can hardly have dried on the Common Stock of ‘The Anglo-SaxonChronicle’ (indeed the text may not have been fully completed) before Asser used

136L. Abrams, ‘Edward the Elder’s Danelaw’, in Edward, edd. Higham & Hill, pp. 128–43, at135–6. Clause five of ‘The Treaty of Alfred and Guthrum’ seems intended to stop Englishmenjoining viking-armies, which could imply that this had happened enough to provoke concern andconsequently legislation. For the text, see Die Gesetze der Angelsachsen, ed. & transl. F. Liebermann(3 vols, Halle a.S. 1898–1916), I.129; English Historical Documents, transl. Whitelock, pp. 416–17(no. 34).

137I have particularly acknowledged their insights in nn. 1, 2, 5, 7, 9, and 106, above.138Dumville, Wessex, p. 89. Cf. David N. Dumville, ‘Latin and Irish in the Annals of Ulster’,

A.D. 431–1050’, in Ireland in Early Mediaeval Europe. Studies in Memory of Kathleen Hughes, edd.Dorothy Whitelock et al. (Cambridge 1982), pp. 320–41, at 333–4, reprinted in his book Historiesand Pseudo-histories of the Insular Middle Ages (Aldershot 1990), essay XVII, with the samepagination.


it as the major source for his ‘Life of King Alfred’.139 His work seems in turn tohave influenced the composition of annals 893 to 896 of ‘The Anglo-SaxonChronicle’. These four annals constitute a coherent group, united by stylisticpeculiarities and by their enumeration of years from the arrival in Kent ofviking-armies from the Continent. This section ends once the threat of thesearmies was perceived as having receded. These annals may be contemporaryreports. The authorship of these annals is anonymous, although use of the firstperson singular in annal 894 suggests that a single person was responsible. Thestrongly pro-Alfredian character of these annals may indicate that the author wasworking directly under royal patronage.140

It can be argued that there was a flurry of historical writing in England in theyears 892 and 893. This may be put in the broader context of Alfred’s programmeof reform and revival. These writings may also have been influenced by the newdanger posed to Greater Wessex in 892 by the arrival of viking-armies in Kent.Annals 865–92 of ‘The Anglo-Saxon Chronicle’ show (as too does Asser’sbiography of the king) how King Alfred overcame previous viking-threats, andthey provide an outline-history of the armies which arrived in 892. Asser also putacross the message that, despite Alfred’s many tribulations, God was on his side.141

These messages may have been politically relevant to the small, but élite, audienceof these histories, perhaps people who had a leading role in protecting GreaterWessex in the war against vikings from 893 to 896.

139Asser used ‘The Anglo-Saxon Chronicle’ as a source as far as annal 887. The evidence for thedate of writing was reviewed in Asser’s Life, ed. Stevenson, pp. lxxi–lxxiv, lxxxii–lxxxiii, and Alfred,transl. Keynes & Lapidge, pp. 41–2, 53, 269–70 (n. 218), 287 (n. 9). It is by no means certain thatthe work is complete. For the most substantial recent contribution to study of the interrelationshipof Asser and ‘The Anglo-Saxon Chronicle’, see The Medieval Life of King Alfred the Great. ATranslation and Commentary on the Text attributed to Asser, transl. Alfred P. Smyth (Basingstoke2002), pp. 162–201, 264–8.

140Cf. Dumville, Wessex, pp. 68–71.141K. DeM. Youmans, ‘Asser’s Life of Alfred and the rhetoric of hagiography’, Mediaevalia 22

(1998/9) 291–305, at p. 297.


In this study I have sought to highlight the artistry in ‘The Anglo-SaxonChronicle’ for the annals 865–96 but also to show some of the crucial limitationsof this source as a guide to viking-activity in these years. The Chronicle isparticularly deficient in providing information on those areas of England,especially Northumbria, which were on the fringes of the Chroniclers’world-view.142

142For my attempt at an overview of the engagement of King Ívarr and his associates in thepolitics of later ninth-century England, see Downham, Viking Kings, pp. 63–80. The present articlewas drafted and redrafted in 2001–3, since when it has been altered only in response to editorialrequirements. In the interim, three books have been published which have added significantly toour understanding of ‘The Anglo-Saxon Chronicle’ and should be specially mentioned: ThomasA. Bredehoft, Textual Histories. Readings in the Anglo-Saxon Chronicle (Toronto 2001); AliceSheppard, Families of the King: writing Identity in the Anglo-Saxon Chronicle (Toronto 2004); andReading The Anglo-Saxon Chronicle. Language, Literature, History, ed. Alice Jorgensen (Turnhout2010).


II

INSULAR IDENTITIESIN THE VIKING-AGE

3

‘Hiberno-Norwegians’ and ‘Anglo-Danes’:anachronistic ethnicities and Viking-Age England1

TWO papers have recently been published, with reference to Irish sources fromthe Viking-Age, challenging the identification of Dubgaill (‘Dark Foreigners’) with‘Danes’ and Finngaill (‘Fair Foreigners’) with ‘Norwegians’.2 In this article I seekto broaden the debate by suggesting that the categorisation of Insular-vikingpolitics as a struggle between opposing Danish and Norwegian factions is similarlyunhelpful.

For example, the use of the term Dene in ‘The Anglo-Saxon Chronicle’ can beregarded as similar to the use of the terms Dani and Nordmanni in Frankishchronicles: that is, as a general name for those of Scandinavian cultural identityrather than a label referring to people of one particular Scandinavian ethnicity.3

I argue that the supposed animosity between ‘Hiberno-Norwegian’ and ‘Anglo-Danish’ factions in English politics before 954 is largely a historiographicalinvention and not a Viking-Age reality. The stereotypes applied to each of theseso-called groups (the ‘Hiberno-Norwegians’ being generally seen as more violent,more heathen, and more chaotic than the ‘Anglo-Danes’) can also be called into question.

1 I should like to thank Judith Jesch and Alex Woolf for the opportunity to present versions ofthis paper at Nottingham and St Andrews in February and April 2007. My thanks also go to PaulBibire, Stefan Brink, and David Roffe for reading and commenting on the text.

2 David N. Dumville, ‘Old Dubliners and New Dubliners in Ireland and Britain: a Viking-Agestory’, Medieval Dublin 6 (2004) 78–93, reprinted in his Celtic Essays, 2001–2007 (2 vols, Aberdeen2007), I.103–22; C. Downham, ‘The good, the bad and the ugly: portrayals of vikings in “TheFragmentary Annals of Ireland”’, The Medieval Chronicle 3 (2004) 28–40 (see chapter 6, below).

3 Quellen des 9. und 11. Jahrhunderts zur Geschichte der Hamburgischen Kirche und des Reiches,edd. & transl. Werner Trillmich & R. Buchner (Darmstadt 1961), p. 450, and Adam of Bremen,History of the Archbishops of Hamburg-Bremen, transl. Francis J. Tschan (New York 1959; rev.imp., by T. Reuter, 2002), p. 195 (IV.12), quoting Einhard, Vita Karoli regis: Dani et Sueonesceterique trans Daniam populi ab hystoricis Francorum omnes uocantur Nortmanni …, ‘Danesand Swedes and other peoples beyond Denmark are all called Northmen by the historians of the Franks’.

Medieaval Scandinavia 19 (2009) ISSN 0076-5864

41

If this argument holds true, then references to ‘Hiberno-Norwegians’ and ‘Anglo-Danes’ in modern narratives about Viking-Age England may merit somereconsideration.4

‘HIBERNO-NORSE’ OR ‘HIBERNO-NORWEGIAN’?

I should first comment on the term ‘Hiberno-Norse’ which frequently appears inmodern historical writing. It is sometimes found in contrast to ‘Anglo-Danes’ butits meaning seems to vary at the hands of different authors. In its broadest sense‘Hiberno-Norse’ is sometimes (inaccurately) used by scholars to refer to allScandinavians linked to the Gaelic-speaking areas of Ireland and North Britain.However, the term is more often used with reference to vikings from Ireland. Themeaning of ‘Norse’ is problematic. Sometimes it is applied to Scandinavians ingeneral, and sometimes it is used specifically to mean ‘Norwegians’. According tothe Oxford English Dictionary the word Norse is first recorded in the Englishlanguage in 1598; it was derived from the now obsolete Dutch word noordsch.5

In origin it means ‘northern’ or ‘nordic’, and its early use was as a linguistic label.

4 In this article I use ‘viking’ as a cultural label; I avoid ‘Norse’, ‘Norsemen’, and ‘Northmen’,because all these terms have been used in English-language historiography with specific reference toNorwegians, and they can therefore be misleading in a more general context. I consider‘Scandinavian’ to be often inappropriate in a colonial situation, for it does not reflect the hybrididentities which developed. Hybrid names including ‘Anglo-Scandinavian’, ‘Hiberno-Scandinavian’,‘Scoto-Scandinavian’, and ‘Britanno-Scandinavian’ might suit, but it would often be difficult whenreading Insular primary sources to know which of these subgroups one is dealing with. In the MiddleAges the name ‘viking’ was used to describe Scandinavian-speaking seaborne raiders, although itsmeaning has broadened considerably in modern popular usage. The term has a drawback: ‘viking’can conjure up a caricature of a warrior in a longship, much as the word ‘Norman’ invites thestereotype of a knight on horseback. Neither emblem is fully representative of a society, althougheach can be said to communicate something vital. ‘Viking’-activity characterised the coloniesfounded in the Scandinavian diaspora where power was based on dominance at sea and on militaryprowess. Here ‘viking’ can be seen to have some relevance as a cultural label, for the resources ofwhole communities were drawn on to build ships and supply expeditions, and the impact of raidingand trading reached beyond those who were involved in seafaring. This can be seen economically,socially, and also in matters of religion: Neil Price, The Viking Way: Religion and War in Late IronAge Scandinavia (Uppsala 2002). The terminology used for these peoples is currently a matter ofdebate, and it will be interesting to see how the arguments develop in future years. For somecomments on the baggage associated with the word ‘viking’ (akin to that associated with the word‘Celt’), see J. Langer, ‘The origins of the imaginary viking’, Viking Heritage Magazine 4 (2002),available online at http://www.abrem.org.br/viking.pdf (accessed 22/02/08).

5 Oxford English Dictionary, edd. John Simpson & E. Weiner (2nd edn, 20 vols, Oxford 1989), X.519.


Walter Scott introduced the term ‘Norseman’ into Scots and English in 1817.6

He also popularised the word ‘Norse’ as a people-name (rather than just a linguisticterm), although it had been used thus by Scottish authors from the seventeenthcentury.7 Since the nineteenth century, use of the term ‘Norse’ has been criticisedas having a Norwegian bias.8 This bias seems to reflect usage of the word inScottish historiography and literature where ‘Norse’ has often been employed tomean ‘Norwegian’, and this interpretation has spread more widely. The Scottishusage may point to an alternative Scots derivation of ‘Norse’ from norsk as meaning‘Norwegian’ in modern Scandinavian languages.9 Or it may be due to the factthat Norway was regarded as the natural homeland of the viking-settlers who cameto North Britain.10 The use of ‘Hiberno-Norse’ and ‘Norse’ as ethnic labels cancause confusion, as different authors have used these terms to mean differentthings. The use of ‘Hiberno-Norse’ meaning ‘Hiberno-Norwegian’ is particularlytroublesome, as I hope to demonstrate in this paper through analysis of theterminology used by mediaeval authors and modern historians.

VIKINGS IN ‘THE ANGLO-SAXON CHRONICLE’

It is evident that Latin words Nordmanni and Dani were used interchangeably inEuropean chronicles composed in Latin during the ninth and tenth centuries.11

One should therefore question whether their Old-English equivalents, Norðmennand Dene, were used to distinguish separate ethnic groups in the First Viking-Age.

6 Ibid., X.520. The first known appearance of the term ‘Norseman’ is due to Walter Scott, Harold the Dauntless. A Poem in Six Cantos (London 1817); P. Roberts, ‘Sir Walter Scott’scontributions to the English vocabulary’, Proceedings of The Modern Language Association of America 68 (1953) 189–210, at p. 198.

7 Alexander Garden, A Theatre of Scottish Worthies: and the Lyf, Doings, and Deathe of WilliamElphinston, Bishop of Aberdeen, ed. Alexander Laing (Glasgow 1878), p. 17: ‘I gave att Largs a foullyet famous foile, where numbers of thy Norces left yet ly’; Walter Scott, The Pirate, edd. Mark Weinstein & A. Lumsden (Edinburgh 2001), pp. 10–11, ‘Land … in the possession of theNorse inhabitants’ (this was first published in 1822).

8 Andrew Wawn, The Vikings and the Victorians. Inventing the Old North in Nineteenth CenturyBritain (Cambridge 2000), pp. 4, 148.

9 Scottish National Dictionary, ed. William Grant (10 vols, Edinburgh 1941–76), VI.443.10 See below, pp. 53~68.11 Jean Renaud, Les Vikings en France (Rennes 2000), p. 78; Eric Christiansen, The Norsemen in

the Viking Age (Oxford 2002), pp. 116–18; P. Bouet, ‘Les chroniqueurs francs et normands face auxinvasions vikings’, in L’Héritage maritime des vikings en Europe de l’ouest, ed. Élisabeth Ridel (Caen2002), pp. 57–73, at p. 59, n. 6.

ANACHRONISTIC ETHNICITIES & VIKING-AGE ENGLAND 43

Paul Bibire has argued that before the late tenth century both Old-English Deneand Old-Scandinavian Danir referred to Scandinavians in a general way.12 It canbe argued of the ninth and early tenth centuries that Denmark and Norway werenot yet politically unified.13 It would thus be anachronistic if ninth-century viking-groups were to be identified primarily by use of the distinct national labels ‘Dane’and ‘Norwegian’.14 The case will be put that Norðmenn and Dene were usedinterchangeably in English sources of the ninth and earlier tenth centuries. Theargument will focus on ‘The Anglo-Saxon Chronicle’ as the most influential sourceused by historians of Viking-Age England. It can be held that the terms used todescribe vikings in this chronicle show little concern to categorise them intodifferent Scandinavian nationalities.

The terms used to describe vikings in ‘The Anglo-Saxon Chronicle’ in annals787–954 varied in popularity over time.15 This gives the impression that authorialpreference was in play, or that concerns other than ethnicity were determining thelanguage used. The word hæðen occurs frequently in the annals from 832 to 872,and it is most popular during the 850s to the exclusion of other labels. It is onlyfound once again, for the year 942. In annals from 860 to 892 the word here(‘invading army’) was the favoured usage.16 It is little used thereafter, withappearances only in the annals for 910, 914, and 915. The term wicing(‘viking’/‘pirate’) is employed in annals 879 and 885 and then only used oncemore, in version A in annal 917. Norðmenn is recorded in annal 787 – unlikely

12 P. Bibire, ‘North Sea language contacts in the early Middle Ages: English and Norse’, in TheNorth Sea World in the Middle Ages. Studies in the Cultural History of North-western Europe, edd.Thomas R. Liszka & L.E.M. Walker (Dublin 2001), pp. 88–107, at p. 89.

13 Christiansen, The Norsemen, pp. 118–23; K. Crag, ‘The early unification of Norway’, in TheCambridge History of Scandinavia, I, ed. Knut Helle (Cambridge 2003), pp. 184–201; I. Skovgaard-Petersen, ‘The making of the Danish kingdom’, ibidem, pp. 168–83. (Cf. also the papers by U.Näsman, J. Ringtved, and M. Axboe in Anglo-Saxon Studies in Archaeology and History 10 [1999]1–10, 49–63, 109–18.) Cf. Uta Goerlitz, Literarische Konstruktion (vor-)nationaler Identität seit dem‘Annolied’. Analysen und Interpretationen zur deutschen Literatur des Mittelalters (11.–16. Jahrhundert)(Berlin 2007), and see n. 28, below.

14 Christiansen, The Norsemen, p. 117.15 The Anglo-Saxon Chronicle. A Collaborative Edition, III, MSA, ed. Janet M. Bately

(Cambridge 1986), pp. 39–74; The Anglo-Saxon Chronicle. A Collaborative Edition, IV, MSB, ed.Simon Taylor (Cambridge 1983), pp. 28–54; The Anglo-Saxon Chronicle. A Collaborative Edition,V, MS C, ed. Katherine O’Brien O’Keeffe (Cambridge 2001), pp. 50–80; The Anglo-Saxon Chronicle.A Collaborative Edition, VI, MSD, ed. G.P. Cubbin (Cambridge 1996), pp. 16–45; The Anglo-SaxonChronicle. A Collaborative Edition, VII, MS.E, ed. Susan Irvine (Cambridge 2004), pp. 41–56; TheAnglo-Saxon Chronicle. A Collaborative Edition, VIII, MSF, ed. Peter S. Baker (Cambridge 2000),pp. 53–81.

16 T.A. Shippey, ‘A missing army: some doubts about the Alfredian Chronicle’, Anglo-Saxon 1(2007) 319–38.


to be a contemporary record – and then reäppears in 920 (version A), 937, and942.17Dene is similarly infrequent, with occurrences in annals 900, 917, 942, and943. English regional names Norþ(an)hymbre and Eastengle are used to describeviking-armies from 893 to 919 and again from 944 to 948.

Overall, Denisc, used as an adjective or a noun, was most frequently employedto label vikings in ‘The Anglo-Saxon Chronicle’ throughout the ninth and earlytenth centuries. Denisc first appears as an adjective in annal 787. Deniscan is usedas a noun in annal 833. Thereafter these labels are employed to the near-exclusionof all others until 845 (when hæðen becomes the dominant term). Denisc(an)reäppears in annal 870 and remains in frequent use until annal 913. Denisc isthereafter used in version A of the Chronicle in annals 918 and 920. Scholars haveusually considered that Denisc means ‘Danish’; but I am grateful to Paul Bibirefor pointing out in conversation that an adjectival form of the word Norðmenn(which is usually taken to mean ‘Norwegian’) does not appear in Old Englishuntil the eleventh century. Denisc might therefore apply to all Scandinavians inthe ninth and tenth centuries.18 It can be argued that the use of labels in ‘TheAnglo-Saxon Chronicle’ reflects authorial preferences across particular time-frames, rather than an attempt to distinguish viking-groups by ‘Danish’ or‘Norwegian’ ethnicity. Furthermore, it can be argued that the words Norðmenn,Denisc, and Dene refer to Scandinavians in general.

In versions B, C, D, and E of ‘The Anglo-Saxon Chronicle’ under the year 787there is an account of a viking-attack, in which Norðmenn is shown to be equivalentto denisce menn: ‘on his dagum coman ærest .iii. scipu Norðmanna; … þæt wæronþa ærestan scipu deniscra manna þe Angelcynnes land gesohtan’ (‘and in his daysthere came for the first time three ships of Norðmenn… Those were the first ships

17 It should be noted that there are references to a ‘northern’ king in 890 and to ‘northern’armies in 910 and 937. ‘North’ is used also as a relative concept (not only as a pseudo-ethnic term),as in 823, norð ofer Temese (‘north across the Thames’). For examples, see: ASC.A, ed. Bately, pp.41, 54, 71; ASC.B, ed. Taylor, p. 47.

18 F. Amory, ‘The dönsk tunga in early medieval Normandy: a note’, in American Indian andIndoeuropean Studies: Papers in Honor of Madison S. Beeler, edd. Kathryn Klar et al. (Den Haag1980), pp. 279–89.


of denisc menn which came to the land of the English’).19 Furthermore in the D-text under the year 943 a king Óláfr and his followers are identified as Dene but,in the previous year, they are taken to be Norðmenn. These records challengeassumptions which scholars have made about the ethnic connotation of theselabels. Even where a contrast between ‘Danes’ and ‘Norwegians’ has been perceivedin records of events in ‘The Anglo-Saxon Chronicle’ – in annals 920A and942ABCD, these distinctions may be shown to result from a preconception thatNorðmenn means ‘Norwegians’ and Dene means ‘Danes’. In both cases it can beargued that there is a repetition of terms with similar meanings to produce aparticular effect.

In annal 920 the subjection of various Insular peoples to King Edward theElder is said to have occurred at a royal meeting at Bakewell (Derbyshire).20 Inthe A-text the list of those who submitted, and the terms of their subjection, aremade to sound as impressive as possible:21

hine geces þa to fæder 7 to hlaforde Scotta cyning 7 eall Scotta þeod 7 Rægnald 7 Eadulfessunu 7 ealle þa þe on Norþhymbrum bugeaþ ægðer ge Englisce ge Denisce ge Norþmenge oþre 7 eac Stræcledweala cyning 7 ealle Stræcledwealas,

‘then the king of the Scots and all the people of the Scots, and Rögnvaldr and the sons ofEadwulf and all who lived in Northumbria, English and Danish, and Northmen andothers, and also the king of the Strathclyde Britons and all the Strathclyde Britons, chosehim as father and lord’.

I interpret the categories being identified here as the people of Alba and theNorthumbrians (namely English Northumbrians and Scandinavian Northumbriansand other Scandinavians and all others dwelling in Northumbria) and alsoStrathclyders. By using this formula, the chronicler sought to be as inclusive as

19 Annal 787: ASC.B, ed. Taylor, p. 28. Cf. ASC.C, ed. O’Keeffe, p. 50; ASC.D, ed. Cubbin, p.16; ASC.E,ed. Irvine, p. 41; ASC.F, ed. Baker, p. 53. Cf. The Chronicle of Æthelweard, ed. & transl. A.Campbell (Edinburgh 1962), p. 27 (III.1). The Anglo-Saxon Chronicle, transl. Michael Swanton(London 1996), p. 54, n. 4; Bibire, ‘North Sea language contacts’, p. 89. Cf. ASC.A, ed. Bately,p. 39, which lacks Norðmanna (as does its copy, version G). Texts DEF specify these Scandinavians’homeland: of Hereðalande.

20 Clare Downham, Viking Kings of Britain and Ireland. The Dynasty of Ívarr to A.D. 1014(Edinburgh 2007; 2nd edn, 2008), pp. 95–7, 150.

21 ASC.A, ed. Bately, p. 69 (s.a. 920). I am grateful to Paul Bibire for pointing out that in theabsence of hooked o the spelling Rögnvaldr is more accurate than Røgnvaldr, the form used byDownham, Viking Kings.


possible in naming the peoples of Northumbria and North Britain.22

The idea that England was peopled by two principal ethnic groups, the Englisc(English) and the Denisc (Scandinavians), is found at various points in ‘The Anglo-Saxon Chronicle’ and in Anglo-Saxon law-texts.23 The pairing of Angelcynn/Ængleand Denisc/Dene as opposing categories can be seen in ‘The Anglo-Saxon Chronicle’in annal 886 where it is said of King Alfred that him all Angelcyn to cirde þæt butondeniscra monna hæftniede was (‘all the English turned to him, except those in captivityto deniscmen’).24 It is said of him again in annal 900, se wæs cyning ofer eall Ongelcynbutan ðæm dæle þe under Dena onwalde wæs (‘he was king over all the English exceptthat part which was under the control of Dene’).25 In 910, Ængle 7 Dene fought atTettenhall.26 Eight years later, King Edward captured Nottingham, and ‘all the peoplesettled in the land of the Mercians, both denisc and englisc, turned to him’.27 It hasbeen argued that Alfred and his successors sought to promote a unified sense of English identity through the use of words Angelcynn and

22 The author of annal 920A in ‘The Anglo-Saxon Chronicle’ was careful to specify thatEdward’s authority extended to those of both English and Scandinavian identity in Northumbria.Such care was perhaps necessary, as the term ‘Northumbrians’ was used fairly fluidly in the lateninth and tenth centuries to refer to either English or Scandinavians or all groups inNorthumbria. For example, in the 890s the term ‘Northumbrian’ is found with reference to aviking-army but in a tenth-century alliterative charter it seems to be used in reference to people ofEnglish rather than Scandinavian identity: ASC.A, ed. Bately, p. 55 (s.a. 893); C. Downham,‘Religious and cultural boundaries between vikings and Irish: the evidence of conversion’, in TheMarch in the Islands of the Medieval West, edd. Jenifer Ní Ghrádaigh & E. O’Byrne (Leiden 2012),pp. 15–34.

23 Dawn Hadley, The Vikings in England. Settlement, Society and Culture (Manchester 2006), p. 32: ‘settlers, whatever their background, were labelled as Danes for legal purposes’. For example,see The Laws of the Kings of England from Edmund to Henry I, ed. & transl. A.J. Robertson(Cambridge 1925), pp. 32/3 (IV Edgar 2a §2), eallum leodscype, ægðer ge Englum ge Denum geBryttum, on ælcum ende mines anwealdes, ‘to the whole nation – to the English, Danes and Britonsin every part of my dominion’.

24 ASC.A, ed. Bately, p. 53 (s.a. 886). Cf. ASC.B, ed. Taylor, p. 39 (s.a. 887); ASC.C, ed.O’Keeffe, p. 64 (s.a. 887); ASC.D, ed. Cubbin, p. 29 (s.a. 886); ASC.E, ed. Irvine, p. 52 (s.a.886).

25 ASC.A, ed. Bately, p. 61 (s.a. 900). Cf. ASC.B, ed. Taylor, p. 46 (s.a. 901); ASC.C, ed.O’Keeffe, p. 71 (s.a. 901).

26 ASC.B, ed. Taylor, p. 49 (s.a. 910); ASC.C, ed. O’Keeffe, p. 75 (s.a. 910); ASC.D, ed.Cubbin, p. 37 (s.a. 910).

27 ASC.A, ed. Bately, p. 69 (s.a. 918).


Englisc.28 One tactic in promoting this sense of unity was to pitch these terms inopposition to a foreign ‘Other’ whose existence might be seen to threaten thepower or success of the self-referred group. It can be argued in this use of opposingpairs of ethnic terms that Denisc or Dene functioned as an inclusive term todescribe all those of Scandinavian identity in Britain. It was contrasted withAngelcynn or Englisc, referring to people of English identity.

Apart from the record for the year 920 in ‘The Anglo-Saxon Chronicle’, theother place where there is a perceived distinction between Dene and Norðmenn isin annal 942, in the poem which celebrates the capture of five boroughs fromviking-control.29

Her Eadmund cyning Engla þeodenmaga mundbora Myrce geeodedyre dædfruma swa Dor scadeþHwitanwylles geat and Humbra eabrada brimstream burga fife 5Ligoraceaster and Lincyleneand Snotingaham swylce Stanford eacand Deoraby Dæne waeran ærorunder Norðmannum nyde gebegdeon hæþenra hæfteclommum 10lange þrage oþ hie alysde eftfor his weorþscipe wiggendra hleoafera Eadweardes Eadmund cyning.

The translation which was published by Allen Mawer a century ago will probablysound familiar to all who have studied this period of English history. From line5b it reads: ‘… the five boroughs, Leicester and Lincoln and Nottingham, likewiseStamford also and Derby. The Danes were before this subject for a long time byforce under the Norsemen, in bonds of captivity to the heathens …’.30 In thisinterpretation Edmund is seen as the liberator of Danes from the evil clutches ofthe heathen Norwegians.

28 P. Wormald, ‘Engla Lond: the making of an allegiance’, Journal of Historical Sociology 7(1994) 1–24; S. Foot, ‘The making of Angelcynn: English identity before the Norman conquest’,Transactions of The Royal Historical Society, 6th series, 6 (1996) 25–49.

29 ASC.A, ed. Bately, p. 73. In quoting the text of this poem, I have removed all internalpunctuation to assist thoroughgoing reëvaluation of the poem’s meaning. Cf. ASC.B, ed. Taylor,p. 53; ASC.C, ed. O’Keeffe, p. 79; ASC.D, ed. Cubbin, p. 43.

30 A. Mawer, ‘The Scandinavian kingdom of Northumbria’, Saga-book 7 (1911/12) 38–64, atpp. 61–2; A. Mawer, ‘The Redemption of the Five Boroughs’, English Historical Review 38 (1923)551–7, at p. 551.


The poem is found in versions A, B, C, and D of ‘The Anglo-Saxon Chronicle’.There is some variation in the grammatical case of Dene given in different versions.Nominative or accusative plurals Dæne and Dene are given in versions A, C, andD. The dative plural Denum is the reading given in the B-text. Scholars workingin the nineteenth and early twentieth centuries tended to prefer the reading of theB-text. The translation given by Henry Petrie (1848), Charles Plummer (1899),Walter Sedgefield (1904), and E.E.C. Gomme (1909) is rather different fromMawer’s and along the lines of ‘… five boroughs … they were under the Danes,under the Northmen in heathen fetter-bonds a long time …’.31 In this earlier habitof translation the Dene are regarded as being in league with, or as being the sameas, the Norðmenn. On text-historical grounds the readings of versions A, C, andD are to be regarded as superior to that of B.32 Nonetheless it is interesting that ashift in the translation of ‘The Anglo-Saxon Chronicle’ reflects a change ofperception by scholars in the early twentieth century, that vikings in Britain mightbe categorised into rival ethnic groups. This historiographical development isdiscussed in the next section of the present chapter.33

If it be considered that Danes and Norwegians were not distinctive rival groupsin early tenth-century England, another translation which respects the readingsgiven in versions A, C, and D of ‘The Anglo-Saxon Chronicle’ can be made. Mawersuggested that a sentence-division should be inserted into the poem following thelist of five boroughs. This interpretation has been followed by most subsequenttranslators of the poem.34 Nevertheless, if one took the poem as a single sentence,then Dæne wæron ær (‘They were previously Dene’) could be an interjection referring back to the Myrce (‘Mercians’) whom Edmund defeated. In thisreïnterpretation a distinction is being made between the Engle whom Edmund leads

31 Monumenta Historica Britannica, or Materials for the History of Britain, from the EarliestPeriod, I, edd. Henry Petrie & J. Sharpe (London 1848), p. 387; Two of the Saxon ChroniclesParallel, with Supplementary Extracts from the Others, ed. Charles Plummer (2 vols, Oxford1892/9; rev. imp., by D. Whitelock, 1952), II.143; The Battle of Maldon and Short Poems fromThe Saxon Chronicle, ed. Walter Sedgefield (Boston, MA 1904), p. 41; The Anglo-Saxon Chronicle,transl. E.E.C. Gomme (London 1909), p. 89.

32 Mawer, ‘The Redemption’, pp. 552–3.33 See below, pp. 51~4.34 Mawer, ‘The Scandinavian kingdom’, pp. 61–2. For example: The Anglo-Saxon Minor

Poems, ed. Elliott van Kirk Dobbie (New York 1942), pp. xli, 20–1; The Anglo-Saxon Chronicle. ARevised Translation, transl. Dorothy Whitelock et al. (London 1961; rev.imp., 1965), p. 71; TheAnglo-Saxon Chronicle, transl. Swanton, p. 110.


and the Mercian Dene whom he conquers. A new translation could read asfollows.35

‘In this year King Edmund, lord of the Engle,protector of men, conquered Mercians,noble doer of deeds, as the Dore divides,Whitewell gate and Humber’s river,broad sea-stream, five boroughs, 5Leicester and Lincoln,and Nottingham and Stamford also,and Derby – they were previously Dene –,oppressed in need under Northmen,in the fetter-chains of heathens 10for a long time, until he freed them againfor his glory, shield of warriors,offspring of Edward: King Edmund.’

This interpretation does not eliminate the option of distinction between Dene andNorðmenn, but it does allow the possibility that here they were members of the samegroup. It could be argued that the poet used the terms Norðmenn and hæðen, in addition to Dene, to belabour the alien domination of the five boroughs.Edmund’s campaign is implicitly justified by the fact that he was conquering thoseMercians who could be identified as foreigners (or under foreign control).36 Thisnew reading also sits better with the record in ‘The Anglo-Saxon Chronicle’ whereKing Óláfr, who was defeated by the capture of the five boroughs in 942, isidentified as leading an army of Dene in 943.37 Furthermore, parallels can be drawnbetween the seizure of boroughs from the control of Dene in 942 and accounts of

35 I am particularly grateful to Paul Bibire for his assistance in achieving this version.36 This translation may resolve the difficulty which Allen Mawer had in explaining the

description of the boroughs as being held down for ‘a long time’. The poet was not, perhaps,referring to the relatively brief period of York’s domination of five boroughs, but to thesettlement and domination of the region by Scandinavian settlers since the late ninth century. DavidRoffe has made the interesting suggestion in e-mail correspondence that ‘The reference tofive boroughs must have in mind the re-organization in local government that led to theestablishment of the Five Boroughs as an institution (clearly English and post-942). Was theintention to promote the new organization?’ If this institution was established in 942 it may havedisenfranchised some of the Scandinavian lords who had wielded power at a local level (albeitunder the rule of West-Saxon kings). Mawer, ‘The Redemption’, p. 555; C. Downham,‘The chronology of the last Scandinavian kings of York’, Northern History 40 (2003) 25–51, atpp. 39–40.

37 ASC.D, ed. Cubbin, p. 43 (s.a. 943).


other West-Saxon victories recorded in ‘The Anglo-Saxon Chronicle’ during theearly tenth century. In 912 King Edward constructed a stronghold at Witham inEssex where him beag god dæl þæs folces to þe ær under deniscra manna anwalde wæron(‘a good part of the people previously under the control of dæniscra mannasubmitted to him’).38 After a vigorous season of campaigns by King Edward in 917,him cirde micel folc to, ægþer ge on Eastenglum ge on Eastseaxum, þe ær under Denaanwalde wæs (‘a great multitude, both in East Anglia and in Essex, which was earlierunder the control of Dene, turned to him’).39 Thus ‘The Anglo-Saxon Chronicle’conveys the impression that people who dwelt in areas under Scandinavian rulewillingly submitted to the West-Saxon king in order to throw off the yoke of viking-oppression. The same image may be intended in annal 942. Whether all theinhabitants felt liberated by the extension of West-Saxon power across areas whichhad previously been ruled by East Angles or Mercians or Northumbrians, as well asvikings, is a matter for speculation. It can only be expected that ‘The Anglo-SaxonChronicle’ gives a rather one-sided view.

VIKINGS IN INSULAR CHRONICLING

If ethnic labels were not used to distinguish different viking-groups in the ninthcentury, this raises the question of the terms used to distinguish one viking-armyfrom another. A brief investigation of Insular chronicles suggests that, in the ninthcentury, Viking-Age chroniclers described armies in a variety of different ways.These included the identity of their leaders; their bases or the area where they lastcampaigned; the areas where they settled; the relative size of their army; and eventheir level of contact with the local population.

Examples of identification of viking-groups by their leaders are found in ‘TheAnglo-Saxon Chronicle’: 871, two divisions, one led by Bagsecg and Hálfdan, theother by earls; 875, Hálfdan; 875, ‘Guthrum, Oscetel, and Anwend’; 876, Hálfdan;878, the brother of Ívarr and Hálfdan; and, for the years 892 and 893, Hæsten.40 InIrish chronicles frequent reference is made to the leaders of viking armies. Forexample, in ‘The Annals of Ulster’ for 853 ‘Amlaíb’ (Old-Scandinavian Óláfr) brings a fleet to Ireland and is referred to as an army-leader until

38 ASC.A, ed. Bately, p. 64.39 ASC.A, ed. Bately, p. 68.40 The Anglo-Saxon Chronicle, transl. Whitelock et al., pp. 46~55.


871.41 From 856 he is often found in alliance with ‘Ímar’ (Old-ScandinavianÍvarr).42

References to vikings by their bases or last campaign-site are found in ‘TheAnglo-Saxon Chronicle’: 867, ‘from East Anglia’; 879, ‘from Chippenham’; 880,‘from Cirencester’ and ‘the army which had encamped at Fulham’; 882, ‘theFrankish empire’; 885, ‘the army in East Anglia’; 886, ‘the army which had goneeast’; 893, ‘army which had been at Milton … which had been at Appledore’; and894, ‘who were encamped on Mersea’.43 In ‘The Annals of Ulster’ there are manyreferences to viking-bases, including ‘heathens from Linn Duachaill ’, ‘heathensfrom Dublin’, and ‘heathens from Cael Uisce’ for 842, and further references to‘heathens from Lough Ree’ for 844 and ‘the ships of Limerick’ for 845.44

References to the areas where vikings settled are found in ‘The Anglo-SaxonChronicle’ and include: 893, ‘those . . . who lived in Northumbria and EastAnglia’; 896, ‘the armies in East Anglia and Northumbria’; and 910, ‘the army inNorthumbria’.45 In Irish reporting, viking-settlement focused on individual sitesrather than areas. Nevertheless there is reference to ‘the Foreigner-Gaels of LethCuinn’ in Chronicum Scotorum for 858.46

The term micel here, ‘big/great army’, was employed in ‘The Anglo-SaxonChronicle’ in annal 866, and the same force is referred to in the account of the870s as simply here. This may refer to the biggest invading army active in southernBritain at that time; it should be noted, however, that the composition of the armyaltered over time, with the addition of new men and the departure of others.47 Inannal 893 there is reference to se micla here … þe ær … sæt æt Apuldre (‘the largearmy which had been encamped at Appledore’).48

In Irish chronicles the trio of names introduced to distinguish viking-groups inthe mid-ninth century – Finngaill, Dubgaill, and Gallgoídil – may be translated as

41 Downham, Viking Kings, pp. 238–40.42 Ibid., pp. 258–9.43 The Anglo-Saxon Chronicle, transl. Whitelock et al., pp. 45–56.44 The Annals of Ulster (to A.D. 1131), I, edd. & transl. Seán Mac Airt & G. Mac Niocaill

(Dublin 1983), pp. 300–3.45 The Anglo-Saxon Chronicle, transl. Whitelock et al., pp. 55, 57, 61.46 Chronicum Scotorum. A Chronicle of Irish Affairs, from the Earliest Times to A.D. 1135; with a

Supplement containing the Events from 1141 to 1150, ed. & transl. William M.Hennessy (London 1866), pp. 156/7.

47 The Anglo-Saxon Chronicle, transl. Whitelock et al., pp. 47–8 (annals 871 and 875).48 Ibid., p. 55; cf. Shippey, ‘A missing army’, pp. 324/5.


‘Old Foreigners’, ‘New Foreigners’, and ‘Foreigner-Gaels’.49 These categories seemto reflect the level of contact or interaction which these groups had with the Gaels.These terms endured, and their meanings adapted over time. Gallgoídil eventuallybecame associated with the area of Galloway in Scotland. Dubgaill soon becameso settled into Irish politics that they may have been considered not so much as‘new’ foreigners but as being under the leadership of descendants of the first leadersof the Dubgaill.

Occasionally, reference seems to be given to the homeland of a Scandinavianforce, although all such references are controversial. The DEF-texts of ‘The Anglo-Saxon Chronicle’ (representing in their agreement ‘The Northern Recension’ ofaround A.D. 1000) identify Hereðaland (which may be Hordaland in Norway) asthe origin of three ships which arrived in Dorset during the reign of Beorhtric, kingof the West Saxons (786–802).50 That Hereðaland is not mentioned in otherrecensions of ‘The Anglo-Saxon Chronicle’ indicates that the information may havebeen added in Northumbria around 1000.51 Asser, the late ninth-centurybiographer of King Alfred, described the army which arrived in East Anglia in 865as having come from the Danube (de Danubia). Here Asser may have confusedDacia (the area of modern Romania, through which the Danube flows) with Dania(Denmark/ Scandinavia). Alternatively he may have been alluding to an origin-legend, which is recorded in the eleventh century, that vikings hailed from Dacia.52

In Irish chronicles reference is made to the arrival of a son of the king of Laithlinnin the mid-ninth century.53 The meaning of Laithlinn is intensely controversial butmay designate an area to the north-east of Ireland, including some of the Scottish

49 A.P. Smyth, ‘The Black Foreigners of York and the White Foreigners of Dublin’, Saga-book19 (1974–7) 101–17; Dumville, ‘Old Dubliners’, especially pp. 83–4, 91–2. Downham, ‘Vikingidentities in Ireland’.

50 Annal 787: ASC.D, ed. Cubbin, p. 16 (of Hæreðalande); ASC.E, ed. Irvine, p. 41 (ofHereðalande); ASC.F, ed. Baker, p. 53 (of Hereðalande).

51 ASC.E, ed. Irvine, pp. xiii, xxxiii.52 Alfred, transl. Keynes & Lapidge, p. 238, n. 44; Dudo of St Quentin, History of the Normans,

transl. Eric Christiansen (Woodbridge 1998), pp. 15–16 (I.i~ii). If Asser was referring to anorigin-legend, this indicates interest in King Alfred’s circle to understand and interpretinformation about Scandinavian political geography and identity, which may have beenundergoing an important stage of conceptual development at the end of the ninth century. It isrelevant to note that the first reference to Denmark thus is found in ‘The Voyage of Ohthere’written at Alfred’s court: The Old English Orosius, ed. Janet Bately (Oxford 1980), pp. 13–16(I.i).

53 ‘The Annals of Ulster’, s.a. 852[=853].2 (edd. & transl. Mac Airt & Mac Niocaill, I.312/13).


islands and/or parts of Scandinavia. It is uncertain how this name fits withLochlann, a term in use from the eleventh century which then means Norway orthe Nordic countries.54

It can be argued that Insular chroniclers in the ninth and early tenth centurieswere not preöccupied with identifying whether groups were Danes or Norwegians,for such distinctions had no apparent relevance in a contemporary context.55 Thesame argument can be made from the Frankish evidence.56 It seems that viking-armies often comprised a diverse range of individuals who had come together forthe purposes of a particular campaign, or following a particular leader.57We cannotexpect that armies were uniformly composed of people from just one part ofScandinavia.

ENGLISH-LANGUAGE HISTORIOGRAPHY IN THE NINETEENTH CENTURY

If the division between Danes and Norwegians was not functional in the ninthcentury, this naturally raises the question why later historians introduced thesecategories in their discussions of the Viking-Age in Britain and Ireland. Until thenineteenth century, ‘Dane’ was often used as a catch-all word in English

54 A. Ahlqvist, ‘Is acher in gáith … úa Lothlind ’, CSANA Yearbook 3/4 (2005) 19–27; C.Etchingham, ‘The location of historical Laithlinn/Lochla(i)nn: Scotland or Scandinavia?’, inProceedings of the Seventh Symposium of Societas Celtologica Nordica, ed. Mícheál Ó Flaithearta(Uppsala 2007), pp. 11–31.

55 N. Higham, ‘Viking-Age settlement in the North-western countryside: lifting the veil?’, inLand, Sea and Home. Proceedings of a Conference on Viking-period Settlement, at Cardiff, July 2001,edd. John Hines et al. (Leeds 2004), pp. 297–311, at p. 303: ‘Nor is it clear that contemporarycommunities distinguished uniformly between Danes, Norse, Irish, Britons and English in thesame ways and for the same reason as modern scholars have been inclined to’.

56 Examples in ‘The Annals of Saint-Bertin’ from the mid-ninth century demonstrate thatviking-groups could be identified by their leader’s name (s.aa. 845, 850, 861); by the size of afleet (s.a. 852); by stating where they last campaigned (s.aa. 845, 856) or where they were based(s.a. 865). But there is no consistent distinction made between Nordmanni and Dani. See Annalesde Saint-Bertin, edd. Félix Grat et al. (Paris 1964), pp. 61, 63, 72, 78, 89, 106–7, 149–51. Cf.chapter 4, below.

57 N. Lund, ‘The armies of Swein Forkbeard and Cnut: leding or lið?’, Anglo-Saxon England 15(1986) 105–18. Cf. Hadley, The Vikings in England, pp. 83, 306; S. Keynes, ‘The vikings inEngland, c. 760–1016’, in The Oxford Illustrated History of the Vikings, ed. Peter Sawyer (Oxford1997), pp. 48–82, at p. 54; M. Innes, ‘Danelaw identities: ethnicity, regionalism and politicalallegiance’, in Cultures in Contact: Scandinavian Settlement in England in the Ninth and TenthCenturies, edd. Dawn Hadley & J. Richards (Turnhout 2000), pp. 65–88, at p. 78.


historiography to describe all Scandinavians or vikings.58 However, during thenineteenth century there were calls for a more specific terminology to be applied toScandinavian raiders and settlers in order to distinguish the impact of Norwegiansand Danes in mediaeval Britain. This concern to demarcate the influence ofdifferent groups seems to have been entangled with contemporary nationalisticagenda. In 1852 J.J.A. Worsaae insisted in his book An Account of the Danes andNorwegians in England, Scotland and Ireland, that ‘the Norwegians in Scotland’were ‘the most numerous of all the Scandinavian colonists’ but observed that theword Dane was commonly used to describe Scandinavians in a North-Britishcontext. Worsaae (a Dane himself ) attributed ‘the preponderance of the Danishname’ to59

the pre-eminent power of the Danes in ancient times, and in the early middle ages; andof course, more particularly to the supreme domination which they had so gloriously wonfor themselves in the neighbouring country of England.

This echoes, in some respects, the opinion expressed by Daniel Wilson in hisArchaeology and Prehistoric Annals of Scotland, published the previous year:60

the invariable adoption of the latter term [Danes] in preference to that of Norwegians orNorsemen,61 shows how completely Scottish and Irish antiquaries have abandonedthemselves to the influence of English literature, even where the appropriation of itsdogmas was opposed to well-known historical facts.

Wilson clearly resented the overt influence of English scholarship oninterpretations of the Viking-Age in Scotland.62

Some scholars objected to the prevalent use of the word ‘Dane’, not merelybecause it was deemed inaccurate but also because it seemed to deny thesignificance of Norway in the Viking-Age. It was the Norwegian, Peter Andreas

58 J. Graham-Campbell, ‘“Danes … in this country”: discovering the vikings in Scotland’,Proceedings of The Society of Antiquaries of Scotland 134 (2004) 201–39, at p. 202.

59 J.J.A. Worsaae, An Account of the Danes and Norwegians in England, Scotland and Ireland(London 1852), p. 199.

60 Daniel Wilson, Archaeology and Prehistoric Annals of Scotland (Edinburgh 1851), pp. xiv–xv;Graham-Campbell, ‘Danes … in this country’, p. 221.

61 Here the equation of Norse and Norwegian, which has caused such confusion of usage, isquite explicit.

62 A. Newby & L. Andersson Burnett, ‘Between Empire and “The North”: Scottish identity inthe nineteenth century’, in Parting the Mists. Views on Scotland as Part of Britain and Europe, ed.Henrik Meinander (Helsinki 2008), pp. 37–53.


Munch, who had urged Wilson to purge the Danish terminology in scholarship.63

Munch’s concerns reflected a contemporary movement to minimise Danishinfluence in Norway. In 1814, Denmark had ceded Norway to Sweden, butNorway took the opportunity to declare independence. After a brief war, Norwayentered a union with Sweden which allowed the country to retain a separateconstitution and gain independence for many of its institutions. From this time itappears that attitudes towards the language, history, and culture of Norway becameincreasingly politicised as emphasis was laid on their distinct national character. Forexample, before 1814 the official language was called Danish. After 1814 the samelanguage was in Norway called Norwegian.64 It was in this environment of growingexpression of Norwegian nationalism in literature and the arts that Peter AndreasMunch and his fellow-historians worked. Munch was concerned with celebratingthe character and historical importance of Norway in his great work Det norske folkshistorie, published in 1863.65 His influence passed over into scholarship written inEnglish. Munch promoted the idea that the Viking-Age culture of Scandinavia wasmore deeply rooted in Norway than in Denmark.66 This complemented existingstereotypes of the Norwegians as more rural and backward-looking, but also moreindividualistic and liberty-loving, and freer of foreign influence, than their Danishneighbours.67 This desire to distinguish Norwegians from Danes in order to reclaimfor Norway a distinct historical impact in Britain and Ireland was expressed byother Scandinavian scholars whose work reached out to an English-speakingaudience. For example, the eminent historian Alexander Bugge complained in 1900that English students of the Viking-Age ‘confound Norwegians and Danes, withoutdistinguishing between the two nations’.68

Some regarded the shift in identifying a strong and distinctive Norwegiancultural impact in the Viking-Age as going too far.69 George Stephens, anineteenth-century English scholar who worked in Denmark, argued against the

63 Ibid.64 A. Elviken, ‘The genesis of Norwegian nationalism’, Journal of Modern History 3 (1931)

365–91, at p. 385.65 P.A. Munch, Det norske folks historie (8 vols, Oslo 1863); Elviken, ‘The genesis’, p. 387.66 P.A. Munch (transl. Sigurd Bernhard Hustvedt), Legends of Gods and Heroes (New York

1926), pp. xiii–xvi.67 Elviken, ‘The genesis’, pp. 371–4; Wawn, The Vikings, p. 9.68 Alexander Bugge, Contributions to the History of the Norsemen in Ireland,

Videnskabsselskabets Skrifter, II, Historisk-filosofisk Klasse, nos 4–6 (3 parts, Oslo 1900), I.3.69 Wawn, The Vikings, pp. 4, 148.


trend towards national separatism in historical interpretation, by contending thatScandinavian language and culture had been united across a large area.70 Stephens’sinsistence on the shared heritage of ‘Anglo-Scandic lands’ can be interpreted as areäction against Norwegian nationalism and German imperialism in the nineteenthcentury.71 German scholars had their own scholarly ideology (which in Stephens’sview downplayed the significance of Scandinavia), and Germany posed a tangiblepolitical threat to Denmark, a country whose culture Stephens had enthusiasticallyembraced.72 Stephens’s views did not gain widespread support.73 However, it isevident that scholarly disputes were steered (and sometimes buffeted) by the twinforces of national separatism and imperialist ideologies which had currency innineteenth- and early twentieth-century Europe.

The discussion thus far demonstrates that in the nineteenth century efforts weremade to distinguish vikings who were active in Scotland from those who were activein England. The former were identified as being predominantly from Norway andthe latter predominantly from Denmark. In Ireland the view that Danes andNorwegians had operated as separate groups was well established in nineteenth-century historiography. In 1860 John O’Donovan published an edition andtranslation of the text now called ‘The Fragmentary Annals of Ireland’. An eleventh-century saga embedded within that text identified two groups of vikings who werecalled ‘Dark foreigners’ and ‘Fair foreigners’ as Danes and Norwegians respectively.I have argued elsewhere that the eleventh-century saga-author was interpreting earlierchronicle-material, which he drew on for his account, in the light of eleventh-centuryrivalries between Danes and Norwegians.74 David Dumville has persuasively madethe case that references to ‘Dark’ and ‘Fair’ foreigners in the ninth century refer togroups under different leadership.75 The terms ‘Dark’ and ‘Fair’ may be interpreted to

70 Ibid., p. 219.71 George Stephens, The Old Northern Runic Monuments of Scandinavia and England, now first

collected and deciphered (4 vols, London 1866–1901), I, dedication, and I.xi–xii. Despiteemphasising the unity of the Old North, Stephens promoted the idea of regional dialects. Inparticular he insisted that English dialects were the ‘best key to the oldest Scandinavian folk-talks’rather than Icelandic, thus illustrating his own national bias.

72 Wawn, The Vikings, pp. 218–28.73 Ibid., p. 243.74 Downham, ‘The good’.75 Dumville, ‘Old Dubliners’; cf. Ailbhe Mac Shamhráin, The Vikings. An Illustrated History

(Dublin 2002), p. 48; Downham, Viking Kings, pp. xvi–xvii.


mean ‘New’ and ‘Old’ in a Gaelic context, rather than being ethnic signifiers.76

However, guided by the nationalist preöccupations of the age, scholars dealing withthe saga-element in ‘The Fragmentary Annals of Ireland’ readily accepted the divisionof vikings into national groups. Some, including James Henthorn Todd and CharlesHaliday, registered their frustration that other sources did not clearly distinguishbetween Danes and Norwegians in accounts of ninth-century events.77 Suchfrustrations were inevitable, I should argue, because ninth-century viking-groupswere not mono-ethnic and probably did not use national labels to identifythemselves.

It can be seen that in the nineteenth century there was a well establishedperception that viking-populations in Ireland were composed of different ethnicgroups which might work in competition with each other. Opinions were dividedon which of the Scandinavian peoples was dominant. In 1891, Heinrich Zimmerconsidered that the royal dynasty of Dublin and the dominant culture in the viking-colonies were Danish.78 Nevertheless a number of prominent Scandinavian scholarsregarded the Norwegians as the dominant cultural element in Ireland’s viking-ports.79

This was partly argued on geographical grounds: the Scottish islands, which werethought to have been settled from western Scandinavia, were but a short trip by seafrom Irish shores.80 Alexander Bugge’s Contributions to the History of the Norsemenin Ireland, published in 1900, presented a vigorous argument in opposition toZimmer that Dublin’s royal dynasty and much of the viking-population wereNorwegian in origin. Bugge’s work was influential, for after 1900 the majority ofscholars writing in English regarded Ireland’s viking-towns as being Norwegian incharacter, although debates continued about the origin of Dublin’s royal dynasty.

Links between Gaelic and Norwegian peoples were also highlighted in mediaeval

76 Smyth, ‘The Black Foreigners’; The Annals of Clonmacnoise, being Annals of Ireland from theEarliest Period to A.D. 1408, translated into English A.D. 1627 by Conell Mageohagan, ed. DenisMurphy (Dublin 1896), p. 148 (s.a. 922).

77 ‘It is to be regretted that our author [of Cogadh Gaedhel re Gallaibh] does not always veryclearly distinguish between them in his descriptions of their devastations in Ireland. We cannoteven be sure that the name Dane is not sometimes given to the Norwegians’: Cogadh Gaedhel reGallaibh. The War of the Gaedhil with the Gaill, or, The Invasions of Ireland by the Danes andOther Norsemen, ed. & transl. James Henthorn Todd (London 1867), p. xxxi. Charles Haliday,The Scandinavian Kingdom of Dublin (2nd edn, Dublin 1884), p. 15: ‘But whether these invaderswere Norwegians, Danes, Swedes, or Jutes, it is difficult to determine’.

78 H. Zimmer, ‘Keltische Beiträge, III: Weitere nordgermanische Einflü� sse in der ältestenÜberlieferung der irischen Heldensage’, Zeitschrift fü� r deutsches Alterthum 35 (1891) 1–176.

79 Worsaae, An Account, pp. xxiii, 297–353; Chronica regum Manniae et Insularum, edd. &transl. P.A. Munch & A. Goss (2 vols, Douglas 1874), I.1–2.


80 Ibid.; cf. Worsaae, An Account, p. xvii.Icelandic historiography. The author of Landnámabók asserted that some of the earlysettlers came to Iceland from the Hebrides and Ireland.81 The Gaelic contribution tomediaeval Icelandic culture has been much explored since the late nineteenthcentury.82 Nevertheless, not all scholars saw Gaelic influence in positive terms. It issometimes possible to see the anti-Irish sentiments which circulated in the latenineteenth and early twentieth centuries surfacing in the scholarship of the times.83

The picture built up so far is that historians in the nineteenth century sought todefine the impact of different Scandinavian groups on Britain and Ireland. TheNorwegians were eventually seen to be the dominant cultural influence in Scotlandand in Ireland, while Danish links with England had long been recognised. Thisprepared the ground for a view which developed in English historiography of rivalviking-factions in Britain which were aligned along ethnic lines, that is the ‘Anglo-Danes’ on one side and the ‘Hiberno-Norwegians’ on the other.

PLACE-NAME EVIDENCE

There is common sense to the argument that travellers from what is now Norwaywould tend to settle in the north and west of Britain and that settlers from what isnow Denmark would settle in the east. There are however some problems in theway in which place-names have been interpreted to draw a sharp distinctionbetween Scandinavian settlements in the west and east of England.84 This is evidentin maps which oversimplify the work of onomastic specialists by showing easternand western Scandinavian settlements in different colours (in particular those with

81 Íslendingabók-Landnámabók, ed. Jakob Benediktsson (Reykjavík 1968).82 Sturlunga Saga including the Islendinga Saga of Lawman Sturla Thordsson and Other Works,

ed. & transl. Gudbrand Vigfusson (2 vols, Oxford 1878), I.xxvi; Israel Gollancz, Hamlet inIceland (London 1898), pp. lii–lv; W. Faraday, ‘On the question of Irish influence on earlyIcelandic literature’, Memoirs and Proceedings of The Manchester Literary and Philosophical Society44 (1899/1900) 1–22.

83 G.T. Flom, review of Islands Kultur ved Aarhundredskiftet 1900 by Valty� r Guðmundsson,The American Anthropologist, new series, 6 (1904) 339–41. Flom described Valty�r’s categorisationof those deemed to be descended from ‘thralls of a non-Aryan race’: ‘for the greater part, perhaps,… the Celt’ (p. 341) who among other things ‘are generally melancholy … characterized by verystrong feelings, are constant, oppose change … pessimistic, easily discouraged, suspicious, jealous… live for the moment, cannot plan for the future’ (p. 339), in contrast to the noble qualitiesascribed to those of Norwegian descent! See also L.P. Curtis, jr, Anglo-Saxons and Celts. A Study ofAnti-Irish Prejudice in Victorian England (Bridgeport, CT 1968), pp. 74–89.

84 Hadley, The Vikings in England, p. 101.


a thin intervening strip of no-man’s land) as if Norwegian and Danish populationslived in geographically separated zones with little interaction.85 The interpretationof ‘The Anglo-Saxon Chronicle’ which gained ground in the early twentiethcentury, that there were two rival ethnic camps of Scandinavian settlers in Britain,may have influenced historians’ interpretations of place-name evidence.J.J.A. Worsaae was one of the earliest scholars to highlight the differences betweenScandinavian place-names in eastern and western England. In 1852 he wrote:86

the names ending in by, thorpe, toft, beck, næs, and ey, appear chiefly in the flat midlandcounties of England; whereas farther towards the north, in the more mountainous districts,these terminations mostly give place to those in thwaite, and more particularly to those indale, force, tarn, fell and haugh. The difference, however, is scarcely founded on the naturalcharacter of the country alone; it may have arisen from the different descent of theinhabitants …. Exactly similar names are met with to this day in the mountains of Norway;whilst they are less common, or altogether wanting, in the flat country of Denmark …Norwegians … appear to have betaken themselves chiefly to the most northern andmountainous districts, which lay not only nearest to them, but which in character mostresembled their own country.

The geographically determined argument makes sense, that Scandinavian names inmountainous areas of England are akin to the names of mountainous areas inScandinavia and the names of lowland-areas are akin to the names found in thelowlands of Scandinavia. However, the ethnically determined view that Norwegiansshunned lowland-areas which would have been richer agriculturally, as they feltdrawn to a harsher landscape which looked more familiar, does not make as muchsense.87 The classification of place-names into ‘Norwegian’ and ‘Danish’ elementstherefore risks being a division between names for ‘upland’-features and names for‘lowland’-features which reflect differences in geography between northwestern andeastern England.

Worsaae went on to identify -by�r or -by�� (‘farm’/‘settlement’) as a Danish place-

85 For a selection online, see: http://medieval.ucdavis.edu/20A/Viking.England.jpg;http://www.trin.cam.ac.uk/sdk13/RPMaps/MapPolDev.jpg; http://members.aol.com/scothist/viks/v0311.jpg; http://webspace.ship.edu/cgboer/evolenglish.html; http://research.uvsc.edu/mcdonald/Anglo-Saxon/wife’slament/wifema5.gif; http://www.variantbank.org/results/rules/v/viking_ra.gif; http://members.lycos.nl/vikings/vikingjourney.html (all accessed 01/08/08).

86 Worsaae, An Account, pp. 72–3.87 If one can imagine an attitude towards landscape as a commodity which requires heavy

manual work (and is not simply cast in aesthetic terms), those who had farmed in a harshlandscape would especially prize good flat land and might regard it as more visually attractive. I should thank my late great-grandfather, Edward Inman, a Westmorland-farmer, for this insight.


name element.88 He noted that these names appear more frequently in northeasterncounties of England than they do in the north-west. Recent research by GillianFellows-Jensen has reïnforced the view that northwestern names ending in -by�� weretransmitted from the east of England, taking into account some complexities of theevidence to draw the following conclusion: 89

there was an anti-clockwise movement from the northern Danelaw across the Pennines anddown the Eden valley to Carlisle, spreading northwards from there into easternDumfriesshire and trickling along the coast to Galloway, and southwards from Carlislealong the coastal plain of Cumberland, across the Irish Sea to Man, and finally perhaps backacross the Irish Sea to Wirral and south-west Lancashire.

However, Fellows-Jensen’s view that all British -by�� names were originallydisseminated from the Northern Danelaw has been challenged by Alison Grant,who has argued on linguistic grounds that -by�� names of Ayrshire and the Hebrideswere introduced in a Gaelic-Scandinavian milieu. In other words, Grant has put thecase that -by�� names were transmitted from the west as well as from the east.90

Grant’s argument may also have implications for some of the -by�� names ofCumbria, Lancashire, and the Isle of Man.91

The other elements discussed by Worsaae were -thveit (‘clearing’) and -thorp(‘secondary settlement’). Worsaae considered -thveit to be a Danish place-nameelement because it was commoner in northeastern England than in Scotland ornorthwestern England. His interpretation was soon challenged by Robert Ferguson,whose book The Northmen in Cumberland and Westmoreland was published in1856. Ferguson identified the work of Worsaae as a springboard to his ownresearches, but he was concerned to highlight the links between the English LakeDistrict and Norway.92 Ferguson noted the frequency of -thveit names inCumberland, suggesting that such names were less frequent in the east becauseareas of Yorkshire and Lincolnshire would have already been cleared before vikingsarrived.93 He therefore regarded the distribution of -thveit names as being

88 Worsaae, An Account, pp. 75–6.89 G. Fellows-Jensen, ‘Scandinavian place-names of the Irish Sea Province’, in Viking Treasure from

the North West. The Cuerdale Hoard in Context, ed. James Graham-Campbell (Liverpool1992), pp. 31–42, at p. 36.

90 A. Grant, ‘The origin of the Ayrshire by names’, in Cultural Contacts in the North AtlanticRegion: the Evidence of Names, edd. Peder Gammeltoft et al. (Lerwick 2005), pp. 127–40.

91 Diana Whaley, A Dictionary of Lake District Place-names (Nottingham 2006), p. 390.92 Robert Ferguson, The Northmen in Cumberland and Westmoreland (London 1856), pp. i–ii.93 Ibid., pp. 36–7.


influenced by geographical factors, but he also linked it with settlement from anarea of southwestern Norway where this element was common. Nevertheless,-thveit is also found in eastern England. It occurs in seven settlement-names inNorfolk, and, as Fellows-Jensen has pointed out, the adoption of the word thwaiteinto northern dialects of English means that a number of -thveit names in Cumbria(Cumberland, Westmorland, and Lancashire North-of-the-Sands) may postdate theninth and tenth centuries.94 In sum, the distribution of -thveit names (like -by��names) does not always conform with a simple east/west division of English place-names.

The Old-Scandinavian place-name element -thorp is found frequently ineastern England but rarely in the west. It has the same meaning as Old-English thropwhich is also focused in the Eastern counties but which is much less common. It ispossible that some -thorp names were adopted from Old-English -throp by speakersof Scandinavian or those speaking a Scandinavianised dialect of English.95 The wordthorp continued to be used in Middle English, and some names seem to have beenformed after the Viking-Age.96 Fellows-Jensen has noted that -thorp names appearfrequently in eastern Norway, which challenges the view that the names in Englandwere all coined by settlers from Denmark.97 Furthermore, the upland-distributionof -thveit names in northwestern England is analogous to where -thorp names lie ineastern England, except that in Cumbria these areas would have been more thicklywooded: ‘when the Vikings were naming dependent settlements in this part ofEngland, they referred to them quite naturally by the term þveit “clearing”’.98 Thesearguments suggest that, despite the marked east/west pattern of division in thedistribution of -thveit and -thorp, it would be unwise to conclude that the formerwere all coined by Norwegians and the latter were all named by Danes.

As part of a debate about the level of dialectal difference between WestScandinavian (Old Icelandic/Old Norwegian) language and East Scandinavian

94 K.I. Sandred, ‘Language contact in East Anglia: some observations of Scandinavian place-names in thwaite in Norfolk’, in Proceedings of the Seventeenth International Congress of OnomasticSciences, ed. Eeva Maria Närhi (Helsinki 1990), pp. 1–8; G. Fellows-Jensen, ‘Little Thwaite, whomade thee?’, in Proceedings of the Nineteenth International Congress of Onomastic Sciences, ed.W.F.H. Nicolaisen (3 vols, Aberdeen 1998), II.101–6; G. Fellows-Jensen, ‘Vikings in the BritishIsles: the place-name evidence’, Acta Archaeologica 71 (2000) 135–46, at p. 143.

95 G. Fellows-Jensen, ‘Place-names in -þorp: in retrospect and in turmoil’, Nomina 15 (1991/2)35–51, at pp. 36, 40.

96 Ibid., pp. 39, 42.97 Ibid., p. 41.98 Ibid.


(Old Swedish/OldDanish) language, Paul Bibire has questioned the analysis ofScandinavian place-names in Britain.99 For example,100

the form Botham and the loan into English, booth, have been used as evidence for EastNorse, and more particularly Danish, settlement in Yorkshire …. It has been alleged thatthere is a dialectal distribution of Old West Norse u, Old East Norse u in this root … but inactual fact, Old West Norse has both u and o… A geographical distribution of u and o inthis root and words derived from it must therefore be regarded as very questionable, and it isfar from certain that all varieties of Norse did not have both vowels in the ninth century.

There is no consensus as to when dialectal differences between West Scandinavianand East Scandinavian became pronounced: opinions range between the sixth andthe eleventh century.101 It is therefore problematic to apply these arguments toScandinavian place-names in an Insular context, where local languages (Gaelic,Brittonic, English) and dialects will also have impacted on the evidence.

In sum, one would expect that Scandinavian place-names in the east ofEngland might show greater influence of East-Scandinavian naming habits and thatnorthwestern England might show greater influence from West-Scandinaviannaming habits. However, the distinctions between eastern and western England havealso been influenced by local geography. Furthermore, the evidence is complex interms of the origin of settlers, the chronology of settlement, and the chronology ofname-formation using onomastic elements of Scandinavian origin. While modernonomastic studies show sensitivity to all these issues, the differences between easternand western England have been exaggerated and sometimes continue to beoverplayed in modern accounts and associated maps. The perception that ‘TheAnglo-Saxon Chronicle’ treats ‘Norwegians’ and ‘Danes’ as distinct groups hasperhaps coloured analyses.

On this basis it had been argued that Denby and Normanby place-namesdistinguished between separate settlements of ‘Danes’ and ‘Norwegians’ in the

99 The West-Scandinavian language is also called West Nordic or West Scandinavian in scholarlyusage, and similarly for East Scandinavian.

100 Bibire, ‘North Sea language contacts’, pp. 100–1; G. Fellows-Jensen, ‘To divide the Danesfrom the Norwegians: on Scandinavian settlement in the British Isles’, Nomina 11 (1987) 35–60,at p. 53.

101 M. Schulte, ‘Language contact in the period between Ancient Nordic and Old Nordic’, inThe Nordic Languages. An International Handbook of the History of the North Germanic Languages,I, edd. Oskar Bandle et al. (Berlin 2002), pp. 882–95 (§88); H. Perriden, ‘Dialects and writtenlanguage in Old Nordic, II: Old Danish and Old Swedish’, ibid., pp. 1018–28 (§114). I am gratefulto Stefan Brink for providing bibliographical guidance on this matter.


ninth and early tenth centuries.102 However, this seems doubtful. All such namesare found in eastern counties rather than across the whole area of Scandinaviansettlement in England.103 As I have argued, during the ninth and early tenthcenturies a distinction was not made between these groups in the Englishlanguage.104 Both toponyms might refer to Scandinavians in a general way, or theircoining may postdate the mid-tenth century.105

One cannot doubt that in the Viking-Age there were observable differencesbetween vikings raised in Gaelic-speaking areas and those who had lived inEngland. These differences would presumably have increased over time asScandinavian settlers and their descendants intermingled with people from thehost-culture. The immigration of vikings from Gaelic-speaking areas into Englandmay have led to the coining of new names such as Ireby (‘farm/settlement of theIrish’), discussed by Mary Higham.106 Differences may also be noted in areas ofGaelic-Scandinavian settlement, with the use of inversion-compounds,combinations of Gaelic personal names with Scandinavian place-name elements,and the use of Gaelic loanwords into Scandinavian, of which the most notableexample is -erg (from Old-Gaelic áirge, Modern Scottish-Gaelic airigh).107 It canbe argued that vikings could be identified by their place of origin, but I do not

102 Worsaae, An Account, p. 73; Mawer, ‘The Redemption’, pp. 556–7.103 ‘The Key to English Place-Names’ database lists two ‘Danby’ names in North Yorkshire, a

‘Denby’ in West Yorkshire and another in Derbyshire. Two ‘Normanby’ names are listed forNorth Yorkshire and two in Lincolnshire. In addition there is a ‘Normancross’ inHuntingdonshire and ten ‘Normanton’ names distributed as follows: Derbyshire (3),Nottinghamshire (3), Leicestershire (1), West Yorkshire (1), Rutland (1), Lincolnshire (1). Seehttp://www.nottingham.ac.uk/english/ins/kepn/results_search.php (accessed 01/08/08).104 It might also be questioned whether all the Normanby and Normanton names were derivedfrom the Old-English plural Norðmenn, or whether some are derived from the personal name Norðmaðr: Lena Peterson, Nordiskt runnamnslexikon (Uppsala 2002), p. 149,http://www.sofi.se/servlet/GetDoc?meta_id=1472 (accessed 01/08/08). Cf. Mawer, ‘TheRedemption’, p. 556; Whaley, A Dictionary, p. 256, s.n. Ormathwaite.

105 For later Scandinavian impact on English place-names, see G. Fellows-Jensen, ‘Danish place-names and personal names in England: the influence of Cnut?’, in The Reign of Cnut, King ofEngland, Denmark and Norway, ed. Alexander R. Rumble (London 1994), pp. 124–40.

106 M.C. Higham, ‘Scandinavian settlement in north-west England, with a special study of Irebynames’, in Scandinavian Settlement in Northern Britain, ed. Barbara E. Crawford (London 1995),pp. 195–205.

107 Fellows-Jensen, ‘Vikings in the British Isles’, pp. 141–2; Higham, ‘Scandinavian settlement’,

pp. 199–205; Alfred P. Smyth, Scandinavian York and Dublin. The History and Archaeology of Tworelated Viking Kingdoms (2 vols, Dublin 1975/9), I.80.


consider that a political distinction was maintained throughout the ninth and tenthcenturies between people whose ancestors hailed from the areas of Denmark andNorway. Rather, viking-groups were primarily bound together by the identity of their leaders and by the bases and areas from which they operated in an Insular context.

ETHNIC STEREOTYPING

The idea that conflict sometimes characterised relations between ‘Anglo-Danish’and ‘Hiberno-Norwegian’ contingents in Viking-Age England was developed intwentieth-century historiography. The rivalry was seen not merely as political butalso as a competition between peoples of contrasting character and world-view. Ithas been concluded by various scholars that ‘Hiberno-Norwegian’ and‘Norwegian’ colonists were more staunchly heathen, more adventurous, moreviolent, and more disorganised than their ‘Anglo-Danish’ rivals. This rhetoric,which sometimes distinguishes relatively domesticated ‘Anglo-Danes’ from thewilder ‘Hiberno-Norwegians’, calls to mind standard stereotypes used to contrastpeoples deemed to be at the core and the margins.108

What follows is not a comprehensive survey of the literature. However, each ofthe stereotypes listed above is illustrated by quotations taken from works which Ihave recently browsed. Taken together, they suggest that stereotyped perceptions of‘Norwegians’ and ‘Hiberno-Norwegians’ in opposition to ‘Danes’ and ‘Anglo-Danes’ have been and remain fairly pervasive in the historical literature.The first stereotype, that ‘Norwegians’ were more adventurous than ‘Danes’ inthe Viking-Age, was linked with the perception that their colonies weregeographically distinct. This notion tends to be found in older historical narratives,probably because recent onomastic research has effectively challenged the idea that‘Norwegians’ and ‘Danes’ were ethnically and geographically separated in England.According to Edward Laborde in 1936:109

The raids on the west coast of England were exclusively Norwegian. Even when attackingthe southeastern shores, they used the same route .… Joint enterprises in later timesconsisting of Norsemen and Danes sometimes used the North Sea route, and sometimespurely Norse invasions like that of Harald Hardrada also went that way. But Danishwickings never went ‘round about’.

108 E.D. Snyder, ‘The Wild Irish: a study of some English satires against the Irish, Scots, andWelsh’, Modern Philology 17 (1919-1920) 687–725; Thomas Hylland Eriksen, ‘Images of theneighbour: reciprocal national stereotypes in Scandinavia’, http://folk.uio.no/geirthe/Scandinavian_images.html (1997), accessed 15/02/2008.

109 E.D. Laborde, Byrhtnoth and Maldon (London [1936]), p. 47. Here ‘Norse(man)’ means‘Norwegian’.


Thus Laborde expressed an extreme view that Danes never sailed around Scotland.This echoes to some extent the description offered by Eleanor Hull:110

The Norse were hardy seafarers who pushed out north-west to the shores of Greenland,Iceland, and North Britain, and thence made their way down the western coasts of Scotlandto Ireland; the Danes, who were not naturally a sea-loving nation, were inclined to hug theshores.

In terms of organisation, the ‘Norwegians’ have tended to be regarded as morechaotic. Thus F.T. Wainwright wrote that ‘they lacked the military organisationwhich characterised the Danish settlements of Eastern England’.111 This may becompared with a comment made by Jean Renaud: ‘L’expansion danoise,contrairement à celle des Suédois et des Norvégiens qui était souvent le fait degroupes d’individus isolés, prit très vite un caractère massif et fortement organisé’.112

The stereotype that ‘Norwegians’ were less organised can also be linked withperceptions of the more violent character of the ‘Hiberno-Norwegians’ in contrastto ‘Anglo-Danes’. For example, T.D. Kendrick wrote:

in 918 … a roving viking of the Dublin house, Ragnvald, came over from Ireland and seizedthe throne of York … the unhappy province was thrown into chaos by the attempt toimpose heathen Irish-Norwegian government upon the Christian Danes of Deira ….

And of a later king of Rögnvaldr’s family he observed that ‘Olaf and hisNorwegians were athirst for conquest’.113

The distinctiveness and aggressiveness of the ‘Hiberno-Norwegian’ iscommented on by a wide range of authors. One example is provided by the recentbook Viking Empires written by Angelo Forte, Richard Oram, and FrederikPedersen:114 ‘Christianised York had become … an accepted part of mainlandpolitical society …. This position stands in sharp contrast to that of the Norse ofDublin, whose aggressive paganism continued to set them apart ….’

110 Eleanor Hull, A History of Ireland and her People (2 vols, London 1931), I.91 (§3), accessedonline at http://gutenberg.net.au/ebooks08/0800111h.html#ch1-3 (15/02/2008). Here again‘Norse’ means ‘Norwegian’.

111 F.T. Wainwright, Scandinavian England. Collected Papers (Chichester 1975), p. 256; cf. p.334.

112 Renaud, Les Vikings, pp. 8–9.113 T.D. Kendrick, A History of the Vikings (London 1930), pp. 251, 254.114 Angelo Forte et al., Viking Empires (Cambridge 2005), p. 78; cf. p. 103.


The association of Scandinavian traditional religion and violence is foundelsewhere. Recently, Alex Woolf has referred to ‘Christian Anglo-Danes, settled ineastern England, and heathen Hiberno-Norse from the Irish Sea province’, arguingthat in the mid-tenth century ‘Cumbrians … were perhaps particularly exposed toaggression from the pagans of the Irish Sea’.115 The view of ‘Hiberno-Norwegians’as violent heathens bears on issues of cultural assimilation. Scandinavian settlers inthe east of England were more likely to live in rural settlements than were the port-dwelling vikings of Ireland. In such circumstances the settlers in eastern Englandmay have more rapidly assimilated to the local population. Nevertheless, the viewthat vikings in Ireland had little contact with their Irish neighbours, and inparticular that they were relatively untouched by christianity in that island, is highlyquestionable.116 The contrasts have sometimes been overplayed.

The image projected by Allen Mawer’s translation of the poem in annal 942 of‘The Anglo-Saxon Chronicle’ was that a Danish population in the East Midlands ofEngland was held down under the force of heathen Norwegians. It has beenassumed that a similar situation prevailed at York. This view is exemplified in theremarks of Katherine Holman with reference to King Rögnvaldr and his successorswho came from Dublin:117

the new rulers of York were pagan conquerors who imposed themselves upon a Christian,Anglo-Scandinavian population. The new political leaders seem to have made no attempt toestablish any permanent roots, but were content to simply milk York and its hinterland forwealth and power, and to use it as a power base for further expansion of their control intothe surrounding areas. The Danes who had colonized north-eastern England in the ninthcentury had long since been converted, had settled down to farming and trading, and werenow an integral part of the political and social structures of the region. They were as keen torid York of its new Norse kings as the kings of Wessex were.

115 Alex Woolf, From Pictland to Alba, 789–1070 (Edinburgh 2007), pp. 182, 184–5.116 Downham, ‘Religious and cultural boundaries’; C. Downham, ‘Viking non-urban

settlements in Ireland to A.D. 1014’, in Proceedings of the Conference on Irish-Norse Relations,800–1200, held in Oslo on 5 November, 2005, ed. Jón Viðar Sigurðsson & T. Bolton (Leiden,forthcoming). Theseminal article on economic contacts is by P.F. Wallace, ‘The economy andcommerce of Viking Age Dublin’, in Untersuchungen zu Handel und Verkehr der vor- undfrü� hgeschichtlichen Zeit in Mittel- und Nordeuropa, Teil IV: Der Handel der Karolinger- undWikingerzeit, edd. Klaus Dü�wel et al. (Göttingen 1987), pp. 200–45. The forthcoming book byMary Valante, The Vikings in Ireland. Settlement, Trade and Urbanization (Dublin 2008), shouldcast further light on economic contacts between Irish and vikings.

117 Katherine Holman, The Northern Conquest. Vikings in Britain and Ireland (Oxford 2007), p. 101. Again, ‘Norse’ is used to mean Norwegian.


In the next section of this article, the portrayal of kings from Dublin as driven byshort-term concerns and ruling by force alone, with little consensus or legitimacy, isquestioned. Although some scholars have avoided applying a theory of ethnicrivalry to viking-politics, it would discredit the evidence to rush to the otherextreme and to assume that everyone of Scandinavian heritage pulled together. Onthe one hand it can be said that factional rivalry was intense, even within the kin-group of ‘the dynasty of Ívarr’ who ruled York. On the other hand, there is littleevidence that the battles which raged between viking-factions were ever drawn upprimarily along ethnic lines.

DYNASTIC CONTINUITY OR ETHNIC CONFLICT?

There is ample evidence that vikings controlled York intermittently from 867 to954. It has generally been recognised that a number of these rulers hailed fromIreland. These include Rögnvaldr, grandson of Ívarr (918–21); Sigtryggr grandsonof Ívarr (921–7); Óláfr and Rögnvaldr, sons of Guðröðr (939×944); and ÓláfrSigtryggsson (941×952). All belonged to the dynasty of Ívarr which dominated theviking-ports of Ireland during the tenth century.

Historians in Ireland have long recognised that a comparison of evidence inIrish and English chronicles suggests that a link between vikings in Northumbriaand Ireland existed before 918. Most prominent among the early exponents of thisview were Charles Haliday and James Henthorn Todd, working in the mid-nineteenth century.118 Both scholars studied in detail the mediaeval sources fromboth sides of the Irish Sea. Historians in Britain have tended to be more cautious inadmitting these links.119 A view was developed in English historiography thatRögnvaldr’s accession to the throne of York after the battle of Corbridge (918) wasan innovation. It was regarded as the imposition of foreign power by ‘Hiberno-Norwegians’ enforced by brutal conquest over the resident ‘Anglo-Danes’.120 Alfred

118 Haliday, The Scandinavian Kingdom, pp. 24–56, 81; Cogadh, ed. & transl. Todd, pp. 263–77.

119 A. Campbell, ‘Two notes on the Norse kingdom in Northumbria’, English Historical Review57 (1942) 85–97; F.M. Stenton, Anglo-Saxon England (3rd edn, Oxford 1971), p. 250. Cf.Smyth, Scandinavian York, I.28; Mawer, ‘The Scandinavian kingdom’, pp. 43–50; W.S. Angus,‘Christianity as a political force in Northumbria in the Danish and Norse periods’, in The FourthViking Congress, York, August 1961, ed. Alan Small (Edinburgh 1965), pp. 142–64, at 144, 164.

120 F.T. Wainwright, ‘The submission to Edward the Elder’, History, new series, 37 (1952)114–30, at pp. 116, 118, 123–4, 129.


Smyth’s volumes on the history of York and Dublin in the later 1970s highlightedthe evidence that two earlier kings from Ireland, Ívarr and Hálfdan, had ruled atYork from 866 to 876.121 However, Smyth and others have argued that powerslipped from the hands of Ívarr’s kin-group following the death of Hálfdan. Thesubsequent history of York was regarded as a battle between two competing camps,the ‘Anglo-Danes’ who forwarded their own candidates for kingship, and the‘Hiberno-Norwegian’ dynasty of Ívarr who sought to regain the power won bytheir glorious ancestor.122

The emerging idea that Rögnvaldr may have had a dynastic claim to the York-kingship in 918 did not therefore dampen perceptions that his rule was imposed byforce over a reluctant ‘Anglo-Danish’ population. In a recent article, DavidDumville has challenged the perception that there was a rapid interchange betweenkings representing ‘Hiberno-Norwegian’ and ‘Anglo-Danish’ factions. He hasargued that the influence of the dynasty of Ívarr was more continuous than hasoften been perceived. After the death of Hálfdan, brother of Ívarr, it is unclear whosucceeded to the throne of York. The next king who can be identified is Guðröðr,who died in 895.123 Historia de Sancto Cuthberto, which may date from the eleventhcentury, identifies Guðröðr as a slave purchased by the abbot of Carlisle who with divine guidance got him elected to the kingship of York. Thishagiographical account cannot be regarded as factually reliable. Dumville haspointed out that the name Guðröðr is common among the descendants of Ívarr,and that he may have belonged to that family.124 Adam of Bremen, writing in thelate eleventh century, linked Guðröðr to this dynasty, although his information wasdrawn from lost Gesta Anglorum of uncertain value.125 Dumville has also pointedout that a trio of kings who led a Northumbrian army in 910 (‘Eowils’, ‘Halfdan’,and ‘Ivar’) all bore names which indicate their association with the dynasty ofÍvarr.126 Arguments have been put forward by other historians to link Sigfröðr who

121 Smyth, Scandinavian York, I.29, 79.122 Ibid., I.102–3, 107–9.123 Æthelweard, Chronicon, IV.3 (The Chronicle, ed. & transl. Campbell, p. 51).124 Dumville, ‘Old Dubliners’, pp. 87–8.125 Quellen, edd. & transl. Trillmich & Buchner, p. 258, and Adam of Bremen, transl. Tschan,

pp. 70–1 (II.25); Downham, Viking Kings, pp. 75–8.126 Annal 910: ASC.A, ed. Bately, p. 64; ASC.B, ed. Taylor, p. 47 (s.a. <911>); ASC.C, ed.

O’Keeffe, p. 73(s.a. 911); ASC.D, ed. Cubbin, p. 38 (s.a. 911); Æthelweard, Chronicon, IV.4 (TheChronicle, ed. & transl. Campbell, p. 53); Dumville, ‘Old Dubliners’, pp. 88–9. (Eowils appears asEcwils in A and Eowilisc in D.) Texts BC have a list of twelve vikings slain; D has only six (nos 1~5and 11).


reigned around 895, and Eiríkr who reigned 946×954, with the same dynasty(although, in the latter case, this is very tentative).127 Overall it seems evident thatthe viking-kings of York were largely members of the dynasty of Ívarr. A case maybe put that some kings of uncertain origin also belonged to that family.

There is very little evidence of organised opposition by an ‘Anglo-Danish’faction raising its own candidates. Members of the West-Saxon royal dynasty ruledYork intermittently from about 900 until the collapse of the viking-kingdom in954.128 They seem to have benefited from struggles within the dynasty of Ívarr inorder to come to power, although there must have been a local support-base foreach party to help effect these changes of government.129 While there was, atdifferent times, some support for southern rule in Northumbria, the rule of thedynasty of Ívarr could be regarded as legitimate, long-term, and backed by a largebody of local supporters.130

The received wisdom that ‘Anglo-Danes’ and ‘Hiberno-Norwegians’ wereopposed factions in Viking-Age England can be called into question. This gives avery simplistic picture of a complex political situation. ‘The Anglo-SaxonChronicle’ does not demonstrate that ‘Danes’ and ‘Norwegians’ were distinctfactions in the ninth and tenth centuries, and neither does the place-name evidence.The desire to distinguish ‘Danish’ and ‘Norwegian’ influence in Viking-AgeEngland can be traced back to the political and cultural agenda of various

127 Smyth, Scandinavian York, I.33–7, 46; C. Downham, ‘Eric Bloodaxe − axed? The mysteryof the last Scandinavian [rectius viking] king of York’, Mediaeval Scandinavia 14 (2004) 51–77 seechapter 9, below; Downham, Viking Kings, pp. 79, 115–20.

128 Æthelwold (ca 900–2); Æthelstan (927–39); Edmund (944–6); Eadred (946×948, 954–5).129 Downham, Viking Kings, pp. 80–2, 88–9, 95, 99–105, 111–13.130 Evidence for concern by members of the dynasty of Ívarr to govern in England and not

merely to plunder may be seen in coinage: M. Blackburn, ‘Expansion and control:aspects of Anglo-Scandinavian minting south of the Humber’, in Vikings and the Danelaw. SelectPapers from the Proceedings of the Thirteenth Viking Congress, edd. James Graham-Campbell et al.(Oxford 2001), pp. 125–42, and M. Blackburn, ‘The coinage of Scandinavian York’, in Anglo-Scandinavian York, VIII, fasc.4, Aspects of Anglo-Scandinavian York, ed. R.A. Hall. (York 2005), pp.325–49. With regard to that dynasty’s accommodation with the Church, see Hadley, The Vikings inEngland, pp. 192–236. For the continuity of local administrative units and assembly-sites, see D.Roffe, ‘Hundreds and wapentakes’, in The Lincolnshire Domesday, edd. Ann Williams & G.H. Martin(London 1992), pp. 32–42; A. Meaney, ‘Gazetteer of hundred and wapentake meeting-places of theCambridgeshire region’, Proceedings of The Cambridge Antiquarian Society 82 (1993) 67–92; S. Turner, ‘Aspects of the development of public assembly in the Danelaw’, Assemblage 5 (2000),available online at http://ads.ahds.ac.uk/catalogue/adsdata/assemblage/html/5/turner.html(accessed 08/02/08).


Scandinavian and Insular scholars working in the nineteenth and early twentiethcenturies. These views were elaborated into theories of ‘Anglo-Danish’ and‘Hiberno-Norwegian’ rivalry.

Although it can be said that vikings from Ireland would have differed culturallyin some respects from vikings in England, evidence of ethnic conflict betweenviking-groups is ambiguous and deficient. Rather than being fly-by-nights, thedynasty of Ívarr supplied kings for both Dublin and York for a sustained period.These rulers would have been attuned to the needs and concerns of theirconstituencies on both sides of the Irish Sea and familiar with the cultures of both.It is possible to envisage a ruling elite who felt equally at home with the variousviking-cultures of Ireland and Britain, rather than conforming with the labels of‘Hiberno-Norwegian’ or ‘Anglo-Dane’.

The contrasting ethnic stereotypes applied to ‘Hiberno-Norwegians’ and‘Anglo-Danes’ reflect modern categories of thought which hinder rather than assistour understanding of the Viking-Age. They take their origin in national boundarieswhich were not embedded in the ninth and earlier tenth centuries, and they reflectattitudes to core and periphery as defined by later mediaeval and modern politics.In discarding these stereotypes, we may find that new possibilities emerge forinterpreting aspects of the Viking-Age in England and beyond.


III

VIKINGS AS SEENFROM IRELAND

4Irish chronicles as a source for rivalry between vikings,

A.D. 795–1014

IRISH chronicles give insights into rivalry between viking-dynasties andviking-settlements. This information may have a comparative value forunderstanding and interpreting viking-activity in other areas, notably Scotland,where there is less written evidence from which to work. Only a skeletal accountof these events will be presented here.

Irish chronicles are the most essential sources for the study of viking-activityin Ireland. Ten principal Irish chronicles have survived which deal with the periodin question (‘The Annals of Boyle’, ‘The Annals of Clonmacnoise’, ‘The Annalsof Inisfallen’, ‘The Annals from “The Book of Leinster”’, ‘The Annals of Roscrea’,‘The Annals of The Four Masters’, ‘The Annals of Tigernach’, ‘The Annals ofUlster’, Chronicum Scotorum, and ‘The Fragmentary Annals of Ireland’). Theseare compilations ranging in date from the eleventh to the seventeenth centurythough based for the most part on contemporary records.1 In some chroniclespseudohistorical saga-narratives have been added, as in ‘The Fragmentary Annalsof Ireland’, ‘The Annals of Clonmacnoise’, and ‘The Annals of The Four Masters’.2

These interpolations can often be identified on stylistic grounds. Apart from suchadditions, mediaeval Irish chronicles are considered to be largely accurate records,albeit partisan in their presentation of events.

While accepting the accuracy of these sources, Colmán Etchingham has raisedquestions about the usefulness of Irish chronicles for providing an overview ofviking-activity in Ireland.3 He has drawn attention to the geographical bias of

1Kathleen Hughes, Early Christian Ireland: Introduction to the Sources (London 1972), pp. 97–159; Gearóid Mac Niocaill, The Medieval Irish Annals (Dublin 1975); Kathryn Grabowski &D. Dumville, Chronicles and Annals of Mediaeval Ireland and Wales. The Clonmacnoise-group Texts(Woodbridge 1984); D.N. Dumville, ‘A millennium of Gaelic chronicling’, The Medieval Chronicle1 (1999) 103–15, at p. 108.

2Ibid., p. 104; D. Ó Corráin, ‘Muirchertach Mac Lochlainn and the Circuit of Ireland ’, inSeanchas. Studies in Early and Medieval Irish Archaeology, History and Literature in Honour of FrancisJ. Byrne, ed. Alfred P. Smyth (Dublin 2000), pp. 238–50.

3Colmán Etchingham, Viking Raids on Irish Church Settlements in the Ninth Century: aReconsideration of the Annals (Maynooth 1996), has included in his analysis the events recorded inthe twelfth-century pseudohistorical text Cogadh Gaedhel re Gallaibh. The War of the Gaedhil withthe Gaill, or The Invasions of Ireland by the Danes and Other Norsemen, ed. & transl. James HenthornTodd (London 1867). For commentary on that text, see R.H. Leech, ‘Cogadh Gaedhel re Gallaibhand The Annals of Inisfallen’, North Munster Antiquarian Journal 11 (1968) 13–21, and M. NíMhaonaigh, ‘Cogad Gáedel re Gallaib and the annals: a comparison’, Ériu 47 (1996) 101–26.

Northern Scotland 26 (2006) ISSN 0306-5278

75

these records towards the midlands of Ireland and the Shannon-basin. In addition,Etchingham has pointed out that the volume of records concerning vikingsfluctuates in the course of the ninth century. He has suggested that thesefluctuations reflect changes in chronicling rather than changes in viking-activityand that therefore the chronicles may be misleading. These are valid questions,and more work is necessary on the chronicles of Viking-Age Ireland to unravelthe story of their origins.4

PATTERNS OF VIKING-ACTIVITY

It is worth noting that Etchingham has admitted that changing levels ofviking-activity suggested in the chronicles of the ninth century are ‘likely to reflectreality and not merely change in the nature of the record’.5 Other types ofannal-entries, such as those about ecclesiastical events, ebb and flow in a fashionsimilar to the pattern of recorded viking-raids. Nevertheless, records of vikingsvary more dramatically in number and geographical distribution over time. Thissuggests a genuine change in the pattern of viking-attacks, rather than in ‘thenature of the record’.6 Ailbhe Mac Shamhráin pointed out that the better-recordedareas of Ireland (the eastern and central regions and river-valleys) were thewealthiest areas, where viking-activity was also focused.7 This is clearlydemonstrated when one compares chronicle-records with the distribution ofviking-artefacts and silver-hoards.8 The overall impression given by the chroniclesis probably correct. Nevertheless, it is inappropriate to quantify fluctuations inthe records of viking-activity in statistical terms without comparison with thewider range of chronicle-evidence.

Foreign sources lend support to the theory that Irish chronicles provide aninsight into patterns of viking-activity. From the 790s until the 840s, the volumeof recorded viking-attacks had been increasing. The pattern of coastal attacks,

4Pioneering work has been done by Nicholas John Evans, ‘The Textual Development of thePrincipal Irish Chronicles in the Tenth and Eleventh Centuries’ (unpublished Ph.D. dissertation,University of Glasgow 2003). See further Nicholas Evans, The Present and the Past in Medieval IrishChronicles (Woodbridge 2010).

5Etchingham, Viking Raids, p. 52.6Mary A. Valante, ‘Urbanization and Economy in Viking-Age Ireland’ (unpublished Ph.D.

dissertation, Pennsylvania State University 1998), and The Vikings in Ireland. Settlement, Trade andUrbanisation (Dublin 2008).

7Ailbhe Mac Shamhráin, The Vikings. An Illustrated History (Dublin 2002), p. 43.8C. Etchingham, ‘Les Vikings dans les sources documentaires irlandaises: le cas des annales’, in

L’Héritage maritime des vikings en Europe de l’Ouest, ed. Élisabeth Ridel (Caen 2002), pp. 35–56, at48–9; J. Sheehan, ‘Early Viking Age silver hoards from Ireland and their Scandinavian elements’, inIreland and Scandinavia in the Early Viking Age, edd. Howard B. Clarke et al. (Dublin 1998),pp. 166–202, at p. 174.


followed by bolder incursions into Irish territory, mirrors the pattern of activityrecorded in other western European countries.9 The first significant change in thenotices of vikings, observed by Etchingham, is the decline in records ofviking-attacks on churches after 850. Donnchadh Ó Corráin has attributed thischange to the establishment of Ívarr (founder of the dynasty called Uí Ímair) andhis associate Óláfr as kings of the vikings in Ireland.10 These rulers established anear-monopoly on viking-warfare in Ireland. It is from this time that we seealliances developing more frequently between vikings and Irish kings. The declinealso followed a series of major Irish victories against vikings in the 840s.11 From850 we also begin to hear more of the activities of vikings from Ireland in Britain,and a number of vikings perhaps left Ireland at that date.12

The volume of viking-attacks in Ireland remains low in the chronicle-recordof the 860s and 870s. A sharp exception to this trend is found in the years 866and 867 when a large number of Irish victories over vikings is reported. Thiscorresponds with the years when vikings from Ireland were campaigning inPictland and Northumbria.13 It may be that Irish rulers were taking advantage ofthe reduced manpower of vikings in Ireland at this point. In consequence offurther campaigns in Britain, Ireland was apparently seen as a safe haven fromvikings in the 870s. In 877, Rhodri son of Merfyn, king of Gwynedd, came toIreland, ‘in flight from the Dubgaill ’.14 In the following year the shrine ofColumba was also brought to Ireland from North Britain, in flight from vikings.15

A further change, noted by Etchingham,16 occurred from the end of the 870s:there was a sudden and notable increase in the number of recorded viking-attacks

9Clare Elizabeth Downham, ‘Britain and Scandinavian Ireland: the Dynasty of Ívarr andPan-Insular Politics to 1014’ (unpublished Ph.D. dissertation, University of Cambridge 2003), p. 55;cf. Clare Downham, Viking Kings of Britain and Ireland. The Dynasty of Ívarr to A.D. 1014(Edinburgh 2007; 2nd edn, 2008), pp. 11–12.

10D. Ó Corráin, ‘The vikings in Scotland and Ireland in the ninth century’, Peritia 12 (1998)296–339, at pp. 328–9; Downham, Viking Kings, pp. 1–23, 28–9 (cf. 63–71, 137–48, 201–4, 231–5).

11The Annals of Ulster (to A.D. 1131), I, edd. & transl. Seán Mac Airt & G. Mac Niocaill (Dublin1983), pp. 306/7 (848.4, 848.5, 848.6, 848.7).

12Downham, ‘Britain and Scandinavian Ireland’, pp. 50, 54; cf. Downham, Viking Kings,pp. 202–3.

13Downham, ‘Britain and Scandinavian Ireland’, pp. 137–8; Downham, Viking Kings, pp. 63–5, 138–41.

14The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.332/3 (877.3); ChronicumScotorum. A Chronicle of Irish Affairs, from the Earliest Times to A.D. 1135; with a Supplement,containing the Events from 1141 to 1150, ed. & transl. William M. Hennessy (London 1866),pp. 166/7 ([877].2).

15The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.334/5 (878.9); ChronicumScotorum, ed. & transl. Hennessy, pp. 166/7 ([878].4).

16Etchingham, Viking Raids, pp. 15–16.

RIVALRY BETWEEN VIKINGS, A.D. 795–1014 77

on churches. Their incidence remained high until the 890s. This increasecorresponds with the reversal of viking-fortunes in Britain at the battle of Edingtonin 878.17 It is a plausible deduction that some vikings may have decided to trytheir luck across the Irish Sea following this major defeat at the hands of KingAlfred. Nevertheless, the overall number of viking-events reported at the end ofthe century was still less than in the second third of the ninth century. This isdespite the proportional increase in recorded ecclesiastical events (excluding obits)noted by Etchingham for these years.18 Peter Sawyer has suggested that this declinein viking-activity in Ireland from the 870s was a consequence of the death of Ívarrin 873.19 It may be possible to demonstrate that changes in the number of recordedviking-campaigns can provide us with an insight into events of the ninth century.20

The same may be said of tenth-century records.21 In 902 the viking-dynastyof Ívarr was expelled from Dublin by the overkings of Leinster and Brega. By thisdate, Dublin had become established as the principal viking-settlement in Ireland,and its loss was a major blow to the family of Ívarr. The sharp decline of referencesto viking-warfare in Ireland from 902 to 914 coïncides with the exile of thisdynasty. Archaeological evidence suggests that Dublin remained settled in theseyears.22 Nevertheless, it seems that many vikings left Ireland to seek their fortuneselsewhere. The dynasty of Ívarr did not return to Ireland until 914. These yearsmark a low point in viking-activity in Ireland, with few recorded campaigns.23

From 914, there is a significant increase in the level of viking-warfare reportedin Ireland. This coïncides with the return of the dynasty of Ívarr. The number ofrecorded viking-campaigns from 917 to 943 rivals any other period of Irish history.

17Dorothy Whitelock, ‘The importance of the battle of Edington’, Report of The Society of Friendsof the Priory Church of Edington, Wiltshire (1975–7) 6–15, reprinted in her book, From Bede to Alfred.Studies in Early Anglo-Saxon Literature and History (London 1980), essay XIII, with the samepagination.

18Etchingham, Viking Raids, p. 16.19P.H. Sawyer, The Age of the Vikings (2nd edn, London 1971), pp. 206–7.20Ó Corráin, ‘The vikings in Scotland and Ireland’, p. 328.21Etchingham, ‘Les vikings’, pp. 51–3.22L. Simpson, ‘Forty years a-digging: a preliminary synthesis of archaeological investigations in

medieval Dublin’, Medieval Dublin 1 (1999) 11–68, at p. 25.23The Annals of Clonmacnoise, being Annals of Ireland from the Earliest Period to A.D. 1408,

translated into English A.D. 1627 by Conell Mageoghagan, ed. Denis Meehan (Dublin 1896), p. 144(901[=909].1); Annala Rioghachta Eireann, Annals of the Kingdom of Ireland, by The Four Masters,from the Earliest Times to the Year 1616, ed. & transl. John O’Donovan (2nd edn, 7 vols, Dublin1856), I.560/1 (900[=904].3), and II.572–5 (904[=909].7); The Annals of Ulster, edd. & transl.Mac Airt & Mac Niocaill, I.354/5 (904.4); Chronicum Scotorum, ed. & transl. Hennessy, pp. 178–81 ([904].6–7); Fragmentary Annals of Ireland, ed. & transl. Joan Newlon Radner (Dublin 1978),pp. 166–73 (§429).


This peak in viking-warfare has been noted by various commentators, and somehave regarded it as the zenith of viking-power in Ireland.24 Many of the campaignsresulted from rivalry between the vikings of Dublin and Limerick. Dublin emergedvictorious from this dispute in 937.25 The vikings of Dublin also led frequentcampaigns against the neighbouring population-group of Uí Dúnchada inNorthern Leinster. However, their most bitter enemy was the overking of CenélnEogain, Muirchertach mac Néill, whom they fought relentlessly from the 920sto the 940s.26 The extent of the Dubliners’ military engagements in these years,and their ability to rival Muirchertach, the most powerful king in Ireland, indicatesthe strength of these viking-rulers.

The number of recorded viking-campaigns declined significantly after 944.Dublin was sacked in this year by the overkings of Leinster and Brega, and a feudwas developing within the dynasty of Ívarr.27 The power of Dublin was furthercompromised by the reversal of their fortunes abroad, as the dynasty of Ívarrstruggled, unsuccessfully, to maintain control of Northumbria.28 The record ofviking-events continues to remain slender during the 950s and 960s. In the latetenth century, vikings campaigned more frequently in alliance with Irish rulersand were drawn more fully into the web of Irish politics. Their independencebecame increasingly compromised as Irish overkings competed to control thetowns.29

The vikings of Dublin fought several battles in an attempt to reverse their ailingpolitical fortunes. These included engagements at Tara in 980, Glenn Máma in999, and finally Clontarf in 1014.30 These major battles followed noticeable risesin the records of viking-activity in the 970s, late 990s, and in 1013. Each can beinterpreted as a build-up of military activity prior to these great events. On allthree occasions the Dublin-vikings and their Irish allies were defeated. Despitethe increasing subordination of viking-towns to Irish overkings, the towns persisted

24J. Ryan, ‘Pre-Norman Dublin’, Journal of The Royal Society of Antiquaries of Ireland 79 (1949)64–83, at p. 72; Alfred P. Smyth, Scandinavian York and Dublin. The History and Archaeology of Tworelated Viking Kingdoms (2 vols, Dublin 1975/9), II.8; T.M. Charles-Edwards, ‘Irish warfare before1100’, in A Military History of Ireland, edd. Thomas Bartlett & K. Jeffery (Cambridge 1996), pp. 26–51, 463–5, at pp. 44–5.

25Annala, ed. & transl. O’Donovan, II.632/3 (935[=937].11).26B. Jaski, ‘The vikings and the kingship of Tara’, Peritia 9 (1995) 310–53, at pp. 329–33.27Downham, ‘Britain and Scandinavian Ireland’, p. 41; Downham, Viking Kings, pp. 43–8.28P. Holm, ‘The slave trade of Dublin, ninth to twelfth centuries’, Peritia 5 (1986) 317–45, at

p. 329; cf. chapter 9, below.29D. Ó Corráin, ‘The vikings in Ireland’, in The Vikings in Ireland, ed. Anne-Christine Larsen

(Roskilde 2001), pp. 17–27, at 24–5.30A. Mac Shamhráin, ‘The battle of Glenn Máma, Dublin and the high-kingship of Ireland: a

millennial commemoration’, Medieval Dublin 2 (2000) 53–64.


as self-governing communities during the eleventh and twelfth centuries andplayed an important role on the Irish political scene.31 The volume of recordedevents can be seen to reflect the fortunes of different viking-groups in Irelandduring the ninth and tenth centuries, despite some of the shortfalls andinadequacies in chronicles as a historical source.

FINNGAILL, DUBGAILL, AND GALL-GOÍDIL

It is not until the late 830s that Irish chronicles begin to give an insight into theidentity of different viking-leaders in Ireland. In 837 two fleets of sixty shipsmounted a great invasion into the heartlands of Ireland. One fleet travelled alongthe River Liffey, and the other entered the River Boyne.32 These were probably anaggregation of forces with different leaders.33 The attack may have beendeliberately planned as a pincer-movement against the heartlands of the powerfulruling dynasty of the Southern Uí Néill, which lay between the two rivers.34 Inthe same year Søxólfr was slain: he is the first viking-leader in Ireland to be namedin our sources.35 In the course of the next two decades’ records, we see differentviking-leaders vying for control over all the vikings settled in Ireland.

In 845 Þórgils seems to have been the most prominent viking in Ireland. Heestablished a base on Lough Ree on the River Shannon and plundered many ofthe surrounding churches and territories.36 He later became a figure of legend.37

In the 1180s Gerald of Wales reported his downfall at the hands of fifteen Irishmendressed as ladies who deceived the men of Þórgils and killed them ‘in the midst of

31Ó Corráin, ‘The vikings in Ireland’, pp. 26–7. See further chapter 8, below.32The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.294/5 (837.3); Chronicum

Scotorum, ed. & transl. Hennessy, pp. 140/1 ([837].2).33One may compare the English examples of this phenomenon discussed in chapter 2, above

(pp. 17,18, 22~3, 23~4).34A.P. Smyth, ‘The effect of Scandinavian raiders on the English and Irish Churches: a

preliminary assessment’, in Britain and Ireland, 900–1300. Insular Responses to Medieval EuropeanChange, ed. Brendan Smith (Cambridge 1999), pp. 1–38, at p. 8.

35The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.294/5 (837.9), where he iscalled toís<ech> na nGall, ‘the leader of the Foreigners’; Chronicum Scotorum, ed. & transl. Hennessy,pp. 142/3 ([837].6).


37Cogadh Gaedhel re Gallaibh, ed. & transl. Todd, pp. 8–15 (§§9–14); J. Stewart, ‘The death ofTurgesius’, Saga-book 18 (1970–3) 47–58.

38J.J O’Meara (ed.), ‘Giraldus Cambrensis in Topographia Hibernie. Text of the First Recension’,Proceedings of The Royal Irish Academy 52 C (1948–50) 113–78, at p. 174; Gerald of Wales, TheHistory and Topography of Ireland, transl. John J. O’Meara (2nd edn, Portlaoise/Harmondsworth1982), pp. 120–1 (§114).


their embraces’.38 However, ‘The Annals of Ulster’ more soberly report his captureand execution by the overking of the Southern Uí Néill, Mael Sechnaill mac MaeleRuanaid.39 Vikings continued to campaign successfully in Ireland for the next twoyears. According to ‘The Annals of Saint-Bertin’ many of the Scotti (Gaels) wereforced to pay tribute to them in 847.40 However, their success during the 840swas short-lived. A large army of vikings fell alongside their leader Hákonn in battleagainst Cerball, overking of the Osraige, in 847.41 Vikings received further defeatsin 848 in four major battles, and Þórir, a deputy of the king of Laithlind (tániserígh Laithlinne) was killed.42 Then, in 849, Dublin was plundered byMaelsechlainn, overking of the Southern Uí Néill.43 David Dumville has suggestedthat the reversal of viking-fortunes may have encouraged another viking-group tointervene and seize power.44 These are identified as the Dubgaill or ‘Dark/NewForeigners’.45

The Dubgaill are first described in ‘The Annals of Ulster’ and ChronicumScotorum for the year 849.46 This new group du thiachtain du tabairt greammaforsna Gaillu ro-bádar ara ciunn, ‘came to exact obedience from the Foreignerswho were [in Ireland] before them’. Those already settled were named Finngaill,‘Fair/Old Foreigners’, to distinguish them from the new arrivals. In 851, Dubgaill– named thus – won battles against Finngaill at Dublin and Linns.47 The Finngaillretaliated by attacking their rivals at Carlingford Lough in the following year, butthey were heavily defeated. During the 850s the Dubgaill rose to prominenceunder the leadership of three kinsmen, Óláfr, Ívarr, and Ásl. It was their successors(the dynasty of Ívarr) who held sway over the main viking-settlements in Ireland,

39The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.302/3 (845.8); cf. ChronicumScotorum, ed. & transl. Hennessy, pp. 146/7 ([845].6).

40Annales de Saint-Bertin, edd. Félix Grat et al. (Paris 1964), p. 54 (847.6); The Annals ofSt-Bertin, transl. Janet L. Nelson (Manchester 1991), p. 65.

41The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.306/7 (847.4); ChronicumScotorum, ed. & transl. Hennessy, pp. 146–9 ([847].3). See further chapter 5, below.


43Ibid., pp. 148/9 ([849].4).44David N. Dumville, ‘Old Dubliners and New Dubliners in Ireland and Britain: a Viking-Age

story’, Medieval Dublin 6 (2004) 78–93, at p. 83, reprinted in his book Celtic Essays, 2001–2007(2 vols, Aberdeen 2007), I.103–22, at pp. 108–10.

45Ibid.; A.P. Smyth, ‘The Black Foreigners of York and the White Foreigners of Dublin’,Saga-book 19 (1974–7) 101–17.

46The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.308/9 (849.6); cf. ChronicumScotorum, ed. & transl. Hennessy, pp. 148–51 ([849].6).

47The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.310/11 (851.3 and 852.3);Chronicum Scotorum, ed. & transl. Hennessy, pp. 150–3 ([851].3–4 and [852].3).

48See, for example, The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.378/9(927.2); Chronicum Scotorum, ed. & transl. Hennessy, pp. 196/7 ([926]=[927].2).


and those leaders are sometimes called ‘kings of the Fair/Old and Dark/NewForeigners’.48 The distinction between the two groups continued to be made inIrish sources until 941 (the terms reäppear later in different contexts).49

In the late 850s Óláfr, Ívarr, and Ásl had to deal with another viking-group.These were called Gallgoídil (literally ‘Foreigner-Gaels’). They have been identifiedas vikings of mixed Gaelic and Scandinavian ethnicity.50 Their origins have beendebated. The argument that Gallgoídil came from Scotland was first advanced inthe nineteenth century.51 This is because modern Galloway in Scotland appearsto have taken its name from these people.52 Nevertheless, Galloway does not seemto have been settled by vikings before about 900; so a ninth-century link seemsimprobable.53 David Dumville has recently argued that Gallgoídil could haveoriginated anywhere there was sustained contact between Gaels and vikings fromthe 830s and 840s, either in the Scottish islands or in Ireland.54

In favour of the theory of polygenesis of Gallgoídil, Irish chronicles indicatethat some Gallgoídil were active in Ireland. In 856 Gallgoídil campaigned in theterritory of the Southern Uí Néill and in the north of Ireland at Glenn Foichle(Co. Tyrone).55 In 857 a group of Gallgoídil led by Caittil find was defeated inMunster by Óláfr and Ívarr.56 Then in 858 a large force of the Gallgoídil fromLeth Cuinn (the northern half of Ireland) was defeated at Ara Tíre (Co. Tipperary)by Ívarr and an Irish regional over-king, Cerball of the Osraige.57 Alfred Smyth

49Smyth, ‘The Black Foreigners’, pp. 116 and 109–10; W. McLeod, ‘Rí Innsi Gall, rí Fionnghall,Ceannas nan Gàidheal: sovereignty and rhetoric in the late medieval Hebrides’, Cambrian MedievalCeltic Studies 43 (2002) 25–48, at pp. 36–8.

50Barbara E. Crawford, Scandinavian Scotland (Leicester 1987), p. 47; Ó Corráin, ‘The vikingsin Scotland’, p. 326; M. Ní Mhaonaigh, ‘The outward look. Britain and beyond in medieval Irishliterature’, in The Medieval World, edd. Peter Linehan & J.L. Nelson (London 2001), pp. 381–97,at 391–2.

51Willliam F. Skene, Celtic Scotland: a History of Ancient Alban (3 vols, Edinburgh 1876–80),I.311–13.

52Alfred P. Smyth, Scandinavian Kings in the British Isles, 850–880 (Oxford 1977), p. 115.53David N. Dumville, The Churches of North Britain in the First Viking-Age (Whithorn 1997),

p. 29; James Graham-Campbell & C.E. Batey, Vikings in Scotland – an Archaeological Survey(Edinburgh 1998), p. 107; Downham, ‘Britain and Scandinavian Ireland’, pp. 193–4; Downham,Viking Kings, pp. 170–5.

54Dumville, The Churches of North Britain, pp. 27–8; J.I. Young, ‘A note on the Norseoccupation of Ireland’, History, new series, 35 (1950) 11–33, at p. 24.



57Annala, ed. & transl. O’Donovan, I.490/1 (856[=858].5); Chronicum Scotorum, ed. & transl.Hennessy, pp. 156/7 ([858].3); Fragmentary Annals, ed. & transl. Radner, pp. 104/5 (§263). Cf.chapter 5, below.


proposed that Caitill find was the same as a Hebridean chieftain, Ketill flatnefr,named in Icelandic sagas.58 This view has been criticised by Donnchadh ÓCorráin.59 There is no reason to suppose that a Ketill named in thirteenth-centurysagas was the same as a man of a similar name active in southern Ireland in 857.Furthermore, the chronology of the two individuals does not match.60 The nameGallgoídil disappears from Irish chronicles in the 860s; it resurfaces in the eleventhcentury, representing the territory of Galloway.61 Perhaps, after the 850s, vikingsof mixed ethnicity were not sufficiently novel or distinct on the Irish politicalscene to warrant special appellation.

The next disputes recorded between vikings in Ireland were among theDubgaill (‘Dark/New Foreigners’). In 867, Ásl was killed by his kinsmen.62 Thenin 875, Hálfdan, brother of Ívarr, killed Eysteinn Óláfsson, who may have beenhis nephew.63 These killings may have been the consequence of political rivalry.Further dissensions are reported in 881, 893, and 896 with the deaths of two sonsof Ívarr.64 The troubles seem to have been focused on possession of Dublin.

Following the deaths of Óláfr and Ívarr, weaknesses within the family of Ívarrled to a renewal of the struggle between Finngaill and Dubgaill. In 877 Hálfdanwas slain in a battle against the Finngaill of Strangford Lough.65 Further troublearose in 883 when a son of Ásl (a Dubgall) was slain by Óttar son of Járnkné inalliance with an Irish princess.66 Óttar’s father, Járnkné, was probably the leaderof the Finngaill who died at the hands of Dubgaill at the battle of CarlingfordLough in 852.67 The killing of Ásl’s son may have been an act of revenge.

The difficulties encountered by the developing dynasty of Ívarr helped to spur

58Laxdæla saga, Halldórs �þættir Snorrasonar, Stúfs �þáttr, ed. Einar Ól. Sveinsson (Reykjavík 1934),pp. 3–8 (§§1–4); The Sagas of Icelanders. A Selection, transl. Katrina C. Attwood et al. (London1997), pp. 270–421, at 276–9. Cf. Smyth, Scandinavian Kings, pp. 118–19.

59D. Ó Corráin, ‘High-kings, vikings and other kings’, Irish Historical Studies 21 (1978/9) 283–323, at p. 301.

60Smyth, Scandinavian Kings, p. 117.61The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.472/3 (1034.10).62The Annals of Clonmacnoise, ed. Murphy, p. 142 (865[=867].4); The Annals of Ulster, edd.

Mac Airt & Mac Niocaill, I.322/3 (867.6); Fragmentary Annals, ed. & transl. Radner, pp. 126/7(§347).

63The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.330/1 (875.4).64Ibid., I.336/7 (881.3), 346/7 (893.4), 348/9 (896.3).65Ibid., I.332/3 (877.5); Chronicum Scotorum, ed. & transl. Hennessy, pp. 166/7 ([877].3).66The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.338/9 (883.4); Chronicum

Scotorum, ed. & transl. Hennessy, pp. 168/9 ([883].3).67The Annals of Ulster, ed. & transl. Mac Airt & Mac Niocaill, I.310/11 (852.3); Chronicum

Scotorum, ed. & transl. Hennessy, pp. 152/3 ([852].3); Fragmentary Annals, ed. & transl. Radner,pp. 90–5 (§235).


their Irish enemies into action. In 902 viking-leaders were chased out of Dublinby the overkings of Leinster and Brega.68 It is not clear what the fate was of othervikings in Ireland. Limerick at this date may still have been in the possession ofFinngaill or another group of rivals to the dynasty of Ívarr. The first leader ofLimerick to be identified in the chronicles is Þórir Helgason. He campaigned from922 along the River Shannon and in Meath, and he died in 927 or 928.69 Hisname is unlike those sported by the dynasty of Ívarr, and it seems correspondinglyunlikely that he was related to them. In 924, war broke out between Þórir andGuðrøðr, who was a grandson of Ívarr and the king of Dublin.70 This disputebetween Dublin and Limerick lasted to 937.

The divisions between the rival viking-groups were not always clear-cut. From924 Kolli Barðarson is named as king of Limerick, and he died in 932.71 Hisparentage is uncertain. Kolli may have been a son of that Barðr Ívarsson who waskilled and burnt at Dublin in 881.72 In 927, there is clearer evidence of dissensionwithin the dynasty of Ívarr. In that year Þórir of Limerick supported the sons ofSigtryggr grandson of Ívarr against their kinsman Guðrøðr at Dublin.73 Inaddition, in 931 Guðrøðr of Dublin expelled from Ossory a ‘grandson of Ívarr’from Limerick.74

In the 920s and 930s Limerick seems to have been a source of support forthose who opposed the rule of Guðrøðr, grandson of Ívarr, at Dublin. After thedeath of Guðrøðr in 934, his son, Óláfr Guðrøðsson, continued the war againstLimerick. The struggle ended when Óláfr Guðrøðsson imprisoned Óláfrcenncairech (‘Scabby Head’), king of Limerick, in 937.75

THE DESCENDANTS OF GUÐRØÐR AND SIGTRYGGR, GRANDSONS OF ÍVARR

From the 940s the term Finngaill ceases to be mentioned in Irish chronicles, andthe dynasty of Ívarr ruled all the major viking-settlements. From 927 until the960s the descendants of two grandsons of Ívarr, Sigtryggr and Guðrøðr, competed


69The Annals of Clonmacnoise, ed. Murphy, p. 148 (922[=927].6 and 8); The Annals of Inisfallen(MS. Rawlinson B.503), ed. & transl. Seán Mac Airt (Dublin 1951), pp. 146/7 ([922].2); The Annalsof Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.374/5 (922.3) and I.378/9 (928.5).

70Ibid., I.376/7 (924.3).71Chronicum Scotorum, ed. & transl. Hennessy, pp. 194/5 ([923]=[924].5) and 198/9

([931]=[932].3).72Annala, ed. & transl. O’Donovan, I.526/7 (878[=881].3).73The Annals of Clonmacnoise, ed. Murphy, p. 148 (922[=927].6).74Ibid., p. 149 (926[=931].3); Annala, ed. & transl. O’Donovan, II.624/5 (928[=930].5 and

929[=931].8).75Ibid., II.632/3 (935[=937].11).


to rule the most prestigious settlements under the family’s control.In 927, the sons of Sigtryggr had attempted to displace their cousins, the sons

of Guðrøðr, from Dublin.76 Nevertheless, Óláfr Guðrøðsson was able to holdpower there until 939. He then left Dublin in order to take control of theScandinavian kingdom of York, and Óláfr Sigtryggsson took his place in Dublin.When Óláfr Guðrøðsson died, he was succeeded at York by Óláfr Sigtryggsson.Blákári Guðrøðsson then became king of Dublin.77 It seems that rival dynasticsegments alternated in control of Dublin and York for a short period in the 930sand 940s. York may have been regarded as the greater prize, as kings of Dublinwould migrate thither when they had opportunity.78

This arrangement soon broke down. ‘The Anglo-Saxon Chronicle’ reports thatBlákári’s brother Røgnvaldr Guðrøðsson went to England in 943.79 Soonafterwards, Røgnvaldr Guðrøðsson and Óláfr Sigtryggsson were both driven outof York by the English king, Edmund I.80 Alex Woolf has suggested that Edmundwas helped by rivalry between the two viking-leaders, which weakened their ruleat York.81 After the expulsion, Óláfr Sigtryggsson returned to Ireland and forcedBlákári Guðrøðsson out of Dublin.82 When Óláfr Sigtryggsson returned to takepower in York for the second time, his brother, Guðrøðr Sigtryggsson, took chargeat Dublin.83 Thus, with Blákári’s death, the descendants of Guðrøðr grandson ofÍvarr were deprived of a position of power in Ireland. They may have sought theirfortunes elsewhere in the Irish Sea.84 In 960 and 962 the sons of Óláfr Guðrøðssonwaged minor campaigns in Ireland.85

76See above, n. 73.77Annala, ed. & transl. O’Donovan, II.638/9 (937[=939].13) and 640/1 (938[=940].12).78Cf. Smyth, Scandinavian York, I.71 and II.1, 11, 90–1. Cf. chapter 9, below.79The Anglo-Saxon Chronicle. A Collaborative Edition, gen. edd. David Dumville & S. Keynes, III,

MS A, ed. Janet M. Bately (Cambridge 1986), p. 73, s.a. 942; ibid., IV, MS B, ed. Simon Taylor(Cambridge 1983), p. 53, s.a. <942>; ibid., V, MS C, ed. Katherine O’Brien O’Keeffe (Cambridge 2001),pp. 79–80, s.a. 942; ibid., VI, MS D, ed. G.P. Cubbin (Cambridge 1996), pp. 43–4, s.a. 943; ibid., VII,MS. E, ed. Susan Irvine (Cambridge 2004), p. 55, s.a. 942; ibid., VIII, MS F, ed. Peter S. Baker(Cambridge 2000), p. 80, s.a. 942.

80Ibid., III, MS A, ed. Bately, p. 74, s.a. 944; ibid., IV, MS B, ed. Taylor, p. 53, s.a.<944>; ibid., V,MS. C, ed. O’Brien O’Keeffe, p. 80, s.a. 944; ibid., VI, MS D, ed. Cubbin, p. 44, s.a. 944; ibid., VII,MS. E, ed. Irvine, p. 55, s.a. 944; ibid., VIII, MS F, ed. Baker, p. 80, s.a. 944.

81A. Woolf, ‘Amlaíb Cuarán and the Gael, 941–81’, Medieval Dublin 3 (2001) 34–43.82The Annals of Clonmacnoise, ed. Murphy, p. 154 (937[=945].9); Annala, ed. & transl. O’Donovan,

II.654/5 (943[=945].6); The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.392/3 (945.6).83The Annals of Clonmacnoise, ed. Murphy, p. 156 (946[=951].3); The Annals of Ulster, edd. & transl.

Mac Airt & Mac Niocaill, I.396/7 (951.3); Chronicum Scotorum, ed. & transl. Hennessy, pp. 210/11([950]=[951].4).

84C. Etchingham, ‘North Wales, Ireland and the Isles: the Insular viking zone’, Peritia 15 (2001)145–87, at p. 170.

85The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.402/3 (960.3); Annala, ed. & transl.O’Donovan, II.680–3 (960[=962].13–15).


VIKING-RIVALRIES BEFORE THE BATTLE OF CLONTARF

After the 960s inter-viking wars focused on members of the dynasty of Ívarrlocated in different settlements. Some of their rivalries were represented at thefamous battle of Clontarf which was fought on Good Friday in 1014. Essentially,this battle was triggered by the efforts of Maelmórda, overking of Leinster, andSigtryggr silkiskeggi, king of Dublin, to stem the growing power of the overkingof Munster, Brian Bóroma, in Ireland.86 Nevertheless, rivalries between differentviking-groups helped to shape the events which led up to this famous conflict.

In the 970s ‘the sons of Haraldr’ appear in the chronicles as powerful leadersof the kingdom of Man and the Isles. They may have been sons of Haraldr, kingof Limerick, who died in 940.87 In 974 Maccus son of Haraldr waged war againstLimerick, imprisoning its king, Ívarr, who later escaped. The overking of Munster,Brian Bóroma, killed Ívarr of Limerick and his sons in 977.88 Brian seems to havebrought the port directly under his authority. Its resources, in wealth and militarypower, no doubt helped in forwarding his ambition to dominate Ireland. Brian isfound in alliance with the Hebridean ‘sons of Haraldr’ in 984, when they made apact against the rulers of Dublin.89 Thus Brian may have exploited the rivalrieswhich developed between different viking-groups’ power-bases.

In 986 another group of vikings, labelled Danair, appeared in the Irish Seaand became involved in Irish politics. These can be identified with Scandinavianfleets which also attacked England during the 980s.90 In 986, a band of Danairraided Iona. There they killed fifteen elders of the monastery and took Abbot MaelCiaráin to Dublin where he was martyred.91 In 987 some Danair were allied withthe men of the Isles and won a victory on Man with the son of Haraldr, althoughtheir enemy is not named.92 In 990 Danair attacked Derry and allied with thearmies of Dublin and Leinster to wage war against Meath.93 Thereafter, less is

86J. Ryan, ‘The battle of Clontarf ’, Journal of The Royal Society of Antiquaries of Ireland 68 (1938)1–50.

87The Annals of Inisfallen, ed. & transl. Mac Airt, pp. 152/3 ([940].1).88Annala, ed. & transl. O’Donovan, II.698/9 (972[=974].9); The Annals of Inisfallen, ed. &

transl. Mac Airt, pp. 160/1 (974.2), 162/3 (977.2); Chronicum Scotorum, ed. & transl. Hennessy,pp. 224/5 ([975]=[977].1).

89The Annals of Inisfallen, ed. & transl. Mac Airt, pp. 164/5 (984.2).90Etchingham, ‘North Wales, Ireland’, p. 177.91The Annals of Clonmacnoise, ed. Murphy, p. 160 (980[=986].2); Annala, ed. & transl.

O’Donovan, II.718/19 (985[=986].7–8); The Annals of Inisfallen, ed. & transl. Mac Airt, pp. 166/7(986.4); The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.420/1 (986.3); ChronicumScotorum, ed. & transl. Hennessy, pp. 230/1 ([984]=[986].2).

92The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.420/1 (987.1).93Ibid., I.422/3 (990.1); Annala, ed. & transl. O’Donovan, II.726/7 (989[=990].5). On the

interpretation of the latter item, cf. chapter 6, n. 46, below.


heard of the Scandinavian fleets, although a contingent may have fought onDublin’s side at the battle of Clontarf in 1014.

Waterford and Dublin were locked in conflict for a number of years before thebattle of Clontarf. In 983 a war broke out between the two ports. One of the sonsof Ívarr, king of Waterford, was killed in this conflict.94 In 984, Ívarr and BrianBóroma joined forces to plan an attack on Dublin. Although their forces did notreach the town, they caused extensive damage to churches and fortifications inLeinster.95 The rivalry between Waterford and Dublin intensified further after989, when the death of Glúniairn (Járnkné), king of Dublin, triggered a disputeover royal succession. The kingship of Dublin was seized by Ívarr, king ofWaterford, on two occasions, in 990 and 995. He was expelled by Glúniairn’sbrother Sigtryggr.96 In the run-up to the battle of Clontarf, vikings of Waterfordsupported Brian Bóroma against Dublin and Leinster and may have participatedat Clontarf itself.97

Sigtryggr’s supporters at the battle of Clontarf included viking-contingentsfrom the Hebrides and the Northern Isles. Óláfr son of Lagmann may have led aHebridean contingent. The Welsh vernacular chronicle Brenhinoedd y Saeson statesthat the king of Dublin, Sigtryggr silkiskeggi, paid vikings led by Brodr toparticipate in the battle.98 The earl of Orkney, Sigurðr Hløðvisson, also foughtand died there.99 (Sigurðr is commemorated in the early thirteenth-century Njálssaga as a powerful ruler who extended his sway over parts of the Hebrides.)100

His presence in the Irish Sea in 1014 may have been due to an extension of histerritorial power. Alternatively he may have played a role in events which led tothe conquest of England by Sveinn, king of Denmark, in 1013. The renewal ofviking-activity in England at the close of the tenth century and in the earlyeleventh century attracted mercenaries from across the viking-world to British

94The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.418/19 (983.2); ChronicumScotorum, ed. & transl. Hennessy, pp. 228/9 ([981]=[983].1).

95The Annals of Inisfallen, ed. & transl. Mac Airt, pp. 164/5 (984.2).96Ibid., pp. 168/9 (990.2) and 170/1 (993.6); Annala, ed. & transl. O’Donovan, II.730/1

(992[=993].17), 732/3 (993[=994].5 and 994[=995].2), 734/5 (994[=995].6).97Cogadh Gaedhel re Gallaibh, ed. & transl. Todd, pp. 146–9 (§84), 168/9 (§96). Cf. Young, ‘A

note’, p. 26.98This statement is found in all three versions (P, R, S) of that chronicle (but not in any version

of Annales Cambriae, its Latin relative): see Brenhinoedd y Saeson, ‘The Kings of the English’, A.D.682–1097: Texts P, R, S in Parallel, ed. & transl. David N. Dumville (Aberdeen 2012), pp. 62/3.

99The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.446–9 (1014.2); ChronicumScotorum, ed. & transl. Hennessy, pp. 252/3 ([1012]=[1014].2).

100Brennu-Njáls saga, ed. Einar Ól. Sveinsson (Reykjavík 1954), pp. 205–6, 220–4, 439–42(§§85, 89, 154); Njal’s Saga, transl. Magnus Magnusson & Hermann Pálsson (Harmondsworth1960), pp. 182–3, 193–6, 341–2.


shores.101 Sigtryggr may have taken the opportunity of hiring some of theseScandinavian fleets in his battle against Brian.102

The battle of Clontarf was fought a short distance north of Dublin. In theengagement, Brian was killed but his forces were victorious. The battle may nothave had a decisive impact on Irish politics.103 Sigtryggr remained in power atDublin. The descendants of Brian also remained a major force in Irish politics. Itcan be argued that the fame of the battle of Clontarf lay not in its immediateimpact but rather in the wide range of its participants and its scale. As aconsequence, the event was commemorated in Irish and Icelandic mediaevalliterature.104 Part of the battle’s fame also lies in the fact that it marks the death ofBrian who had perhaps done more than any king before him in winningsupremacy throughout Ireland.105

CONCLUSIONS

It is quite clear from a brief study of viking-rivalries in Ireland that vikings werefar from united in their political outlook. At least four groups who were active inIreland earned distinct appellations, namely Dubgaill, Finngaill, Gallgoídil, andDanair. Other viking-groups may have settled, whose names are not reported inthe sources. Viking-attacks also continued to be led against Ireland from theContinent, as demonstrated by the record in Annales Xantenses for 870: ‘and theheathens who were ravaging nearly all Ireland at that time returned with manyspoils, and through the waterways of Francia and Gaul they brought many miseriesto humankind’.106

As a result of the rivalries between viking-groups in Ireland, there is no clear,general dichotomy between Scandinavian and Gael in the Viking-Age wars(despite the claims of later propagandists). Vikings joined Irish leaders to fightagainst their Scandinavian or Irish rivals. In addition there was stiff competition

101Annala, ed. & transl. O’Donovan, II.726/7 (989[=990].5); cf. n. 93, above. The Annals ofUlster, edd. & transl. Mac Airt & Mac Niocaill, I.418/19 (986.2), 420/1 (987.1, 987.3), 422/3(990.1).

102C. Downham, ‘England and the Irish Sea zone in the eleventh century’, Anglo-Norman Studies26 (2003) 55–73, at p. 62.

103Ryan, ‘The battle of Clontarf’, pp. 47–9; Young, ‘A note’, pp. 20–1; Donncha Ó Corráin,Ireland before the Normans (Dublin 1972), p. 130.

104C. Downham, ‘The battle of Clontarf in Irish history and legend’, History Ireland 13 (2005),no. 5, pp. 19–23.

105Francis John Byrne, Irish Kings and High-kings (2nd edn, Dublin 2001), pp. 266–7. Cf. MáireNí Mhaonaigh, Brian Boru, Ireland’s Greatest King? (Stroud 2007).

106Annales Xantenses et Annales Vedastini, ed. B. von Simson (Hannover 1909), p. 30(s.a. 871[=870]).


for the leadership of viking-settlements. Settlements also fought to determine theshare of Irish territory with which they raided and traded.

Studies of viking-activity have naturally focused on relations betweenScandinavians and the native populations of the places where they settled. Theevidence from Ireland gives an insight into the rivalries and tensions betweenopposing viking-factions, which helped to shape political events. This evidencemay have a comparative value for other areas of viking-settlement where similarcompetition for resources and authority may have arisen, but where the writtensources are of a different character or not so abundant.


DURING the reign of Cerball mac Dúnlainge (842–88) the overkingdom of theOsraige rose from relative obscurity to become a major player in Irish politics.1

Cerball’s career is remarkable for its longevity and success. This is particularlyimpressive, as he lived through an intense era of viking-activity in Ireland. I havedivided this chapter into three sections, dealing with written sources, the politicalbackground to Cerball’s life, and his deeds. In so doing, I hope to highlight thecircumstances in which Cerball rose to power, and to illustrate how his fame wascultivated after his death.2

WRITTEN SOURCES

‘The Osraige Chronicle’The most lavish account of Cerball’s deeds has survived in an eleventh-centurysaga which has been entitled ‘The Osraige Chronicle’ by Joan Radner.3 Anincomplete and edited version of this saga has survived, embedded in ‘TheFragmentary Annals of Ireland’.4 In addition, some brief summaries drawn froma lost antecedent of ‘The Fragmentary Annals’ are preserved in ‘The Annals ofThe Four Masters’.5

Ossory, Laois and Leinster 1 (2004) ISSN 1649-4938

5

The career of Cerball mac Dúnlainge, overking of the Osraige

1J.N. Radner, ‘Writing history: early Irish historiography and the significance of form’, Celtica23 (1999) 312–25, at p. 322.

2I should like to thank Alex Woolf for reading and correcting a draft of this chapter.3Fragmentary Annals of Ireland, ed. & transl. Joan Newlon Radner (Dublin 1978), pp. xxii–xxv.4Radner’s title for her book could apply to almost every extant Irish chronicle: I have therefore

made her title minimally specific and renamed the chronicle as ‘The Fragmentary Annals of Ireland’.5Fragmentary Annals, ed. & transl. Radner, pp. xvi–xvii, xxix–xxx; see further chapter 6, below.

The fourth fragment of ‘The Fragmentary Annals of Ireland’ shares twenty-nine records exclusivelywith ‘The Annals of The Four Masters’: for that text see Annala Rioghachta Eireann, Annals of theKingdom of Ireland, by The Four Masters, from the Earliest Times to the Year 1616, ed. & transl. JohnO’Donovan (2nd edn, 7 vols, Dublin 1856). These two groups of entries represent differentcomponents within ‘The Fragmentary Annals’. Thus it seems that the compilers of ‘The Annals ofThe Four Masters’, working in 1632–6, drew on a version of ‘The Fragmentary Annals’ which hasnow been lost. At times this lost version seems to have given a fuller record of events: this is suggestedby comparison of ‘The Fragmentary Annals’, §§293 and 318, with the equivalent records in ‘TheAnnals of The Four Masters’.

91

‘The Osraige Chronicle’6 may have been written under the sponsorship ofCerball’s great-great-grandson Donnchad mac Gilla Phátraic. He ruled theOsraige 1003–39 and briefly extended his sway to become provincial overking ofLeinster (1033–9).7 Donnchad seems to have promoted a cult of Cerball as anillustrious ancestor whose achievements served as a precursor to, and justificationof, his own ambitions.8 To evaluate the saga as a source for Cerball’s reign, it isnecessary to question how it has been transmitted and how its composition wasaffected by events in the eleventh century.

Only one transcript of ‘The Fragmentary Annals’ survives.9 ‘The FragmentaryAnnals’, as we now have that chronicle, was a late seventeenth-century copy of an‘old broken book’, the text of which survived in five parts. Cerball’s deeds aredescribed in the fourth fragment, which covers the years 849–73.10 The beginningand end of Cerball’s reign are therefore missing from this account. This is a pity,for what has survived is a most entertaining read.

This section of ‘The Fragmentary Annals’ is a compilation of two mainelements, with minor additions from other sources.11 One major component isthe pseudohistorical ‘Osraige Chronicle’ which is interspersed with entries froma version of the lost ‘Chronicle of Clonmacnoise’. The latter source underliesmost of the mediaeval Irish chronicles.12 The brief, formulaic records of ‘TheChronicle of Clonmacnoise’ stand in contrast to the flamboyant narrative of ‘TheOsraige Chronicle’. These two sources have also been awkwardly combined inrespect of their chronological order, with the result that inconsistencies appearthroughout the text. As an appendix to this article, I have set out the chronologyof entries deriving from ‘The Osraige Chronicle’ in the relevant section of ‘TheFragmentary Annals’. Other elements in this section of ‘The Fragmentary Annals’seem to have been culled from various sources.13 The tenuous nature of thetransmission of ‘The Osraige Chronicle’ serves as a reminder of how manymediaeval Irish texts must have been lost. It is also a testimony to the work of theseventeenth-century scribes who laboured to preserve them.

6Future references to ‘The Osraige Chronicle’ (now lost) rely on this analysis.7Fragmentary Annals, ed. & transl. Radner, pp. xxv–xxvii; Radner, ‘Writing history’, p. 322.8Ibid., p. 323.9Bruxelles, Bibliothèque royale de Belgique, MS. 5301–5320 (4641), folios 3–38: Joseph van

den Gheyn et al., Catalogue des manuscrits de La Bibliothèque royale de Belgique (13 vols, Bruxelles1901–48), VII.48–50.

10Fragmentary Annals, ed. & transl. Radner, pp. 89–148.11Ibid., pp. xxix–xxx.12D.N. Dumville, ‘A millennium of Gaelic chronicling’, The Medieval Chronicle 1 (1999) 103–15.13See, for example, D. Ó Corráin, ‘Viking Ireland – afterthoughts’, in Ireland and Scandinavia

in the Early Viking Age, edd. Howard B. Clarke et al. (Dublin 1998), pp. 421–52, at p. 447.


The character of the entries from ‘The Osraige Chronicle’ suggests that, onthe one hand, the author sought to entertain his audience with tales of heroism,passion, and bloodshed, based on earlier accounts; on the other, he sought toeulogise Cerball and to justify the ambitions of later overkings of the Osraige torule large parts of Ireland. It can be argued that the author remodelled theninth-century past to suit literary tastes and political aims of the eleventh century.This rather qualifies the usefulness of ‘The Osraige Chronicle’ as a historical sourcefor earlier centuries.

As Radner has noted, ‘The Osraige Chronicle’ ‘represents an important stagein the development of Middle Irish dynastic propaganda’.14 As such, it can becompared with the famous twelfth-century sagas Cocad Gaedel re Gallaib andCaithréim Chellacháin Chaisil.15 These texts share a common concern with usingrecords of the past (notably genealogies and annals) to give a sense of authenticityto their interpretation of events. These records were then reïnterpreted andelaborated to eulogise the hero of the saga, and his royal descendants.

Joan Radner has suggested that Cerball’s career was presented in ‘The OsraigeChronicle’ as a model for the life of its patron, Donnchad mac Gilla Phátraic.16

Indeed, a comparison of this saga-element of ‘The Fragmentary Annals’ with otherchronicles suggests how records were reïnterpreted in ‘The Osraige Chronicle’ toserve Donnchad’s interests. This can be seen in a series of examples relating to therelations of the Osraige with different population-groups.

‘The Osraige Chronicle’ exaggerated Cerball’s supremacy over Leinster, perhapsas a precursor to Donnchad’s overkingship established in 1033. Its author assertedthat Cerball took tribute from the province, whereas other chronicles merely notethat he raided there.17 ‘The Osraige Chronicle’ also describes a battle betweenCerball and the Leinstermen. Before the battle, Cerball’s supporters are shownto pray to St Ciarán (patron of Seirkieran and the Osraige), while the Leinstermenprayed to St Brigit (patron of Kildare and Northern Leinster).18 The prayers toCiarán are shown to be more successful in bringing victory. In a subtle way,therefore, ‘The Osraige Chronicle’ projected Leinster’s subjection to Donnchadas both preördained and sanctioned by God.

14Fragmentary Annals, ed. & transl. Radner, p. xxvi.15D. Ó Corráin, ‘Nationality and kingship in pre-Norman Ireland’, Historical Studies (Irish

Conference of Historians) 11 (1975) 1–35, at p. 31; M. Ní Mhaonaigh, ‘Friend and foe: vikings inninth- and tenth-century Irish literature’, in Ireland and Scandinavia, edd. Clarke et al., pp. 382–402, at p. 401.

16Radner, ‘Writing history’, p. 323.17Fragmentary Annals, ed. & transl. Radner, pp. 102–5 (§260).18Ibid., pp. 140–3 (§387).

CERBALL MAC DÚNLAINGE (842–88) 93

‘The Osraige Chronicle’ also emphasises Cerball’s friendship with the churchesof Leighlin (Co. Carlow) and Seirkieran (Co. Offaly).19 Seirkieran was the burialsite of the royal dynasty of the Osraige, and a northern outpost of their power.20

It was closely linked with Leighlin, which was one of the most important churchesin Leinster. During Donnchad’s reign one individual held high office in bothchurches.21 He would have been an important ally to support Donnchad’s powerin both Ossory and Leinster. These eleventh-century links could explain whyflattering references to Cerball’s relations with Seirkieran and Leighlin wereincluded in ‘The Osraige Chronicle’.

Cerball’s dealings with vikings in ‘The Osraige Chronicle’ are also manipulatedin a way which could reflect positively on Donnchad. Descriptions of vikings inthe text are unflattering, while Cerball’s power over them is exaggerated.22 It seemsthat Donnchad had good reason to oppose vikings: his father had been killed bythe men of Waterford in 996.23 As overking of Leinster, Donnchad also soughtto bring the viking-towns of the province under his control.24 Cerball’s dealingswith vikings, as portrayed in ‘The Osraige Chronicle’, might have helped to justifyDonnchad’s agenda for the viking-towns.25

The leaders of rival provinces, Munster and Meath, are denigrated in ‘TheOsraige Chronicle’. Cerball is said to collect the hostages of Munster on behalfof Mael Sechnaill, overking of Mide (while other chronicles report that MaelSechnaill did this himself ).26 According to ‘The Osraige Chronicle’, the peopleof Munster appealed to Cerball for help and then acted treacherously towards him(again, this is unverifiable from other chronicle-accounts).27 Mael Sechnaill ofMide was portrayed in ‘The Osraige Chronicle’ as a coward, only able to asserthis authority over Cerball through the aid of a sorcerer.28 This negative

19Ibid., pp. 110/11 (§281), 130/1 (§365), 140–3 (§387).20Ibid., p. xxiv, n. 43. See also A. Harrison (ed. & transl.), ‘Séanad Saighre’, Éigse 20 (1984)

136–48 (I am grateful to Dr Margo Griffin-Wilson for drawing this article to my attention).21Annala, ed. & transl. O’Donovan, II.752/3 (1004[=1005].5) and 856/7 (1050.1a). Fogartach,

abbot of Leighlin and Seirkieran, died in 1005, and Cléirchén, bishop of Leighlin and ‘the head ofthe piety of Ossory’, died in 1050.

22See further chapter 6, below.23Annala, ed. & transl. O’Donovan, II.734/5 (995.3).24Radner, ‘Writing history’, p. 324.25Annala, ed. & transl. O’Donovan, II.808/9 (1024.9); The Annals of Tigernach, ed. & transl.

Whitley Stokes (2nd edn, 2 vols, Felinfach 1993), II.252 (1022.2).26Fragmentary Annals, ed. & transl. Radner, pp. 96/7 (§246).27Ibid., pp. 100/1 (§254) and 116/17 (§314).28Ibid., pp. 102–5 (§260) and 110/11 (§279). There is no surviving account from ‘The Osraige

Chronicle’ dealing with Cerball’s subjection to Mael Sechnaill at Rathugh.


presentation corresponds with Donnchad’s attacks against both Munster and Mideduring his reign.29

Several features of Cerball’s career described in ‘The Osraige Chronicle’ canbe seen to have parallels with Donnchad’s career. These similarities may havebeen deliberately cultivated. Similarly, Máire Ní Mhaonaigh has demonstratedhow the author of the early twelfth-century propaganda-text Cocad Gaedel reGallaib used the life of Brian Boru as a model for that of his great-grandsonMuirchertach Ua Briain.30 The similar techniques employed in both sagashighlight how they can be considered part of the same literary genre.

The exaggerations and anachronisms found in ‘The Osraige Chronicle’ imposea degree of caution in its use as a historical source for the life of Cerball macDúnlainge. Nevertheless, there is much interesting material within it, and it mayhave drawn on earlier sources of Osraige-history which have since been lost. Thesaga is also central to understanding the development of a post mortem cult of KingCerball.

Icelandic textsIn the Middle Ages stories of Cerball’s deeds travelled as far as Iceland. Accordingto Landnámabók (a twelfth-century text supplemented substantially in thethirteenth century) a number of Icelandic families claimed descent throughCerball’s sons and daughters.31 References to such marriages, and to Cerball asking of Dublin, can also be found in Icelandic sagas. Donnchadh Ó Corráin hassuggested that these mentions of Cerball in Old-Scandinavian literature do notreflect the transfer of oral memories from the ninth century. Rather, they reflecteleventh-century contacts between Ireland and Iceland.32 Stories concerningCerball may have passed to Iceland following the promotion of his cult byDonnchad mac Gilla Phátraic. If so, the references in Scandinavian literatureindicate the success of Donnchad’s efforts to raise the profile of his great-greatgrandfather, as part of his own cult of power.

29Annala, ed. & transl. O’Donovan, II.822/3 (1031.12) and 836/7 (1039.7); The Annals ofInisfallen (MS. Rawlinson B.503), ed. & transl. Seán Mac Airt (Dublin 1951), pp. 198/9 (1031.6).

30M. Ní Mhaonaigh, ‘Cogad Gáedel re Gallaib: some dating considerations’, Peritia 9 (1995)354–77.

31Cogadh Gaedhel re Gallaibh. The War of the Gaedhil with the Gaill or the Invasions of Irelandby the Danes and Other Norsemen, ed. & transl. James Henthorn Todd (London 1867), pp. 297–302; Ó Corráin, ‘Viking Ireland’, pp. 440–1.

32Ibid., p. 444.


At least one of the marriages of Cerball’s descendants recorded inLandnámabók is chronologically implausible.33 Furthermore, Irish records provideno evidence to support the assertions made about Cerball in Icelandic literature.They may all be fiction. Landnámabók was a partially invented record of theIcelandic past, written ‘with the knowledge that many [settlers] came to Icelandnot from Norway, but from the British Isles’,34 while seeking (in the version inórðarbók) to offset accusations that Icelanders were ‘descended from slaves orscoundrels’.35 Some Icelandic families may have asserted a royal Irish pedigree asa neat solution to both problems. Further research needs to be done on theportrayal of Cerball in Old-Scandinavian literature.

More Irish sources‘The Osraige Chronicle’ and Icelandic sagas are the most frequently cited sourcesfor Cerball’s reign. Nevertheless, they are the least reliable. Fortunately, muchevidence about Cerball can be gleaned from other sources. Irish chronicles arethe most valuable resource. ‘The Annals of Inisfallen’, ‘The Annals of Ulster’, andChronicum Scotorum tell much about his career.36 Records from ‘The Annals ofThe Four Masters’ relating to Cerball’s career need to be treated with caution,however, as some of its entries were drawn from a version of ‘The FragmentaryAnnals’ which has now been lost, and these are less trustworthy.37

Additional sources for Cerball’s career may include royal genealogies, whichshow his relationship to overkings of the Osraige.38 The creation and evolutionof these genealogies are still obscure. In some cases the names of rulers can becross-checked in chronicle-records. The ‘early’ sections of the genealogies,

33Cerball’s daughter Eithne was identified as the mother of Sigurðr, earl of Orkney, who died inbattle in 1014: Orkneyinga saga, ed. Finnbogi Guðmundsson (Reykjavík 1965), pp. 24–5 (§11);Orkneyinga Saga, The History of the Earls of Orkney, transl. Hermann Pálsson & P. Edwards (2ndedn, Harmondsworth 1981), p. 36.

34P.H. Sawyer, ‘Harald Fairhair and the British Isles’, in Les Vikings et leur civilisation. Problèmesactuels, ed. Régis Boyer (Paris 1976), pp. 105–9, at 107–8.

35For the principal published version(s), see Íslendingabók, Landnámabók, ed. Jakob Benediktsson(Reykjavík 1968); for the section of the version quoted here, see The Book of Settlements,Landnámabók, transl. Hermann Pálsson & P. Edwards ([Winnipeg,] Manitoba 1972), p. 6.

36For a guide to these sources, see Gearóid Mac Niocaill, The Medieval Irish Annals (Dublin1975).

37See below, chapter 6.38Corpus Genealogiarum Hiberniae, I, ed. M.A. O’Brien (Dublin 1962; 2nd edn, by J.V. Kelleher,

1976), pp. 15–17 (Genelach Osrithe) and 101–16 (Genelach Dáil Birn .i. Osairge); Muireann NíBhrolcháin, ‘The Prose Banshenchas’ (unpublished Ph.D. dissertation, University College, Galway1980), pp. 265–6, 268 (§§392, 403).


however, tend to reflect political concerns rather than biological reality.39 F.J. Byrnehas argued that during Cerball’s reign a Leinster-genealogy was fabricated for hisdistant ancestors.40 This ploy may have developed later, however, to legitimise theambitions of the overkings of the Osraige in Leinster.

Other evidence of Cerball’s familial links can be found in the twelfth-centuryBanshenchas Érenn (‘The Woman-lore of Ireland’) and in various chronicles. Thesehighlight the significance of marital ties in Cerball’s political career. Accordingto Banshenchas Érenn and ‘The Osraige Chronicle’, Cerball’s sister Lann wasmarried first to Gaethíne, overking of the neighbouring territory of the Loígis,then to Mael Sechlainn son of Mael Ruanaid, overking of the Southern Uí Néill,and finally to Aed son of Niall, overking of the Northern Uí Néill.41 Each of Lann’smarriages brought considerable benefit to Cerball who made alliances with herhusbands. Cerball is said to have married a daughter of Mael Sechlainn son ofMael Ruanaid, thus cementing a political alliance developed between them.42

Cerball’s daughters married into the Southern-Leinster dynasties of Uí Dróna andUí Cheinnselaig.43 These ties indicate Cerball’s concern to secure or extend thesoutheastern borders of the Osraige.

‘The Fragmentary Annals of Ireland’ reports that many praise-poems werecomposed for Cerball. Oengus, abbot of Clonfertmulloe (Kyle, bar. Clandonagh,Co. Laois), is said to have composed the most, but none of these poems hassurvived.44 The record suggests that Cerball enjoyed good relations with thechurch of Clonfertmulloe, which lay near the border of the Loígis. In addition,Cerball probably enjoyed good relations with the church of Seirkieran, burial siteof the overkings of the Osraige. A cross-incised slab at Seirkieran bears theinscription Or. do Cherball, and it may commemorate the king in question. AsJoan Radner has noted, the style of the cross would be consistent with a lateninth-century date.45

39David N. Dumville, ‘Kingship, genealogies and regnal lists’, in Early Medieval Kingship, edd.P.H. Sawyer & I.N. Wood (Leeds 1977), pp. 72–104, at p. 74, reprinted in his book Histories andPseudo-histories of the Insular Middle Ages (Aldershot 1990), essay XV, with the same pagination.

40Francis John Byrne, Irish Kings and High-kings (2nd edn, Dublin 2001), p. 163.41Fragmentary Annals, ed. & transl. Radner, pp. 96/7 (§246), 118/19 (§327), 130–5 (§366);

Ní Bhrolcháin, ‘The Prose Banshenchas’, pp. 265–6 (§392).42Ibid., p. 268 (§403).43Chronicum Scotorum. A Chronicle of Irish Affairs, from the Earliest Times to A.D. 1135; with a

Supplement containing the Events from 1141 to 1150, ed. & transl. William M. Hennessy (London1866), pp. 190/1 ([917]=[918].3); Fragmentary Annals, ed. & transl. Radner, pp. 176/7 (§441),176–9 (§443).

44Ibid., pp. 104/5 (§265); cf. ibid., p. 200, n. 31.45Ibid., p. xxiv, n. 43.


Another source relevant to an understanding of Cerball’s career is the politicalgeography of the Osraige-kingdom itself. Only the barest outline of this topicwill be given here. The overkingdom of the Osraige was focused on the NoreValley. Economically the most significant areas were the fertile lowlands south ofKilkenny, and the northern lowlands which now comprise the southwesternbaronies of Co. Laois. The boundaries of the diocese of Ossory provide someguide to the extent of Osraige-territory in the early twelfth century, although it isprobable that the boundaries fluctuated over time.46 The eastern boundary of theOsraige corresponded, more or less, with the watershed of the River Barrow. Thewestern boundary was partly delineated by the River Suir, the uplands ofSlieveardargh, and the watershed of the Nore. Slieve Bloom marked the northernboundary. As I have already mentioned, Seirkieran may have served as a northernoutpost of the Osraige during Cerball’s reign. It is possible that the church waslinked to the rest of the overkingdom by a narrow strip of territory across SlieveBloom.

Chronicle-records suggest that the most important churches of the Osraige inthe ninth century were Seirkieran, Aghaboe, and Freshford.47 Remains of highcrosses indicate that the southwestern churches of Kilkieran, Killamery, and Kilreemay also have been significant, although written sources are lacking.48 As far assecular sites go, ringforts are more densely distributed in the north ofOsraige-territory, near the borders of Munster, Leinster, and Mide.49 The patternmay reflect defensive needs in these border-areas. As the Osraige lay between theheartlands of the provinces of Munster and Leinster, the region was politicallysignificant for any ruler who sought to extend his sway across southern Ireland.Passes on the borders of the Osraige – notably Gowran, and Graine (par.Urlingford, bar. Crannagh) – thus became the focus of political rivalries duringthe ninth century.50 To a certain extent, the success of overkings of the Osraigedepended on the relative strength of neighbouring polities and their own abilityto play off one rival against another.

46William Carrigan, The History and Antiquities of the Diocese of Ossory (4 vols, Kilkenny 1905),I.1–26.

47Ibid., II.1, 26, 125, 246.48E. Rynne, ‘Ireland’s earliest “Celtic” high crosses: the Ossory and related crosses’, in Early

Medieval Munster. Archaeology, History and Society, edd. Michael A. Monk & J. Sheehan (Cork1998), pp. 125–37, at p. 126.

49Matthew Stout, The Irish Ringfort (Dublin 1997), pp. 82–93.50Annala, ed. & transl. O’Donovan, I.548/9 (891[=896].13), 550/1 (893[=898].8).


POLITICAL BACKGROUND

Cerball had both the luck and the ability to use political developments to hisadvantage. At the beginning of his reign, the Osraige were considered part of theprovince of Munster, and Cerball owed allegiance to its overkings. However, fromthe mid-ninth century, the provincial overkingship suffered a reversal in fortunes.This enabled Cerball to exercise a greater degree of political independence than hispredecessors had found possible. This temporary lapse in Munster’s power can betraced in the events following the death of Feidlimid mac Crimthainn in 847. Thishighly successful politician has been described as ‘one of the most colourfulcharacters in the whole of Irish history’.51 His immediate successor, Ólchobar sonof Cinaed (847–51) led successful campaigns against vikings, but he wasshort-lived. A series of weaker overkings of Munster followed. Cerball exploitedthis misfortune. He led several campaigns in flagrant breach of the control of theMunster overkings, and against rebellious subgroups within the province.

Leinster, the eastern neighbour of the Osraige, was also beset by politicaldifficulties during the central years of the ninth century. A series ofviking-campaigns in northern Leinster followed immediately after the foundationof the viking-longphort at Dublin in 841.52 This helped to weaken Uí Dúnlainge,the northern provincial overkings. Their decline was exacerbated as a result ofboth dynastic infighting and intervention by Uí Néill. For example, the overkingRuarc mac Broen was slain by Uí Néill in 862, and vikings killed his successor inthe following year. ‘The Annals of Ulster’ represent this downturn by assigningmore modest titles to kings of Uí Dúnlainge.53 Their head is identified as ‘a kingover the Laigin’ in 835, ‘the king of Uí Dúnlainge’ in 862, and merely as ‘the kingof Naas and Airther Liphi’ in 863.54

The weakening of the eastern and western neighbours of the Osraige can beseen to have assisted in Cerball’s political advance.55 A sense of their decline isreïnforced by the emergence of other marginal groups into the political limelightin these years, including Ciarraige Luachra in Munster and the Loígis in Leinster.Among the difficulties faced by the overkings of Leinster and of Munster in theseyears, vikings and Uí Néill figure prominently. These political rivals also had asignificant impact on Cerball’s career.

51Donncha Ó Corráin, Ireland before the Normans (Dublin 1972), p. 97.52The Annals of Ulster (to A.D. 1131), I, edd. & transl. Seán Mac Airt & G. Mac Niocaill (Dublin

1983), pp. 298/9 (841.4, comprising two entries).53Byrne, Irish Kings, p. 163.54The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.292/3 (835.1), 318/19 (862.6),

318/19 (863.3).55Byrne, Irish Kings, pp. 162–3; Ó Corráin, Ireland, p. 99.


Cerball had to contend with the growing ambitions of the Uí Néill dynasty inrespect of southern Ireland. Mael Sechnaill son of Mael Ruanaid of ClannCholmáin (846–62) was one of the most successful overkings of Uí Néill.56 Hesucceeded in gaining recognition of the power of Uí Néill in Munster and Leinster,a position which necessarily involved him with Cerball.

During Cerball’s reign there was also a shift in the pattern of viking-campaignsfrom the East Midlands and the Shannon-basin to the valleys of the Rivers Noreand Barrow, which adjoin the territory of the Osraige. These activities can beassociated with the foundation of a viking-camp at Waterford and an increase ofviking-activity in Munster during the central years of the ninth century.57 Inconsequence, Cerball was forced to have extensive dealings with vikings. It is inhis battles against them that Cerball has been most famously commemorated.

EVENTS IN CERBALL’S CAREER

Cerball’s first military victory over vikings is recorded in 846. In this year, vikingsseem to have invaded the Osraige from across the Leinster-border. The church atCoolcashin (bar. Galmoy, Co. Kilkenny) was one of the victims of their attack. Itwas located near to the border of the Loígis and the Osraige.58 According to ‘TheAnnals of The Four Masters’, in this year the fleet of the ‘Cailli’ (loinges nacCaillech) attacked Cúl Maine, and Cerball besieged them for a fortnight beforeslaughtering them in battle. Two possible locations for Cúl Maine were suggestedby Edmund Hogan: either Clonmany in Co. Donegal, or Collooney in Co. Sligo.59

However, both sites are far beyond the sphere of Cerball’s other activities at thisstage in his career. It seems more likely that Cúl Maine was in or nearOsraige-territory but has not yet been identified.

There is also some difficulty in identifying the fleet which Cerball defeated,although it has been assumed that it consisted of vikings. Loinnges na cCaillechliterarally translates as ‘the fleet of the nuns’, but this is a rather unlikely prospect!Nor does Caillech seem to derive from a personal or population-group name. Onepossibility, first raised by Charles O’Conor, is that it derives from a river-name.60

It may be linked with the River Burren in Co. Carlow. This used to be called

56Ibid., pp. 99–101; Byrne, Irish Kings, pp. 262–7.57The first reference to Waterford is in relation to 860: Annala, ed. & transl. O’Donovan, I.492/3

(858[=860].5). See also chapter 7, below.58Annala, ed. & transl. O’Donovan, I.470/1 (844[=846].7); Edmund Hogan, Onomasticon

Goedelicum Locorum et Tribuum Hiberniae et Scotiae. An Index, with Identifications, to the GaelicNames of Places and Tribes (Dublin 1910), p. 318 (s.n. Cúl Caissine).

59Ibid., p. 321 (s.n. Cúl Maine).60Annala, ed. & transl. O’Donovan, I.470, n. 1.


Caille, a guttural-stem noun, which is rendered Caillech in the genitive singularand genitive plural.61 As John O’Donovan noted, (Loinnges) na cCaillech is givenas a genitive plural in ‘The Annals of The Four Masters’, demonstrated by thenasalisation of the initial consonant. However, the river-name should be renderedin the singular. It is possible that the mutation was added in an error of copying,if the original meaning of Caille was forgotten, and if it was interpreted as agenitive plural (perhaps under the influence of caillech, ‘nun’). In other words,the compilers of ‘The Annals of The Four Masters’, or an antecedent redactor,may have over-corrected the grammar, by adding nasalisation where it was notnecessary. It is possible that this fleet originated in Carlow.

In 847, according to two chronicles, Cerball achieved another military victory,this time over the vikings of Dublin, at Carn Brammit, an unidentified site. Heremany vikings were killed with their leader Hákonn.62 The 840s were a decadewhen a number of longphuirt or viking-bases were established in Ireland.63 It ispossible that Hákonn and his followers were seeking to establish a base in Cerball’sterritory or seeking a route, across the overkingdom of the Osraige, from Dublinto the southwest of Ireland.

No further viking-activity is recorded which involved the Osraige until the late850s. In the intervening years, vikings were often locked in battles with eachother, or against the overking of Uí Néill, Mael Sechnaill. In the interim, it seems,Cerball became involved in Leinster-politics. In 848, the overking of UíCheinnselaig, Echthigern son of Guaire, invaded the Osraige and won a battle atUachtar Garada (location uncertain).64 Cerball took his revenge by participatingin the treacherous murder of Echthigern in 853.65

In the late 850s vikings stepped up their activities in southwestern Ireland.‘The Osraige Chronicle’ (and derivative entries in ‘The Annals of The FourMasters’) records a feud between Cerball and a viking-leader named Rauðúlfr.According to this saga-account, Rauðúlfr plundered the Osraige about 856.However, Cerball defeated his troops at a ford over the River Nore at Áth Muiceda

61The King’s River, Co. Kilkenny, is named Caille, but this belongs to a different (nasal)stem-class.

62Cerball’s troops are said to have killed 1,200 men, but the frequent recurrence of this numberin estimates of battle-deaths casts doubt on its accuracy. Cf. Annala, ed. & transl. O’Donovan,I.474/5 (845[=847].10); The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.306/7(848.5–6).

63Sites include Annagassan/Linn Duachaill (841), Cork (848), Dublin (841), Limerick (845),Lough Ree (844), and Loch Swilly (842).

64Annala, ed. & transl. O’Donovan, I.476/7 (846[=848].11).65The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.312/13 (853.3).


(Ballamuicca, bar. Ballyconra, Co. Kilkenny).66 Cerball then allied with a rivalband of vikings (‘The Dark Foreigners’, Dubgaill) under the leadership of Horm(Old-Scandinavian Ormr). Together they defeated ‘The Fair Foreigners’ (Finngaill)who were plundering Munster. After Horm left for Wales, where he was killed byKing Rhodri ap Merfyn, Cerball retained the support of those of Horm’s followerswho remained in Ireland.67 These entries from ‘The Osraige Chronicle’ may besuspect in detail. Nevertheless, some corroboration is provided by other sources.For example, ‘The Annals of Ulster’ and Chronicum Scotorum report the death ofHorm in Wales in 856.68

‘The Osraige Chronicle’ asserted that Cerball won two more battles againstRauðúlfr. The first took place about 861, at Slievemargy, on the southeasternborder of Loígis-territory. This report has an air of divine vengeance, as Rauðúlfr’sfollowers are said to have recently plundered the church of Leighlin.69 Cerballthen allied with his nephew Cennétig mac Gaethíne, overking of the Loígis, todestroy Rauðúlfr’s camp.70 John O’Donovan identified this site as Dunrally (DúnRothlaibh) near Vicarstown on the River Barrow.71 Eamonn Kelly and John Maashave suggested that some earthworks in this area mark the site, which was on theborders of the Loígis; however, this identification remains uncertain.72 Rauðúlfr’scampaigns and the battles waged against him by Cerball and Cennétig aredescribed only in ‘The Fragmentary Annals’ and ‘The Annals of The FourMasters’. Both sources derive their accounts ultimately from ‘The OsraigeChronicle’.73 The stories concerning Rauðúlfr may have been partly or whollyinvented in the eleventh century, or they may have developed from a local sourcewhich has since been lost.

There is clearer evidence that Cerball made an alliance with Ívarr, a king of‘the Dark Foreigners’, in the late 850s. Ívarr was one of the most formidablemilitary leaders in ninth-century Insular history, and Cerball may have felt that it

66Fragmentary Annals, ed. & transl. Radner, pp. 98/9 (§249); Carrigan, The History, I.38; D. ÓMurchadha, ‘A reconsideration of some place-names from Fragmentary Annals of Ireland ’, Ainm 8(1998–2000) 41–51, at p. 42.

67Fragmentary Annals, ed. & transl. Radner, pp. 100/1 (§254), 102–5 (§260).68The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.314/15 (856.5); Chronicum

Scotorum, ed. & transl. Hennessy, pp. 154/5 ([856].7). This is not reported in either AnnalesCambriae or Brenhinoedd y Saeson (cf. n. 110, below).

69Fragmentary Annals, ed. & transl. Radner, pp. 110/11 (§281).70Ibid., pp. 114/15 (§308).71Annala, ed. & transl. O’Donovan, I.496, n. x.72E.P. Kelly & J. Maas, ‘Vikings on the Barrow: Dunrally Fort, a possible viking longphort in

County Laois’, Archaeology Ireland 9 (1995), no. 3, pp. 30–2; R. Ó Floinn, ‘The archaeology of theearly Viking Age in Ireland’, in Ireland and Scandinavia, edd. Clarke et al., pp. 131–65, at p. 163.

73See chapter 6, below.


was more beneficial to ally with this formidable character than to oppose him.74

In 858, they defeated a force of Gallgoídil (‘Foreigner-Gaels’), a group of mixedculture or ethnicity, who had been battling against Ívarr in the previous year.75

This group seems to have travelled from the northern half of Ireland and was alliedwith Cenél Fiachach, who occupied the southwestern borderlands of Mide. Theengagement took place at Ara Tíre (bar. Arra or Duharra, Co. Tipperary) inMunster.76 According to ‘The Osraige Chronicle’, Mael Sechnaill was troubledby Cerball’s political strength, and he undertook an expedition to take hostagesfrom the Osraige and Munster in the same year.77 In response, Cerball raided theSouthern Uí Néill with Ívarr and Óláfr, another king of ‘the Dark Foreigners’, inthe following winter.78

In an attempt to restore peace, a royal meeting (rígdál) was arranged at Rahugh(Co. Westmeath), in the territory of Cenél Fiachach, in 859. The relatively lengthyrecord of this meeting in Irish sources indicates that it was perceived as significantby contemporaries.79 Cerball and Mael Guala, overking of Munster, submitted toMael Sechnaill. As a result, the overkingdom of the Osraige was formally cededfrom the authority of Munster and placed under the authority of Mael Sechnaill.Cerball also paid dues to Féthgna, abbot of Armagh.80 The event confirmed MaelSechnaill’s success in establishing his supremacy in southern Ireland, and hisachievements were to influence the ambitions of later overkings of Uí Néill.81 Itmay be that the marriage between Cerball’s sister Lann and Mael Sechnaill wasarranged at this time to strengthen the new political arrangement. The event

74On Ívarr, see Clare Downham, Viking Kings of Britain and Ireland. The Dynasty of Ívarr toA.D. 1014 (Edinburgh 2007; 2nd edn, 2008).

75Fragmentary Annals, ed. & transl. Radner, pp. 104/5 (§263).76Chronicum Scotorum, ed. & transl. Hennessy, pp. 156/7 ([858].3); Hogan, Onomasticon, p. 34

(s.n. Ara Tíre).77Fragmentary Annals, ed. & transl. Radner, pp. 102–5 (§260).78Ibid., pp. 104/5 (§264), 106/7 (§268); The Annals of Inisfallen, ed. & transl. Mac Airt,

pp. 132/3 ([859].2).79Annala, ed. & transl. O’Donovan, I.490–3 (857[=859].3); The Annals of Ulster, edd. & transl.

Mac Airt & Mac Niocaill, I.316/17 (859.3); Chronicum Scotorum, ed. & transl. Hennessy, pp. 156/7([859].3); Fragmentary Annals, ed. & transl. Radner, pp. 106/7 (§268).

80Previously, the Osraige may have resisted recognising the supremacy of Armagh. This issuggested by the Life of St Ciarán, which has been dated to the eighth century: R. Sharpe, ‘Quatuorsanctissimi episcopi: Irish saints before St Patrick’, in Sages, Saints and Storytellers. Celtic Studies inHonour of Professor James Carney, edd. Donnchadh Ó Corráin et al. (Maynooth 1989), pp. 376–99;I. Sperber, ‘The Life of St Ciarán of Saigir’, in Offaly – History and Society. Interdisciplinary Essayson the History of an Irish County, edd. William Nolan & T.P. O’Neill (Dublin 1998), pp. 131–52.

81Ó Corráin, Ireland, pp. 99–100.


caused Cerball to reject his alliance with Ívarr and Óláfr. This threatened toundermine viking-influence in Munster. Perhaps in retaliation, vikings stonedMael Guala to death soon after the meeting.82 Cerball was engaged in conflictswith vikings (including his erstwhile allies) in the following year.

In 860, ‘The Osraige Chronicle’ reported, Cerball defeated a fleet of vikingsfrom Waterford at Achad meic Erclaige (near present-day Kilkenny City, on thebanks of the River Nore).83 This is the earliest chronicle-reference to the existenceof a viking-settlement at Waterford. The account in ‘The Fragmentary Annals’gives a highly dramatised description of the event, relating how Cerball was ableto despatch two enemy-fleets while suffering a hangover: ‘the amount which hehad drunk the night before hampered him greatly; and he vomited much, andthat gave him immense strength’.84 It is possible that the saga-account developedfrom an earlier written record or an oral account, but the source of the informationis lost. Because of this, its accuracy is uncertain.

The other major event in which Cerball was involved in 860 is partlyconfirmed by other chronicle-reports. Mael Sechnaill led a coalition of troopsfrom Leinster, Munster, Connaught, and the Southern Uí Néill, to Moy (nearArmagh, Co. Tyrone).85 ‘The Osraige Chronicle’ identified Cerball as one of themembers of this troop. One night, Aed, overking of the Northern Uí Néill, andÓláfr, a viking-king, attacked this army but were defeated.86 The eventdemonstrates Cerball’s political transition from being an ally of Óláfr to becominghis enemy. Similarly, after the death of Mael Sechnaill in 862, Cerball forged anew alliance with Aed, who had been his enemy in this conflict at Moy.

It can be argued that, during the 850s and early 860s, Uí Néill and differentviking-groups vied for influence in Munster. All were seeking to benefit from thepolitical weakness of contemporary provincial overkings. Mael Sechnaillsuccessfully led expeditions to gather hostages in Munster in 854 and 858, whileat least three different viking-groups were active in the region. In 856, vikingskilled Gormán, heir to provincial overkingship of Munster.87 Then in 859, as

82Fragmentary Annals, ed. & transl. Radner, pp. 104/5 (§264); Chronicum Scotorum, ed. & transl.Hennessy, pp. 156/7 ([859].4).

83Fragmentary Annals, ed. & transl. Radner, pp. 108/9 (§277).84Ibid.: ‘tainig ris go mór a méd attibh an aidhche remhe, 7 ra sgé go mór, 7 tug sonairte mór

do-s[o]mh sain’.85Annala, ed. & transl. O’Donovan, I.492/3 (858[=860].3); The Annals of Ulster, edd. & transl.

Mac Airt & Mac Niocaill, I.316/17 (860.1); Chronicum Scotorum, ed. & transl. Hennessy, pp. 156/7([860].1).

86Fragmentary Annals, ed. & transl. Radner, pp. 110/11 (§279).87Annala, ed. & transl. O’Donovan, I.486/7 (853[=855].6); Chronicum Scotorum, ed. & transl.

Hennessy, pp. 154/5 ([856].5), a marginal addition of uncertain status. The two reports, it must besaid, do not fit very comfortably together.


already mentioned, they killed the provincial overking, Mael Guala. ‘The OsraigeChronicle’ also described perilous journeys to or from Limerick, undertaken byvikings in attempts to establish overland-routes between settlements.88 Even ifthese travel-stories were invented in the eleventh century, they nevertheless reflectplausible concerns in the ninth century as vikings attempted to establishthemselves in Munster. It seems that Cerball’s method of dealing with these rivalforces was to ally with those whom he considered to be the most fearful as enemies.He associated first with ‘the Dark Foreigners’ and then with Uí Néill. BecauseCerball was an able military leader who controlled some of the main routes intoMunster, both vikings and Uí Néill seem to have been willing to court his supportto bolster their influence in the region.

In 861, Cerball felt secure enough in his power within his overkingdom of theOsraige to hold an oenach (‘assembly’) at Mag rRaigne.89 This plain representedthe political heartland of the Osraige, comprising much if not all of the lowlandstretching from the River Nore to Slievardagh and Slievenamon.90 The oenach asan institution served many purposes in mediaeval Ireland, as various scholars havedemonstrated.91 On the one hand it could bolster royal power as an occasion toenact law and receive tribute, and to have royal genealogy publicly recited. Itmade manifest the figure of the king to a wide cross-section of his subjects. Anoenach could also win popularity for a king, as it fulfilled a wide range of socialfunctions for his people. It combined aspects of a country-fair, a sporting event,and a festival of music and storytelling. It also served as a place to agree contracts.In summing up the functions of the oenach, Catherine Swift has concluded thatit was ‘a safety valve through which a community could resolve any internaltensions’.92 Cerball’s ability to hold this assembly symbolised his power, sincesuch a meeting required a degree of peace within his territories.

In order to maintain political stability, Cerball allied with the Loígis for defenceagainst viking-attacks in the early 860s. His most determined enemy at this pointwas perhaps Rauðúlfr, who is mentioned above.93 Nevertheless, Cerball was alsoprepared to enlist the support of viking-groups when it suited him. In 864 he

88Fragmentary Annals, ed. & transl. Radner, pp. 108–11 (§278), 122/3 (§337).89Ibid., pp. 110/11 (§280).90Ibid., pp. 94/5 (§236).91D.A. Binchy, ‘The Fair of Tailtiu and The Feast of Tara’, Ériu 18 (1958) 113–38; C. Doherty,

‘Exchange and trade in early medieval Ireland’, Journal of The Royal Society of Antiquaries of Ireland110 (1980) 67–89; C. Swift, ‘Óenach Tailten, the Blackwater Valley, and the Uí Néill kings of Tara’,in Seanchas. Studies in Early and Medieval Irish Archaeology, History and Literature in Honour ofFrancis J. Byrne, ed. Alfred P. Smyth (Dublin 2000), pp. 109–20.

92Ibid., p. 118.93See pp. 101–2. Fragmentary Annals, ed. & transl. Radner, pp. 114/15 (§310). Cerball also

defeated a viking-army at Fertagh (bar. Galmoy, Co. Kilkenny) in 863.


allied with a band of Lochlannaig to attack Munster.94 According to ‘The OsraigeChronicle’, in 866 Cerball coöperated again with Cennétig of the Loígis to defeata viking-army in Munster.95 From 866 until 878 the viking-threat waned inIreland. (This coïncided with the departure of Óláfr and Ívarr to Britain, and itlasted until the major defeat of vikings at Edington, which may have causedvikings in Britain to travel to Ireland.)96 The threat posed by vikings did notrecede entirely: Cerball’s son Culén was slain by vikings in 886.97 Nevertheless,the temporary decrease in viking-campaigns in the late 860s enabled Cerball topursue wider political ambitions.

In the remainder of this chapter I shall focus on Cerball’s involvement in theneighbouring provinces of Munster and Leinster. In 864, Cerball led an ambitiouscampaign into the heartlands of Munster (in Co. Tipperary and in the area ofFermoy, Co. Cork).98 The expedition was primarily aimed against theEoganacht-dynasties, who controlled the overkingship of Munster. A branch ofthe Déisi called Uí Oengusa also suffered in the attack, and Cerball is said to havetaken hostages from some of the non-Eoganacht peoples.99 According to ‘TheOsraige Chronicle’ his raid was motivated by revenge, as the Eoganacht had killedsome people who fled from Ossory earlier in that year.100

Cerball later chose to ally with Dúnchad mac Duib dá Bairenn, who becameoverking of Munster in 872. Together they plundered Connaught in 871 and873.101 After the first attack on Connaught, Cerball raided western Munster. Hemay have done so with Dúnchad’s approval in order to curb the growing politicalstrength of Ciarraige Luachra (in northern Co. Kerry).102 After the second attack

94Ibid., pp. 116/17 (§314).95Ibid., pp. 122–5 (§338).96The decline in viking-power in 866 and 867 is marked by the destruction of viking-camps by

Irish kings. The decline in chronicle-records of viking-activity, which lasted until the late 870s,could reflect a real change in viking-activity, and not merely a change in the nature of thechronicle-record. Until 878, Ireland could be considered as a refuge from viking-activity for peoplefleeing Britain: The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.320/1 (866.4), 322/3(867.8), 332/3 (877.3), 334/5 (878.9): Colmán Etchingham, Viking Raids on Irish ChurchSettlements in the Ninth Century. A Reconsideration of the Annals (Maynooth 1996), pp. 15, 16, 52.

97Annala, ed. & transl. O’Donovan, I.536/7 (884[=886].6).98Fragmentary Annals, ed. & transl. Radner, pp. 116/17 (§314); Hogan, Onomasticon, pp. 408–

9, 424.99Annala, ed. & transl. O’Donovan, I.498/9 (862[=864].9).100Fragmentary Annals, ed. & transl. Radner, pp. 116/17 (§314).101Annala, ed. & transl. O’Donovan, I.514/15 (869[=871].7), 518/19 (871[=873].8).102D.N. Dumville, ‘Two approaches to the dating of “Nauigatio Sancti Brendani”’, Studi

medievali, 3rd series, 29 (1988) 87–102, reïssued as ‘Two approaches to the dating of NauigatioSancti Brendani’, in The Otherworld Voyage in Early Irish Literature. An Anthology of Criticism, ed.Jonathan M. Wooding (Dublin 2000), pp. 120–32; see pp. 128–31.


on Connaught, Cerball plundered the Déisi of eastern Munster in 874. Hisalliance with Dúnchad seems to have collapsed soon afterwards. This may be seenin 878 when Cerball joined with the Déisi to defeat the Eoganachta and theirallies at Inneoin (Mullaghnoney, par. Newchapel, Co. Tipperary).103 In sum,Cerball does not seem to have remained the consistent ally of any particular groupin Munster. From the mid-860s his policy towards his western neighbour appearsto have been opportunistic. His activities focused on the territories adjoining theOsraige, but he also undertook several campaigns farther afield in alliance withDúnchad mac Duib dá Bairenn. Although Cerball began his reign subordinateto the provincial overkings of Munster, he rose to a position of influence inMunster-affairs.

Cerball’s attitude towards Leinster from the mid-860s was more uniformlyhostile. In 864 he raided the province; the Leinstermen launched a counter-raidtwo weeks later in league with some vikings.104 Cerball may then have allied withvikings to attack the churches of Shrule, Sleaty, and Agha (Co. Carlow) in 866.105

This struggle between the Osraige and Leinster threatened to continue in 868,but (‘The Osraige Chronicle’ asserted) the abbot of Leighlin intervened to makepeace.106 This agreement was broken two years later when Cerball plunderedLeinster as far as Dún Bolg. This took place while his ally, the overking of UíNéill, Aed son of Niall, attacked the province from the north in order to levytribute.107 In 878 an army from Southern Leinster invaded the Osraige.According to ‘The Annals of The Four Masters’, Cerball slaughtered this army atFulachta. Edmund Hogan identified this site as Foulksrath (par. Coolcraheen,bar. Fassadinin), near the River Nore.108 The marriages of Cerball’s daughters tooverkings of Uí Cheinnselaig and of Uí Dróna suggest a counterbalance to thislist of hostilities.109 In the long term, Cerball may have hoped to end theborder-warfare which dominated his relations with Leinster. He may have alsohoped to make common cause with South-Leinster polities against their northernoverkings.

103Annala, ed. & transl. O’Donovan, I.522/3 (876[=878].9).104Ibid., I.498/9 (862[=864].4); Fragmentary Annals, ed. & transl. Radner, pp. 116/17 (§314).105Annala, ed. & transl. O’Donovan, I.500/1 (864[=866].4); Fragmentary Annals, ed. & transl.

Radner, pp. 126/7 (§345).106Ibid., pp. 130/1 (§365).107Cerball’s nephew Cennétig and Aed son of Niall were both supporters of Mael Ciaráin son

of Rónán, whose death at the hands of the Leinstermen is said to have instigated this attack: ibid.,pp. 140–3 (§387).

108Annala, ed. & transl. O’Donovan, I.522/3 (876[=878].8); Hogan, Onomasticon, p. 432 (s.n.Fulachta).

109Chronicum Scotorum, ed. & transl. Hennessy, pp. 190/1 ([917]=[918].3); Fragmentary Annals,ed. & transl. Radner, pp. 176/7 (§441), 176–9 (§443).


CONCLUSIONS

Cerball’s death is recorded in Irish chroncles for the year 888. News of his deathis also reported in Welsh chronicles.110 Cerball’s dealings with the viking-leaderHorm, who was active in both Ireland and Wales, could explain such a connexion.The evidence from Iceland suggests that stories of Cerball’s adventures travelledalong vikings’ trading routes over a century and a half after his death.

How was it that Cerball enjoyed such success? From the start, Cerball seemsto have been an able military leader. His early victories perhaps earned him afierce reputation which made him a desirable ally for those rulers who wished toextend their power in southern Ireland. Another key to Cerball’s success mayhave been his ability to adapt quickly to changing political situations. He seemsto have made good use of the advantages offered by the relative weakness ofMunster and Leinster in these years. Cerball also dealt ably with the threats offeredby Uí Néill and by vikings who were ambitious for power in southern Ireland.

In the late 860s and early 870s the threat posed to the Osraige by vikings andUí Néill decreased but did not disappear. During these years Cerball flexed hismuscles in neighbouring territories, leading a series of campaigns into bothMunster and Leinster, as well as participating in two invasions of Connaught. Bythese actions Cerball succeeded in establishing the Osraige as an independentpolity and a strong power in southern Ireland.

The final decade of Cerball’s reign was quiet by comparison, and no furthercampaigns are recorded. One might assume that his youthful vigour had passedaway by this stage. Cerball reigned for a total of forty-six years. His long rule, hissuccess, and the dramatic potential of the events of his life made Cerball a suitablerole-model for later kings of the Osraige to emulate and eulogise. Thus it isnecessary to keep in mind the influence which the post mortem development ofCerball’s reputation, and indeed cult, may have on our perception of his career.In this chapter I have attempted to reveal the ‘historical’ Cerball. Nevertheless,the entertaining accounts written long after Cerball’s death, in ‘The OsraigeChronicle’ and in Icelandic sagas, offer an interesting puzzle for historians and anengaging resource for students of literature.


110Annales Cambriae, A.D. 682–954: Texts A–C in Parallel, ed. & transl. David N. Dumville(Cambridge 2002), pp. 14/15 (s.a. 888A); Brenhinoedd y Saeson, ‘The Kings of the English’, A.D. 682–1097: Texts P, R, S in Parallel, ed. & transl. David N. Dumville (Aberdeen 2012), pp. 30/1(s.a. 887PS).

APPENDIX

A suggested chronology of the entries from ‘The Osraige Chronicle’ in Section Four of ‘The Fragmentary Annals of Ireland’

Elsewhere I have argued that ‘The Osraige Chronicle’ was originally composedindependently of ‘The Fragmentary Annals of Ireland’.111 I suggest that theredactor of ‘The Fragmentary Annals’ divided up that text to incorporate it intohis own compilation. By separating the entries from ‘The Osraige Chronicle’ andthe rest of ‘The Fragmentary Annals’, it can be shown that the records run in(rough) chronological order. This can help to date entries where chronologicalinformation cannot be found in other chronicles (viz, by filling in the gaps). Theprovenance of three entries (§§262, 263, 280) is uncertain. The dates given belowwithout question-marks are verified by cross-reference with extant chronicles. I have suggested possible dates, with question-marks, for other entries.112

Entries from ‘The Osraige Chronicle’ in ‘The Fragmentary Annals of Ireland’

ENTRY YEAR ENTRY YEAR ENTRY YEAR

233 851 ?263 858 340 866234 851 265 859 341 ?866/7235 852 266 ?860 342 867236 852 267 861 347 867239 853 277 860 348 867243 ?852/3 278 ?860/1 349 867244 852 279 860 350 ?867/8246 854 ?280 861 365 ?867/8247 856 281 ?861/2 366 868249 ?856 292 862 377 869251 ?856 294 862 387 870252 ?856 314 864 388 870253 ?856 327 866 400 ?254 856 337 866 401 ?260 858 338 ?866 408 ??262 858 339 ?866 410 ?

111See below, chapter 6.112Cf. Fragmentary Annals, ed. & transl. Radner, pp. xxix–xxx, 94–7 (§239), 100/1 (§254), 136–

9 (§377); The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.312/13 (853.2), 314/15(856.6), 324/5 (869.4); Chronicum Scotorum, ed. & transl. Hennessy, pp. 152/3 ([853].1), 154/5([856].7), 162/3 ([869].4). Three further items can perhaps be linked to ‘The Osraige Chronicle’:Fragmentary Annals, ed. & transl. Radner, pp. 104/5 (§262), 104/5 (§263), and 110/11 (§280).


The Medieval Chronicle 3 (2004) ISBN 90-420-1834-8

THE lacunose compilation known as ‘The Fragmentary Annals of Ireland’ survivesonly in a seventeenth-century transcript (Bruxelles, Bibliothèque royale deBelgique, MS. 5301–5320 [4641], folios 3–38).1 The text is in five fragments.It begins with the year 573 and ends at 914. The chronicle as a whole ischaracterised by a combination of short annalistic entries and longerpseudohistorical narratives.2 In this chapter I shall examine the portrayals ofvikings in the pseudohistorical accounts found in the fourth section of thischronicle, which covers the years from 849 to 873. The text exhibits interestingvariation in the ways in which different groups of vikings are portrayed.

Vikings had terrorised the coasts of Ireland from the end of the eighth centuryand began to play a role in Irish politics soon after. As a result, vikings areprominent in ‘The Fragmentary Annals’. However, the hero of the narrative isCerball mac Dúnlainge, an overking who ruled the Osraige from 842 to 888.Cerball is portrayed ‘as a man worthy to possess all Ireland’ (§260),3 and hisworthiness is often measured in relation to his victories over different groups ofvikings. Therefore the portrayal of various viking-contingents can be seen as afoil to enhance the reputation of this king.

The composition of the pseudohistorical narrative about Cerball has beenargued to have been sponsored by his descendant, King Donnchad.4 This highlysuccessful overking extended his sway over the province of Leinster in the 1030sand may have desired to control the wealthy viking-towns on the coast of thatprovince. Donnchad may have developed the cult of his ancestor, Cerball, tojustify his ambitions and to glorify his lineage.

The pseudohistorical narratives in ‘The Fragmentary Annals of Ireland’ can be

6The good, the bad, and the ugly:

portrayals of vikings in ‘The Fragmentary Annals of Ireland’

1Joseph Van den Gheyn et al., Catalogue des manuscrits de la Bibliothèque royale de Belgique(13 vols, Bruxelles 1901–48), VII.48–50.

2Fragmentary Annals of Ireland, ed. & transl. Joan Newlon Radner (Dublin 1978). Radner’stitle for her book could apply to almost every extant Irish Chronicle: I have therefore made her titleminimally specific and renamed the chronicle as ‘The Fragmentary Annals of Ireland’.

3Ibid., pp. 102–5.4Ibid., p. xxvi; cf. J.N. Radner, ‘Writing history: early Irish historiography and the significance

of form’, Celtica 23 (1999) 312–25.

111

seen to belong to a genre of Irish literature which developed in the eleventh andtwelfth centuries, celebrating the victories of earlier Irish kings over theirviking-enemies.5 Contemporary kings seemed eager to benefit from the prestigeof their ancestors who overcame foreign enemies. This xenophobic trend in Irishliterature has been attributed to a growing sense of Irish identity which could bepositively pitched against a negative foreign Other.6 These stories can be seen asa means of justifying Irish control over the viking-settlements in Ireland, whichhad been subject to Irish overkingship from the late tenth century, although theyretained a degree of local autonomy.7

Ninth-century chronicle-reports provided the basis for the pseudohistoricalaccount in ‘The Fragmentary Annals of Ireland’. This can be shown by comparingthe main outline of events in that text with other Irish chronicles. A sense ofhistorical authenticity was clearly important to the author of this pseudohistoricalchronicle, as indeed to the creators of others of its type.8 However, the devil isoften in the detail, for it is in the way in which these events are interpreted in thetext that the eleventh-century author of these narrative sections exhibited hiscreativity and inventiveness. It seems that ninth-century reports were remodelledand elaborated to suit eleventh-century political aims, literary tastes, andperceptions of vikings.

It is fortunate that a good deal of near-contemporary evidence about the careerof King Cerball can be gleaned from other Irish chronicles. This demonstratesthat Cerball was in contact with three main viking-groups during his career.9 Theseare the Finngaill (‘Fair Foreigners’), Dubgaill (‘Dark Foreigners’), and Gall-Goídel(‘Foreigner-Gaels’). These three groups are also represented in ‘The Fragmentary

5D. Ó Corráin, ‘Nationality and kingship in pre-Norman Ireland’, Historical Studies (IrishConference of Historians) 11 (1975) 1–35, and ‘Viking Ireland – afterthoughts’, in Ireland andScandinavia in the Early Viking Age, edd. Howard B. Clarke et al. (Dublin 1998), pp. 421–52, atp. 443; M. Ní Mhaonaigh, ‘Friend and foe: vikings in ninth- and tenth-century Irish literature’,ibid., pp. 381–404, at p. 401; D.N. Dumville, ‘A millennium of Gaelic chronicling’, The MedievalChronicle 1 (1999) 103–15, at pp. 104–5.

6Ó Corráin, ‘Nationality’, pp. 31–2.7S. Duffy, ‘Irishmen and Islesmen in the kingdoms of Dublin and Man, 1052–1171’, Ériu 43

(1992) 93–133, at pp. 94–5.8D. Ó Corráin, ‘Caithréim Chellacháin Chaisil: history or propaganda?’, Ériu 25 (1974) 1–69,

at pp. 60–9; M. Ní Mhaonaigh, ‘Cogad Gáedel re Gallaib and the annals: a comparison’, Ériu 47(1996) 101–26; Radner, ‘Writing history’, p. 320.

9Chronicum Scotorum. A Chronicle of Irish Affairs, from the Earliest Times to A.D. 1135; with aSupplement, containing the Events from 1141 to 1150, ed. & transl. William M. Hennessy (London1866), pp. 146–9 ([847].3), 156/7 ([858].3); Annala Rioghachta Eireann, Annals of the Kingdom ofIreland, by The Four Masters, from the Earliest Times to the Year 1616, ed. & transl. John O’Donovan(2nd edn, 7 vols, Dublin 1856), I.474/5, 490–3, 492/3 (845[=847].10; 856[=858].5; 857[=859].2–3; 858[=860].5).


Annals of Ireland’. The records show that Cerball fought against, and allied with,vikings at different times during his career.

During Cerball’s reign, the Osraige rose from relative political obscurity tobecome major players in Irish politics.10 Because of this, Cerball was a suitablerole-model for later kings of the Osraige to emulate and eulogise. Cerball’spost-mortem fame extended even as far as Scandinavia. According to Landnámabókand Orkneyinga saga, Cerball’s daughters married prominent vikings, and anumber of Icelandic families claimed descent from this Irish overking.11 However,Irish records provide no mention of Cerball having marriage-alliances with vikings.Donnchadh Ó Corráin has argued that these references to ‘Kjarvalr’ inOld-Scandinavian literature owe more to eleventh-century contact between Irelandand Iceland than to ninth-century oral memory.12 It is possible that these storiesdeveloped as Cerball’s hero-cult was promoted in Ireland in the eleventh century.‘The Fragmentary Annals of Ireland’ can therefore be seen as a key-text in thedevelopment of Cerball’s literary persona.

Some discussion of this chronicle is necessary before I analyse the portrayalsof vikings within it. The scribal rubric at the head of ‘The Fragmentary Annalsof Ireland’ gives us some evidence of its textual history. The surviving text is atranscript of a seventeenth-century copy of a book of Giolla na Naomh MacAodhagáin.13 Mac Aodhagáin’s floruitmay be placed around 1400.14 Before thisdate, the textual history of the chronicle is less certain. In the fourth fragment ofthe chronicle, which deals with King Cerball, we can identify two maincomponents of the compilation.

One is the pseudohistorical narrative which celebrates the deeds of Cerball.Joan Radner has called this component ‘The Osraige Chronicle’: it is the focus ofmy discussion. The second main component is a derivative of the lost ‘Chronicleof Clonmacnoise’, an early- or mid-tenth-century source (911×954) which

10Radner, ‘Writing history’, p. 322. See now also chapter 5, above.11Cogadh Gaedhel re Gallaibh. The War of the Gaedhil with the Gaill, or The Invasions of Ireland

by the Danes and Other Norsemen, ed. & transl. James Henthorn Todd (London 1867), pp. 297–302; Ó Corráin, ‘Viking Ireland’, pp. 440–1.

12Ibid., p. 444.13Fragmentary Annals, ed. & transl. Radner, p. vii (her ‘headnote’) and p.1; Nollaig Ó Muraíle,

The Celebrated Antiquary Dubhaltach Mac Fhirbhisigh (c. 1600–1671), his Lineage, Life and Learning(2nd edn, Maynooth 2002), p. 88. For discussion of forms of reference to this chronicle, on theevidence of the sole manuscript, see F.T. Wainright, ‘Duald’s “Three Fragments”’, Scriptorium 2(1948) 56–8 + plate 11, illustrating ‘the title’ on folio lr (p.1) and subheadings on folios 9r (p. 17)and 17r (p. 33).

14Cath Almaine, ed. Pádraig Ó Riain (Dublin 1978), pp. xxvi–xxvii, where the floruit of Giollana Naomh Mac Aodhagáin (senior) was established with reference to the death-date of hishomonymic son (1443).

THE GOOD, THE BAD, AND THE UGLY 113

underlies several mediaeval Irish chronicles.15 Occasionally an entry from anothersource was included. An example is §330,16 an account of the sack of York byvikings in 867 and an adventurous journey to North Africa. Donnchadh ÓCorráin has suggested that this was derived from legends of Ragnarr loðbrók,which were circulating in the eleventh century.17

It is relevant that when the pseudohistorical sections are extrapolated andcompared with other chronicles they can be shown to run in chronological order.The chronological errors which abound in this section of the chronicle tend tooccur at the interface between ‘Clonmacnoise-chronicle’ entries and the longerpseudohistorical narratives.18 It seems that these two texts once existed as separateentities but that they were spliced together to create ‘The Fragmentary Annals ofIreland’ in its present form. The key-point is that the present ordering of thepseudohistorical entries in ‘The Fragmentary Annals of Ireland’ preserves largelyor entirely the sequence of entries found in the antecedent texts.

By identifying the separate sequences of entries in the compilation, we may also be able to fill some of the gaps in the dating of individual annals or annal-entries.

15Dumville, ‘A millennium’, pp. 107–10. For a full study, see Kathryn Grabowski &D. Dumville, Chronicles and Annals of Mediaeval Ireland and Wales. The Clonmacnoise-group Texts(Woodbridge 1984).

16Fragmentary Annals, ed. & transl. Radner, pp. 118–21.17Ó Corráin, ‘Viking Ireland’, p. 447; cf. Rory McTurk, Studies in Ragnars Saga Loðbrókar and

its Major Scandinavian Analogues (Oxford 1991), p. 1.18I have given fuller details in the appendix to this chapter: see pp. 122–5.


Table showing entries from ‘The Osraige Chronicle’ in Section Four of ‘The Fragmentary Annals of Ireland’

Entry- Year Entry- Year Entry- Yearnumber number number

233 851 ?263 858 340 866234 851 265 859 341 ?866/7235 852 266 ?860 342 867236 852 267 861 347 867239 853 277 860 348 867243 ?852/3 278 ?860/1 349 867244 852 279 860 350 ?867/8246 854 ?280 861 365 ?867/8247 856 281 ?861/2 366 868249 ?856 292 862 377 869251 ?856 294 862 387 870252 ?856 314 864 388 870253 ?856 327 866 400 ?254 856 337 866 401 ?260 858 338 ?866 408 ??262 858 339 ?866 410 ?

Notes to this Table

1. This table is based on Fragmentary Annals, ed. & transl. Radner, pp. xxix–xxx.2. The dates given have been verified by cross-reference with otherchronicle-sources: ibid., §§239, 254, 377 (pp. 94–7, 100/1, 136–9).

3. Compare The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill,I.312/13 (853.2), 314/15 (856.6), 324/5 (869.4).

4. Compare also Chronicum Scotorum, ed. & transl. Hennessy, pp. 152/3([853].1), 154/5 ([856].7), 162/3 ([869].4).

5. Fragmentary Annals, ed. & transl. Radner, §§262 (pp. 104/5), 263 (pp. 104/5),and 280 (pp. 110/11), can perhaps be linked to ‘The Osraige Chronicle’.


By isolating the sequence of pseudohistorical entries in our chronicle, we can studynarrative developments in the portrayals of vikings. The eleventh-century authorof the pseudohistorical narrative wished to praise King Cerball in conventionalterms as an Irish hero who fought vikings. However, the author faced a majordifficulty when looking back over the historical records. Cerball had allied withvikings as well as fighting against them.19 Rather than deny the evidence of otherIrish chronicles, the author of the pseudohistory devised ways of justifyingCerball’s actions.

This justification was based on the portrayal of the three main viking-groupswhich Cerball is said to have encountered. In eleventh-century terms the authoridentified Dubgaill (‘Dark Foreigners’) as Danes. They are described as unpleasantbut useful to Irish kings (§§235, 251).20 The Finngaill (‘Fair Foreigners’) heidentified as Norwegians: they were shown to be greedy, treacherous, and heathen(§§243, 254).21 The final group, the Gall-Goídil (‘Foreigner-Gaels’) he describedas Irish foster-children of the vikings – they are said to have adopted vikings’ ways,and they are sometimes called Northmen (§§247, 260).22 This group is presentedas the most despicable of all. It is notable that ‘The Fragmentary Annals of Ireland’is the earliest surviving text to define ‘Dark’ and ‘Fair’ Foreigners in ethnic termsas ‘Danes’ and ‘Norwegians’.

Quality of viking in ‘The Fragmentary Annals of Ireland’

BAD Danes/Dark Foreigners (Danair / Dubgaill) semi-christian

WORSE Norwegians/Fair Foreigners (Lochlannaig / Finngaill) heathen

WORST Foreign-Irish/Northmen (Gall-Goídil / Normannaig) apostate

By categorising viking-groups in levels of depravity, the author of ‘TheFragmentary Annals of Ireland’ developed in two ways a defensive strategy forCerball’s activities. The most powerful overking in Ireland, Mael Sechnaill, wasshown to ally with vikings before Cerball. Furthermore, Cerball was shown toally with some viking-groups with the forgivable intention of defeating worseviking-enemies.

The section under discussion begins abruptly in 851. It opens with adescription of the arrival of ‘Danes’ in Ireland and their first battle against the

19Chronicum Scotorum, ed. & transl. Hennessy, pp. 156/7 ([858].3); The Annals of Ulster (toA.D. 1131), I, edd. & transl. Seán Mac Airt & G. Mac Niocaill (Dublin 1983), pp. 316/17 (859.2).

20Fragmentary Annals, ed. & transl. Radner, pp. 90–5, 98/9.21Ibid., pp. 96/7, 100/1.22Ibid., pp. 96–9, 102–5.


‘Norwegians’ who were already raiding there. The prowess of the new arrivals isapplauded. The ‘Danes’ killed three times their own number and beheadedeveryone whom they killed (§233).23 Furthermore, they are perceived as agentsof divine retribution: ‘Afterwards the Danes seized the women and gold and allthe goods of the Norwegians, and thus the Lord took from them [the Norwegians]all the wealth they had taken from the churches and holy places and shrines ofthe saints of Ireland’ (§233: Ra gabsat tra na Danair ar sain mna, 7 ór, 7 uilemhaithius na Lochlannac[h]; go rug an Coimdhe uatha amhlaidh sin gach maithrugsat a ceallaibh 7 nemeadaibh 7 sgrínib naomh Eireann).

The next reference to the ‘Danes’ again shows them as victors over the‘Norwegians’, and the scene after the battle is described as follows (§235).24

As in tand sin ra chuir Maoilseachlainn, rí Teamhra, teacht[a] uadh d’ionnsoighidh nanDanar. As amlaidh ro bhattar na Danair, ag luchtaireacht ara gcionn, 7 as iad ba gabhlado ccoireadhaibh, cairn do corpaibh na Loc[h]lannac[h], 7 cidh na beara ara mbiodh anfheóil, as for corpaibh Loc[h]lannac[h] no bhídis a leithcinn, 7 an tine ag losgadh na ccorp,go mbiodh an fheóil 7 an meathradh ra chaithsiot an adaigh remhe ag maidhm asa ngailibhamach. Ra battar dna teachta Maoilseachlainn ’ga fféghadh amhlaid sin, 7 ra battar ’gathathaoír um na Danaraibh sin. As eadh ra raidhsiot na Danair: “As amhlaidh sin budhmaith leo-sum ar mbeith-ne.” Clas mór lán aca do ór 7 da airgead da thabhairt do Pádraicc.Uair as amhlaidh ra bhattar na Danair, 7 cinele crabhaidh aca, .i. gabhaid sealad fri fheóil7 fri mhnáibh ar chrabhudh.

‘It was at that time that Máel Sechlainn, king of Temair, sent messengers to the Danes.When they arrived the Danes were cooking, and the supports of the cooking-pots wereheaps of the bodies of the Norwegians, and even the spits on which the meat was [roasting]rested their ends on the bodies of Norwegians, and the fire was burning the bodies, sothat the meat and fat that they had eaten the night before was bursting out of their bellies.The messengers of Máel Sechlainn were looking at them thus, and they were reproachingthe Danes for it. That is what the Danes said: “They would like to have us like that.”They had a huge ditch full of gold and silver to give to Patrick. For the Danes were likethat, and they had kinds of piety − that is, they abstained from meat and from women fora while, for the sake of piety.’

The type of piety expressed by the ‘Danes’, and the image of human flesh roastingwhile the food is being cooked, may be suggestive of near cannibalism and nearheathenism, two traits which represent the apotheosis of evil in mediaevalliterature, as can be seen in the Old-English poem Andreas but is also found inthe Icelandic Sørla saga sterka where the giantess Mána is seen preparing human

23Ibid., pp. 88–91.24Ibid., pp. 90–5.


flesh and horsemeat.25 This image in ‘The Fragmentary Annals of Ireland’ is linkedwith representations of ‘Danish’ generosity and bravery, creating a scene whichseems to be deliberately ambiguous, situating the ‘Danes’ between good and evil.

That the ‘Danes’ are considered to be better than the ‘Norwegians’ is illustratedin succeeding references. The following entry identifies the ‘Norwegians’ asentirely heathen and shortly afterwards shows the ‘Norwegians’ to be levying heavytaxes and breaking their oaths to the Irish (§§236, 239, 243).26 The portrayalindicates a highly unfavourable view of this viking-group.

Such a picture provides a suitable backdrop to Cerball’s ‘great massacre’ of‘Norwegians’, which is soon related (§249).27 It also justifies Cerball’s acceptanceof the request by the ‘Danes’ to ally with him against the ‘Norwegians’ (§251).28

This pact allows more ‘Norwegians’ to be slain (§254),29 and Cerball is shown toinspire great fear among his enemies, demonstrated during a nocturnal attack onhis camp: ‘This is how Cerball came out of his chamber: with a huge royal candlebefore him, and the light of that candle shone far in every direction. Great terrorseized the “Norwegians”, and they fled to the nearby mountains and to the woods’(§277:30 As amlaidh táinig Cearball immach asa grianán 7 rioghchainnel mhórre(a)imhe, 7 rá bhoí soilsi na caindle sin go fada ar gach leith. Ra ghabh úamhanmór na Lochlannaig; 7 ra theichsiot fona sleibhtibh faigsibh dhóibh 7 fona cailltibh).This is not the only incident where ‘Norwegian’ cowardice is exaggerated: in alater battle ‘they fled into the bog and were all killed in the bog, and dogs devouredtheir corpses’ (§338:31 ro theichsiod ’san mhónaidh gur ro marbhaidh ’san mónaidhuile iad, go nduattar coin a ccolla). As in the incident of the cooking spits, thisseems to be regarded as an appropriate end for the various crimes of the‘Norwegians’.

The most heinous people, Gall-Goídil, are introduced part-way through thisfourth section of ‘The Fragmentary Annals of Ireland’. They are initially describedas Gaels who abandon their faith and join Scandinavians in plundering churches(§244).32 In the next allusion they are called vikings’ foster-children who deservedto be killed (§247).33 Finally we are told that ‘though the original Nor<th>men

25Anglo-Saxon Poetry. An Anthology of Old English Poems, transl. S.A.J. Bradley (London 1982),pp. 110–53; cf. N. Lukman, ‘An Irish source and some Icelandic fornaldarsögur ’, MediaevalScandinavia 10 (1977) 41–57, at p. 46.

26Fragmentary Annals, ed. & transl. Radner, pp. 94/5, 94–7, 96/7.27Ibid., pp. 98/9.28Ibid.29Ibid., pp. 100/1.30Ibid., pp. 108/9.31Ibid., pp. 122–5.32Ibid., pp. 96/7.33Ibid., pp. 96–9.


were evil to the churches, these were much worse’ (§260:34 ger bó olc naNormannaigh bunaidh dona h-eaglaisibh, bá measa go mór iad-saidhe). Thus theimage of the Gall-Goídil deteriorates with each description, and emphasis is placedon their apostasy. It can be argued from mediaeval Irish literature (as fromelsewhere in christian writing) that this makes them worse than heathens, as peoplewho knowingly reject God rather than merely being ignorant of Him.35 Thesedescriptions may serve as a prelude to and a vindication of Cerball’s alliance with‘Norwegians’ to wage war against Gall-Goídil in 858 (§263).36

Cerball then proceeds to ally with ‘Norwegians’ against the people of Mide.The only way in which the author defended this action was by referring in thesame entry to praise-poems made for Cerball. There follows a brief lament onIrish disunity (§§265, 266).37 Thus the author promoted the idea that it wasacceptable for Cerball to ally with some vikings as long as he was seen to use themagainst worse afflictions. The images of different viking-groups seem to have beendeliberately manipulated for this purpose.

The final point which I wish to develop concerns the extent to which theportrayal of viking-groups in ‘The Fragmentary Annals of Ireland’ reflectseleventh-century perceptions of viking-identities as much as ninth-centuryrealities. Gall-Goídil are named for a brief period in the mid-ninth century in anIrish context. They reäppear in eleventh-century reference to Galloway inScotland.38 The statement in ‘The Fragmentary Annals of Ireland’ that theninth-century Gall-Goídil were fosterchildren of the vikings has been rejected byrecent commentators.39 Instead it has been argued that they represent a group ofmixed ethnicity who first appear in the Gaelic world in the 850s followingviking-settlement in the Hebrides and Ireland.40

How valid then are the other ethnic interpretations? The identification, madein ‘The Fragmentary Annals’, of eleventh-century Danes (Danair) withninth-century Dark Foreigners (Dubgaill) on the one hand, and Norwegians

34Ibid., pp. 102–5.35See for example The Irish Liber Hymnorum, edd. & transl. J.H. Bernard & R. Atkinson (2 vols,

London 1898), I.68 and II.151, 156; cf. King of Mysteries. Early Irish Religious Writings, transl. JohnCarey (Dublin 1998), p. 35.

36Fragmentary Annals, ed. & transl. Radner, pp. 104/5.37Ibid., pp. 104/5, 104–7.38The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.472/3 (1034.10).39Barbara E. Crawford, Scandinavian Scotland (Leicester 1987), p. 47; David N. Dumville, The

Churches of North Britain in the First Viking-Age (Whithorn 1997), p. 26; D. Ó Corráin, ‘The vikingsin Scotland and Ireland in the ninth century’, Peritia 12 (1998) 296–339, at p. 326; M. NíMhaonaigh, ‘The outward look. Britain and beyond in medieval Irish literature’, in The MedievalWorld, edd. Peter Linehan & J.L. Nelson (London 2001), pp. 381–97, at 391–2.

40Dumville, The Churches of North Britain, pp. 26–9.


(Lochlannaig) with Fair Foreigners (Finngaill) on the other, has recently beencriticised.41 These colour-terms, ‘fair’ and ‘dark’, which were used in the ninthand tenth centuries, cannot satisfactorily be applied to ethnic distinctions such ashair-colour or dress, although various commentators have attempted to do so.Alfred Smyth has argued from the basis of Irish genealogical literature that ‘light’or ‘fair’ is associated with older lineages or individuals. ‘Dark’ is applied toyounger ones. A possible translation for these terms is ‘new’ and ‘old’ Foreigners.42

The Dark Foreigners arrived in Ireland in the mid-ninth century: they weretherefore ‘newer’ than the vikings who were there before. The vikings who werealready established in Ireland were identified as Fair Foreigners following the arrivalof this new group.43

It has been argued that the ethnic distinctions, Norwegians and Danes, areunlikely to have been recognised by Irish chroniclers in the ninth century.Therefore their later equation with the names Finngaill and Dubgaill is mistaken.Dubgaill can be consistently linked with the family and descendants of theninth-century viking-leader Ívarr. David Dumville has suggested that the namedescribes vikings under the leadership of this family.44 Comparison with otherEuropean chronicles also suggests a lack of concern to categorise viking-groups interms of ethnicity in the ninth century. Distinctions are rather based on namedviking-leaders and their situation in an immediate local context. An example isprovided by the vikings of Seine and Loire, two rival groups named in Frankishchronicles.45

One solution to these difficulties is to present the possibility that the authorof ‘The Fragmentary Annals of Ireland’ reïnvented (or adopted a reïnvention) ofthe terms Fair and Dark Foreigner along ethnic lines, guided by politicalcircumstances in the eleventh century. According to ‘The Annals of Ulster’,Danair or Danes as an identifiable group first appeared on the Irish political scene

41David N. Dumville, ‘Old Dubliners and New Dubliners in Ireland and Britain: a Viking-Agestory’, Medieval Dublin 6 (2004) 78–93, reprinted in his book Celtic Essays, 2001–2007 (2 vols,Aberdeen 2007), I.103–22; C. Downham, ‘Viking identities in Ireland: it’s not all black and white’,Medieval Dublin 11 (2009) 185–201.

42A.P. Smyth, ‘The Black Foreigners of York and the White Foreigners of Dublin’, Saga-book 19(1974–7) 101–17.

43The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.310/11 (851.3); Annala, ed.& transl. O’Donovan, I.480–3 (849[=851].4); Chronicum Scotorum, ed. & transl. Hennessy,pp. 150/1 ([851].3–4).

44See n. 41, above. For my full treatment of that family, see Clare Downham, Viking Kings ofBritain and Ireland. The Dynasty of Ívarr to A.D. 1014 (Edinburgh 2007; 2nd edn, 2008).

45C. Gillmor, ‘War on rivers: viking numbers and mobility on the Seine and Loire, 841–886’,Viator 19 (1988) 79–109. Cf. discussion in chapters 3 and 4, above.


in the late 980s when Sveinn Forkbeard began mounting campaigns againstEngland.46

By the 1030s, which is when our narrative seems to have been written, Daneshad become significant players in Insular politics. This decade also witnessedmuch friction in the relationship between Danes and Norwegians, who wereconsequently equated with the inveterate enemies Dubgaill and Finngaill in ‘TheFragmentary Annals of Ireland’.47 It may have made sense to an eleventh-centurychronicler to interpret ninth-century rival groups in terms of contemporarypolitics. As ‘The Fragmentary Annals of Ireland’ is the first surviving text toprovide this ethnic interpretation, the subsequent widespread use of thoseinterpretations deserves to be questioned.

In summary, one can say that the author of the pseudohistorical narrative inthe fourth section of ‘The Fragmentary Annals of Ireland’ based his account onrecords of events found in earlier chronicles. These reports were elaborated andreworked in order to enhance the image of Cerball, overking of the Osraige. Thisserved the political ends of his descendants in the eleventh century. The portrayalsof vikings are thus a blend of fact and fiction, and ‘The Fragmentary Annals ofIreland’ is the earliest known text to make explicit ethnic identifications for thelabels Fair and Dark Foreigner. This pair of identifications has been accepted bymost subsequent historians but may be a retrospective interpretation which wasdeveloped in this fascinating chronicle.48

46The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.418/19 (986.2) and 420/1(986.3). A distinction between Danair and Gaill is suggested by ‘The Annals of The Four Masters’–Annala, ed. & transl. O’Donovan, II.726/7 (989[=990].5): Sluagh na nGall 7 na nDanar 7 Laighen.... Cf. chapter 4, n. 93, above.

47Martin Syrett, Scandinavian History in the Viking Age. A Select Bibliography (Cambridge 2001),p. 75.

48I should like to thank Erik Kooper who made possible my attendance at the 3rd InternationalConference on the Medieval Chronicle (Doorn / Utrecht, 12–17 July, 2002) through a subsidy fromUtrecht University. I should further like to acknowledge his friendly intermediacy in facilitatingthe present republication of this paper. My thanks also go to David Dumville who read and correcteda draught of this paper, and to Dan McCarthy who read and discussed the Appendix.


APPENDIXChronology and structure in Section Four of

‘The Fragmentary Annals of Ireland’

What we seem to have in this section of ‘The Fragmentary Annals of Ireland’ is acombination of two major texts with minor additions. Both of the majorsource-texts seem to have been chronicles. Joan Radner has argued that briefannalistic records from a version of the lost ‘Chronicle of Clonmacnoise’ (whichshe called ‘The Irish World Annals’) provided a chronological backbone into whichlonger pseudohistorical narratives from ‘The Osraige Chronicle’ were inserted.49

Some additional matter seems to have been included in the compilation: thiscannot readily be assigned to either the ‘The Chronicle of Clonmacnoise’ or ‘TheOsraige Chronicle’.

The argument that most of the shorter entries in ‘The Fragmentary Annals’derive from a single chronicle is supported by the succession of kalend-markerswhich are predominantly associated with the shorter ‘Clonmacnoise-chronicle’entries. This is demonstrated in the following table.

Entries with Year Other method of datingkalend-markers within entry237 849240 850241 851245 852256 853(260 *858 12th regnal year of Maelshechlainn)264 859271 856282 857285 *858308 862312 863325 *865331 *865343 866351 867367 869378 870389 871402 872

49Fragmentary Annals, ed. & transl. Radner, p. xiii.


Out of twenty-one kalend-markers there are two cases where one year (858, 865)seems to have two such markers. I have shown these with an asterisk, above. Theseexamples could imply that two texts organised by kalend-markers were used.

Often a kalend-marked entry is followed by a group of short entries belongingto the same year, which may have been lifted together from a single text (forexample, §§238, 257–9, 272–6, 283–5, 287–91, 308, 313, 332–6, 344–5, 352–60, 368–76, 379–86, 390–9, and 403–7).50 Such order does not always prevail,as some of the shorter entries are not chronologically linked to the nearestkalend-marker, and their order is confused (for example, §§295–307, and 361–4).51 For the most part, however, the short entries seem to stem from a singlechronicle-text.

The impression that two major source-chronicles have been spliced togetherin this section of ‘The Fragmentary Annals of Ireland’ is reïnforced bychronological dislocations in the text, which are often introduced at the interfacebetween the shorter annal-entries and the pseudohistorical narratives (for exampleat §§237, 256, 260, 270, 277, 282, 292, 295, 316, 343).52 The two types of entrymaintain a degree of chronological order when considered as separate elements,but they have been poorly combined with respect to the sequence of years.53

The compilation of the two main elements in ‘The Fragmentary Annals’ seemsto antedate the copy in ‘The Book of Mac Aodhagáin’ which was in existence byabout 1400.54 The evidence of ‘The Annals of The Four Masters’ suggests anintermediate stage (or stages) in the textual history: this seventeenth-centurychronicle depends on both main elements of our chronicle. Joan Radner hasidentified twenty-nine entries uniquely shared by ‘The Annals of The FourMasters’ and ‘The Fragmentary Annals’ for the years from 849 to 873.55

Occasionally ‘The Annals of The Four Masters’ has a fuller record of events inthese entries, which indicates some relationship with another, earlier, version ofour chronicle (§§293, 318).56

The dating of the compilation called ‘The Fragmentary Annals of Ireland’ hasbeen linked to the pseudohistorical element within the text. This describes affairsin the overkingdom of the Osraige, and it has been assigned (as mentioned

50Ibid., pp. 94/5, 102/3, 106–9, 110–13, 112/13, 114/15, 116/17, 120–3, 126/7, 128–31,136/7, 138–41, 144/5, 146/7.

51Ibid., pp. 114/15, 130/1.52Ibid., pp. 94/5, 102/3, 102–5, 106/7, 108/9, 110/11, 112/13, 114/15, 116/17, 126/7. Cf.

discussion ibid., pp. xxix–xxx.53Ibid., p. xxxiii.54See above, p. 113 and nn. 13–14.55Fragmentary Annals, ed. & transl. Radner, pp. xxix–xxx.56Ibid., pp. 112/13, 116/17.


above)57 to the eleventh century. It is not clear to me that the pseudohistoricalnarrative in the text was composed at the same time as ‘The Fragmentary Annalsof Ireland’ was compiled.58 Instead I propose a date for the compilation of ‘TheFragmentary Annals of Ireland’ after the composition of ‘The Osraige Chronicle’and before the latest likely date for ‘The Book of Giolla na Naomh MacAodhagáin’ (therefore ca 1030 ca 1400).

57See above, p. 111~12.58Radner, ‘Writing history’, p. 323.


STEMMATIC REPRESENTATIONOF THE DEDUCED TEXTUAL HISTORY

‘The Chronicle ofClonmacnoise’(911×954)

‘The Osraige Chronicle’(ca 1030)

minor other elements

the original compilationnow represented by

‘The Fragmentary Annals of Ireland’(ca 1030 × ca 1400)

‘The Book of Giolla na NaomhMac Aodhagáin(× ca 1400)

‘The Annals of The Four Masters’(1632–6)

copy by An DubhaltachMac Fhirbhisigh forJohn Lynch in 1643

Bruxelles, Bibliothèque royale,MS. 5301–5320 (4641)

(saec. xvii2)


IV

SCANDINAVIAN SETTLEMENT IN IRELAND

The historical importance of Viking-Age Waterford

THE recent Viking-Age discoveries at Woodstown, near Waterford, in the Irishprovince of Munster, have highlighted the need to assess the importance ofWaterford as a viking-settlement in the ninth and tenth centuries.1 Mainly drawingon written sources, I set out in this paper to discuss: (a) the site of Woodstownand the origins of Waterford; (b) Waterford’s relationship with other viking-settlements in Ireland and (c) links with neighbouring Irish polities; (d)Waterford’s economic significance; and, finally, (e) the external contacts of the port. I shall restrict my analysis to the years when Waterford was ruled byviking-kings before 1035.

In Ireland, the Viking-Age is conventionally dated from the first recordedviking-raids in 795 until the Angevin invasion of 1171/2. The enduringcontribution of these centuries is the foundation of major Irish ports – includingWaterford, Dublin, and Limerick – which brought Ireland into closer contactwith viking-colonies throughout Europe. The nature of vikings’ impact on Irishhistory is still hotly debated, and it is hoped that further research at Woodstownwill shed new light on this formative period.

The site at Woodstown, alias Woodhouse, is located on the banks of the River Suir, roughly three miles west of the centre of modern Waterford. Themodern Irish name of Woodstown, Baile na Coille, seems to be a direct translationof the English.2 The earliest names recorded for the site are Balleode andBaliowodam found in a charter of 1191 issued when John, Lord of Ireland, grantedit to the new Priory of St John which he founded in the town of Waterford.3 This early post-Conquest form suggests that the ‘wood’-element of the name

1 This paper was written for, and presented at, the Eighteenth Irish Conference of Medievalists,held on 26 June 2004, at St Kieran’s College, Kilkenny. I should like to thank Colmán Etchinghamand Kenneth Nicholls for helpful remarks offered after my lecture and Kristin Bornholdt forcomments on numismatic material from Woodstown and Dunmore Cave. I am also grateful toBridgitte Schaffer for checking a reference in a printed source not directly available to me. The mapsof south-west and south-east Ireland are reproduced from http://www.ireland-information.com/.

2 Pace P. Power, ‘Place names of the Decies’, Journal of The Waterford and South-east of IrelandArchæological Society 9 (1906) 12–39, 80–106, 154–70, 228–65, 10 (1907) 14–41, 88–134, 184–240, 263–310, 11 (1908) 1–34, 45–78, 135–68, 179–210, 12 (1909) 25–48, at 9 (1906) 86: ‘TheEnglish, or official, form is merely a literal translation of the Irish name’.

3 P. Power, ‘The priory church and hospital of St. John the Evangelist, Waterford’, Journal ofThe Waterford and South-east of Ireland Archæological Society 2 (1896) 81–97, at p. 83.

7

The Journal of Celtic Studies 4 (2004) ISSN 1781-1406

129


Map 1: South-west Ireland

VIKING-AGE WATERFORD 131

Map 2: South-east Ireland

Woodstown is not derived simply from the English word but was suggested insome way by an antecedent Irish usage.

To date, an area of Woodstown 400 metres by 60 metres has been investigated, and this has yielded a remarkable series of finds. Ship-nails, locks,and balance-weights (some decorated with Irish ecclesiastical metalwork), a heathen warrior-burial, and hacksilver all illustrate viking-activity. Finds which give dating-evidence for the site include a Kufic dirham (a silver coin from the Arab world)which can be dated to the ninth century. A fragment of a Hiberno-Scandinavianarm-ring of the type which circulated in Ireland in the late ninth and tenthcenturies has also been discovered. A sword found in the warrior’s burial may beof a type datable to the tenth century, but that has not yet been confirmed.Evidence of house-gullies and a defensive ditch require further analysis todetermine both the function and longevity of the site and its extent. A fullexcavation could radically alter our views on the early Viking-Age in Ireland.

ORIGINS

A crucial question to address is the relationship of Woodstown to the nearbysettlement at Waterford. I shall approach this topic in three parts, discussing firstthe origins of viking-settlement in Ireland, secondly the evidence for early viking-activity in the Waterford-region, and finally drawing on the historical references to viking-settlement at Waterford. One vital point is that we should not suppose thebeginnings of viking-activity in Ireland to be fully charted in the survivingchronicles.

The first recorded raids on Ireland took place in 795, more or lesscontemporaneously with the first documented attacks in other areas of western Europe.4 The beginnings of viking-activity in Ireland are often divided by scholarsinto phases, beginning with hit-and-run raids from the 790s to the 820s,proceeding to more intensive campaigns in the late 830s when the first annalisticreferences to viking-bases, sometimes called longphuirt (‘ship-ports’), are found.5

4 [The] A[nnals of] Clon[macnoise, being Annals of Ireland from the Earliest Times to A.D. 1408], ed. Denis Murphy (Dublin 1896),792[=795]. [Annala Rioghachta Eireann,] A[nnals of the Kingdom of Ireland, by The] F[our] M[asters, from the Earliest Times to the Year 1616], ed. & transl. John O’Donovan (2nd edn, 7 vols, Dublin 1856), I.790[=795]. [The] A[nnals of] I[nisfallen (MS. Rawlinson B 503)], ed. & transl. Seán Mac Airt (Dublin 1951), [795].2. [The] A[nnals of] U[lster (to A.D. 1131)], I, edd. & transl. Séan Mac Airt & G. Mac Niocaill (Dublin 1983), 794[=795].3. Un-bracketed dates are those found in the chronicle-text; editorial dates are given in square brackets. Where necessary I have added a corrected date inbrackets following an equals-sign.

5 For example, see Else Roesdahl, The Vikings (London 1991), pp. 221–9.


The first example on record is that of Inber Dée (Co. Wicklow) in 836.6

This ‘longphort-phase’ ends in 902 when Uí Ímair were expelled from Dublin.7

Nevertheless, recent excavations conducted by Linzi Simpson in Dublin may causea revision of this linear presentation of the origins of viking-settlement in Ireland.Two viking-burials excavated at Ship Street yielded radiocarbon-dates pointing to the late 780s and 790s.8 Burial evidence is often used to indicate settlement, though perhaps of a temporary nature.

A short-term viking-base may have been founded at Dublin in the time of theearliest recorded attacks, such as the raid on Rechru, arguably Lambay Island nearDublin, in 795, and the more certain attack on Holmpatrick near the Skerries in798.9 ‘The Annals of Ulster’ report that on this occasion the cattle-tribute ofneighbouring territories was collected by vikings.10 One might suppose that thegathering of this tribute could not have been done instantaneously and thereforeenvisage the existence of a temporary post while cattle were extorted, counted,selected, and driven onto longships. Cattle-tribute may not have been the mostpractical way of transferring wealth overseas, and it is possible that the animalsrepresented supplies for a camp in Ireland. Rather than sticking wholly to hit-and-run raids (drawing one’s ship to shore, sacking the local church, then rushinghome), some of the early raiding-bands may have chosen a more leisurelyapproach, anchoring their ships offshore for a short period, or even establishing acamp on land.

It may sound far-fetched or over-hasty to drag back the origins of viking-bases in Ireland by forty years on the evidence of a restricted number ofradiocarbon-dates, and further investigation is needed. Nevertheless, helpfulcomparison may be made with the situation in England. ‘The Anglo-SaxonChronicle’ makes no reference to viking-settlement in England in the early ninthcentury. However, a series of contemporary royal diplomas issued by kings of the

6 AFM 835[=836]; AU 835[=836].5; C[hronicum] S[cotorum. A Chronicle of Irish Affairs, fromthe Earliest Times to A.D. 1135; with a Supplement, containing the Events from 1141 to 1150], ed. &transl. William M. Hennessy (London 1866), [836]. E. Bhreathnach, ‘Saint Patrick, vikings andInber Dée – longphort in the early Irish literary tradition’, Wicklow Archaeology and History 1 (1998)36–40.

7 A second series of longphuirt-foundations is recorded in the 920s and 930s: AU 920[=921].7,922[=923].4, 923[=924].1, 925[=926].5, 925[=926].6, 929[=930].3, 935[=936].2.

8 L. Simpson, ‘Viking warrior burials in Dublin: is this the longphort?’, Medieval Dublin 6 (2004)11–62. The Ship Street burials were located on the west side of the dark pool of Dub-linn. Theradiocarbon-dates come from two male heathen graves, one accompanied by a shield-boss anddagger, the other with a finger-ring, bead, metal disc, and sword.

9 AClon 792[=795]; AFM 790[=795].6; AU 794[=795].3; C. Downham, ‘An imaginary vikingraid on Skye in 795?’, Scottish Gaelic Studies 20 (2000) 192–6 (see chapter 1, above).

10 AClon 795[=798]; AFM 793[=798]; AU 797[=798].2.


Mercians from 792 to 827 refers to viking-camps and campaigns in Kent.11

Chronicles cannot necessarily be relied on to record the first stages of viking-settlement.12

An exodus of vikings from Ireland to Britain in the 860s led to the destructionof some longphuirt by Irish kings.13 Most of the camps – as at Youghal, Clondalkin,and those in the territory of the Northern Uí Néill – are first recorded at themoment when they were demolished.14 We do not know when they were founded.For this reason the site at Woodstown will shed new light on the progression ofviking-settlement along the south coast of Ireland. It will also facilitate betterunderstanding of viking-colonisation across Europe and illustrate how Ireland fitsinto the bigger picture of these revolutionary events.

When we move from general to local evidence, we find that Irish chroniclesdisplay particular interest in viking-attacks on churches. Colmán Etchingham hasnoted a geographical bias towards the Shannon-basin and the East Midlands; so,although annal-entries pinpoint early viking-activity near Waterford, the coverageof this area may be less detailed.15 The earliest recorded viking-campaigns in Munstercan be dated to 812.16 In that year Cobthach son of Mael Dúin, overking ofEóganacht Locha Léin, and his men slaughtered a band of vikings. The early twelfth-century saga Cocad Gaedel re Gallaib gives a fuller account of events in this year andmentions a viking-fleet at Camas in the territory of Uí Fhothaid Thíre, presumablynear Waterford.17 This account could provide evidence for viking-activity along theRiver Suir. However, the reliability of this saga-account is uncertain: in particular,the author tended to telescope together events from a range of years.18

11 P. H. Sawyer, Anglo-Saxon Charters. An Annotated List and Bibliography (London 1968), nos134, 160, 168, 177, 186, 1264 (online at http://www.trin.cam.ac.uk/sdk13/chartwww/NewRegReg.html).

12 David N. Dumville, ‘Vikings in the British Isles: a question of sources’, in Scandinavians fromthe Vendel Period to the Tenth Century, ed. Judith Jesch (San Marino 2003), pp. 209–50, at pp. 214, 220–2; revised reprint in his Celtic Essays, 2001–2007 (2 vols, Aberdeen 2007), II.115–40.

13 Clare Elizabeth Downham, ‘Britain and Scandinavian Ireland: the Dynasty of Ívarr and Pan-Insular Politics to 1014’ (unpublished Ph.D. dissertation, University of Cambridge 2003), p. 18.

14 AFM 864[=866], 865[=867]; AU 865[=866].4, 866[=867].8.15 Colmán Etchingham, Viking Raids on Irish Church Settlements in the Ninth Century. A

Reconsideration of the Annals (Maynooth 1996), pp. 17–31.1 AClon 809[=812]; AFM 807[=812]; AU 811[=812].11; CS [812].1 Cogadh Gaedhel re Gallaibh. The War of the Gaedhil with the Gaill; or, The Invasions of Ireland

by Danes and Other Norsemen, ed. & transl. James Henthorn Todd (London 1867), pp. 4/5 (§4);Edmund Hogan, Onomasticon Goedelicum Locorum et Tribuum Hiberniae et Scotiae (Dublin 1910),pp. 154, 672. Following Hogan, I had thought that this might refer to Camus Bridge, near Clonmel,bar. Iffa and Offa, Co. Tipperary. However, Breandán Ó Ciobháin suggested at the Eighteenth IrishConference of Medievalists that the Camas mentioned in Cocad Gaedel re Gallaib may be a site onthe River Suir closer to Woodstown.

18 M. Ní Mhaonaigh, ‘Cogad Gáedhel re Gallaib and the annals: a comparison’, Ériu 48 (1996)101–26, at p. 105.


There is more secure evidence for viking-activity in the sea-lough known asWaterford Harbour in 822. In this year the church at Inis Doimle was ransackedby vikings.19 The site has been convincingly identified by Máirín Ní Dhonnchadhaas Great Island on the River Suir.20 Vikings returned in 825 and not onlyplundered Great Island but also campaigned against the Osraige.21 In 837 attackson the churches at Freshford (Co. Kilkenny) and Killinny (Co. Kilkenny) inOssory indicate further use of the River Barrow and River Nore by vikings oncampaign.22 These raids may highlight the significance of Waterford Harbour asthe key to the inland river-routes. Located at the border between mediaevalprovinces, Waterford Harbour provided riverine access to the heartlands ofMunster, Leinster, and Ossory. These raids of the 820s and 830s occurred duringone of the most intense phases of viking-activity in Ireland, but as yet withoutreference in the chronicles to a settlement at Waterford. These attacks may haveencouraged overkings of Munster and Leinster to ally against the aggressors, andin 848 these native rulers won a significant victory at Skeenagun nearCastledermot, Co. Kildare.23 This is one of a number of Irish victories in that year,news of which travelled to the carolingian court.24

Following the notices of those defeats there is a decline in records of viking-activity. Colmán Etchingham has linked this to a change in chronicling practice,noting a decline in other types of record at this time.25 Alternatively, DonnchadhÓ Corráin has associated it with the arrival of Dubgaill, ‘Dark Foreigners’. He hasargued that they established a monopoly of viking-led violence, thus reducing thenumber of reported attacks.26 Whatever the cause, it is precisely when this declinein record occurs that there is a radical increase in record of viking-attacksemanating from Waterford Harbour. This could be explained by the foundationof a viking-base in the area.

19 AClon 819[=822]; AFM 820[=822].20 M. Ní Dhonnchadha, ‘Inis Teimle, between Uí Chennselaig and the Déissi’, Peritia 16 (2002)

451–8.21 According to Cocad Gaedel re Gallaib Uí Cheinnselaig and the churches of St Mullins (Co.

Carlow) and Ennistogue (Co. Kilkenny) were also attacked: Cogadh, ed. & transl. Todd, pp. 6/7(§7).

22 AFM 836[=837]; Hogan, Onomasticon, pp. 192, 196.23 AI [848].2; AU 847[=848].5; CS [848].24 Annales de Saint-Bertin, edd. Félix Grat et al. (Paris 1964), s.a. 848; AFM 846[=848]; AU

847[=848].4, 847[=848].6, 847[=848].7; CS [848]. Cf. David N. Dumville, Three Men in a Boat.Scribe, Language, and Culture in the Church of Viking-Age Europe (Cambridge 1997), pp. 56–7;Francis John Byrne, Irish Kings and High-kings (2nd edn, Dublin 2001), p. 262.

25 Etchingham, Viking Raids, pp. 12–15.26 D. Ó Corráin, ‘The vikings in Scotland and Ireland in the ninth century’, Peritia 12 (1998)

296–339, at pp. 328–9.



Figure 1: Locations of viking-activity recorded in Irish chronicles, to show the significance of campaigns along the River Shannon

and the River Liffey

VIKING AGE WATERFORD 137

Figure 2: Locations of viking-activity recorded in Irish chronicles, to show the significance of campaigns along the River Nore

and the River Barrow.

The change can be seen in maps plotting viking-activity (Figures 1 and 2). In the840s reported viking-activity focused on the Shannon and the Liffey. In the 860sthere is a notable reörientation of attacks along the valleys of the Nore and Barrow.Some of this increase can be attributed to the commencement of records for theViking-Age in the so-called ‘Fragmentary Annals of Ireland’.27 This text is biasedtowards the Osraige (both in favour and in coverage), but, even without itsinformation, increased viking-interest in the region is notable.

This concentration of viking-raids may have been intended to take advantageboth of the political weakness of Munster following the death of its overkingÓlchobur in 851 and of the decline of the overkingship of Leinster because of theambitions of Uí Néill in the area.28 The campaigns could also be interpreted asrivalry between Dubgaill (‘Dark Foreigners’) and Finngaill (‘Fair Foreigners’), nowbased respectively at Dublin and Limerick, spilling over into this area.29 WaterfordHarbour represented a strategically significant middle-ground over which rivalviking-groups vied for control. These campaigns threw the polities of the Osraige(focused on the Nore Valley) and their northern neighbours the Loígis into thepolitical limelight. Their rulers, Cerball of the Osraige and Cennétig of the Loígis,were able skilfully to play off one band of vikings against another.30

A saga-account of these years, written in the eleventh century, was used as asource for ‘The Fragmentary Annals of Ireland’. It gives a colourful narration inwhich Cerball plays the hero. The viking Rodolb (Old Scandinavian Hróðólfr) isportrayed as the arch-enemy, and Cerball defeats him on three occasions,culminating in the destruction of Rodolb’s base beside the River Barrow.31

According to the saga, Cerball also defeated the vikings of Waterford in 860, andthis is the first reference to a viking-base there. The account provided in the ‘TheFragmentary Annals of Ireland’ is worthy of quotation.32

As amlaidh táinig Cearball immach asa grianán 7 rioghchainnel mhór re(a)imhe, 7 rá bhoísoilsi na caindle sin go fada ar gach leith. Ra ghabh úamhan mór na Lochlannaig; 7 ratheichsiot fona sleibhtibh faigsibh dhóibh 7 fona cailltibh. An lucht imorro ra thairis

27 Fragmentary Annals of Ireland, ed. & transl. Joan Newlon Radner (Dublin 1978), pp. 88–183(§§233–459).

28 Byrne, Irish Kings, pp. 162–3; Donncha Ó Corráin, Ireland before the Normans (Dublin 1972),p. 99.

29 For the early history of Limerick see: AU 844[=845].1, 921[=922].3, 927[=928].5; CS[845].1, [887].6. The Dubgaill seized Dublin in the 850s: AU 848[=849].6, 850[=851].3,851[=852].3, 851[=852].3; CS [849], [851], [852].

30 C. Downham, ‘The career of Cearbhall of Osraighe’, Ossory, Laois and Leinster 1 (2004) 1–18; C. Downham, ‘The good, the bad and the ugly: portrayals of vikings in “The FragmentaryAnnals of Ireland”’, The Medieval Chronicle 3 (2004) 28–40. (See above, chapters 5–6.)

31 Downham, ‘The career’, p. 11.32 Fragmentary Annals, ed. & transl. Radner, pp. 108/9 (§277).


ra h-eangn[a]m díobh ra marbadh uile. O thainig maidin a mmucha arnamharach, rachuaidh / / Cearbhall gona shoc[h]raidhe ’na cceann uile, 7 ní ra ghabh uatha ar marbadha ndeargáir, go ra cuirit a mmadhmuim, 7 go ro sgaoilit íad for gach leith. Ra immirCearbhall féine go crúaidh isin ammus sain, 7 tainig ris go mór a méd attibh an aidhcheremhe, 7 ra sgé go mór, 7 tug sonairte mór do-s[o]mh sain; 7 ra ghreiss go mór a muinntirgo dioc[h]ra forsna Lochlannac[h]aib.

‘This is how Cerball came out of his chamber: with a huge royal candle before him, andthe light of the candle shone far in every direction. Great terror seized the Norwegians,and they fled to the nearby mountains and to the woods. Those who stayed behind out ofvalour, moreover, were all killed. When daybreak came the next morning, Cerball attackedall of them with his troops and he did not give up after they had been slaughtered untilthey had been routed, and they had scattered in all directions. Cerball himself fought hardin this battle, and the amount he had drunk the night before hampered him greatly; andhe vomited much, and that gave him immense strength; and he urged his people loudlyand harshly against the Norwegians.’

In the same year, according to this saga, two viking-chieftains who were travellingfrom Limerick to Waterford were killed by the men of Munster.33 Caution isnecessary in interpreting these references to Waterford. They are dependent on anaccount written two centuries after the events which it described. It can be arguedthat the author(s) of the saga embedded in ‘The Fragmentary Annals’ drew theskeleton of the narrative from earlier chronicle-accounts and elaborated on,distorted, and dramatised it to flatter and entertain the ruling dynasty of theOsraige.34 It is credible that Scandinavian Waterford had been founded by 860,but ‘The Fragmentary Annals of Ireland’ cannot be relied on in their moreflamboyant details.

One frequently repeated story of Waterford’s origins is that it was establishedby a viking-chieftain named Sitric in 853. However, this assertion seems highlydubious. According to the eighteenth-century writer Charles Smith in his historyof Waterford, the port was created by a viking-leader called Sitricus in the year853, at the same time as Limerick was founded by Ivarus and Dublin byAmlavus.35 This account is based on a story by Gerald of Wales in the late twelfthcentury. According to Gerald, three brothers – Amlavus (Old Scandinavian Óláfr), Sitricus (Old Scandinavian Sigtryggr), and Ivarus (Old Scandinavian Ívarr) – werepermitted to settle in Ireland for the purposes of commerce, and they founded

33 Ibid. (§278). The chieftains are named as Hona the sorcerer and Tomrar Torra.34 Ibid., pp. xxii–xxxiv.35 Charles Smith, The Ancient and Present State of the County and City of Waterford (2nd edn,

Dublin 1774), p. 96.


the three towns of Dublin, Waterford, and Limerick.36 Gerald added the barb thatthe Irish allowed this because they were too lazy to trade. As well as being a taleto discredit the Irish, this reads like an origin-legend for the dynasty of Ímar, whichimposed its authority over Dublin and Limerick after these viking-bases werefounded.37 Like so much of what Gerald wrote, the story has been taken to heartand elaborated by later commentators, who have assigned a date to the event, butI see no reason to accept it as accurate. Another foundation-myth for Waterfordis found in the thirteenth-century Ystoria Gruffudd ap Cynan. In this account theNorwegian king Harald Finehair founded Dublin and gave Waterford to hisbrother whose descendants ruled it thereafter.38 Because of Waterford’s continuingsignificance, it is not surprising that mediaeval legends circulated about the originsof the port.

Vikings from Waterford could have participated in some of the campaignswhich took place on the Rivers Nore and Barrow from the 860s to the 880s,including the slaying of Cuilén son of Cerball of the Osraige in 887.39 The nextunambiguous reference to Waterford is in 892 when Riagán mac Dúnlainge,overking of the Osraige, defeated vikings from Waterford, Wexford, and St Mullins(Co. Wexford).40 Ten years later, the dynasty ruling Dublin was expelled fromIreland. When vikings associated with this dynasty returned to Ireland in 914,Waterford was their first recorded port of call.41

The earliest references to a viking-settlement at Waterford can be identified forthe years 860, 892, and 914, but what names were given to this settlement? Andwhat was the relationship of Woodstown to the site underlying the present city ofWaterford? The English name Waterford is derived from an Old-Scandinavianname meaning ‘windy fjord’ or ‘ram-fjord’: it is not found in Irish written sourcesof the Viking-Age.42 In Irish chronicles the name given to the port in 860, 892,and 914 is the familiar Irish name, Port Láirge. The eminent nineteenth-century

36 J.J. O’Meara, ‘Giraldus Cambrensis in Topographia Hibernie: Text of the First Recension’,Proceedings of The Royal Irish Academy 52 C (1948–50) 113–78, at p. 175 (§117); Gerald of Wales,The History and Topography of Ireland, transl. John J. O’Meara (2nd edn, Portlaoise/Harmondsworth1982), p. 122.

37 See n. 29, above.38 A Mediaeval Prince of Wales. The Life of Gruffudd ap Cynan, ed. & transl. D. Simon Evans

(Felinfach 1990), pp. 25, 56 (§5). For the original twelfth-century Latin text, and see now Vita GriffiniFilii Conani. The Medieval Latin Life of Gruffudd ap Cynan, ed. & transl. Paul Russell (Cardiff 2005),

39 AFM 884[=887].40 AFM 888[=892].41 AFM 910[=914]; AU 913[=914].5; CS [913][=914].42 M. Oftedal, ‘Scandinavian place names in Ireland’, in Proceedings of the Seventh Viking Congress,

Dublin, 15–21, August 1973, edd. Bo Almqvist & D. Greene (Dublin 1976), pp. 125–33, at p. 133; D. Mac Giolla Easpaig, ‘L’influence scandinave sur la toponymie irlandaise’, in L’Héritagemaritime des vikings en Europe de l’ouest, ed. Élisabeth Ridel (Caen 2002), pp. 441–82, at p. 47


scholar John O’Donovan thought that the settlement was named after a viking-leader Laraic mentioned uniquely in ‘The Annals of The Four Masters’ for theyear 953.43 However, if Laraic was a person rather than a place, he lived much toolate to be the founder of Waterford, and his name-form seems to have beencorrupted.44 It therefore seems more likely that the name Port Láirge is entirelyIrish. Lárac is an Irish word meaning ‘limb’ or ‘thigh’, and it is found in otherGaelic place-names, for example Áth dá Loracc near Kells (Co. Meath) recordedin 1176 and Áth do Laaraic near Boyle (Co. Roscommon) first mentioned in1197.45 In the Middle Ages this was the accepted origin of the name ‘Port of athigh’, and fantastic tales were invented to explain it. In Táin bó Cuailnge, whenthe Brown Bull slaughters his rival, the thigh-bone of Findbennach is said to havebeen cast to Port Láirge.46 In the eleventh-century Dinnshenchas Érenn it is saidthat a young prince called Rot died at sea after being lured by nubile sirens, whotore him to pieces, and his thigh-bone was washed ashore at Port Láirge.47

Curiously, Rot was the name given in Cocad Gaedel re Gallaib to a viking ofWaterford who was defeated by Munstermen about the year 916.48

Beginning in 914, and exclusively from 915 to 918, an alternative name Lochdá Chaech (‘The lake of the two blind people’), which is otherwise employed forWaterford Harbour, is used in the chronicles for a viking-settlement at Waterford(see Appendix 1). The name Loch dá Chaech ceases to be used for a viking-settlement at Waterford after 918, corresponding with the time when ‘theForeigners of Loch dá Chaech’ departed from Ireland under the leadership ofRagnall ua Ímair. (The name nevertheless continued to apply to the sea-loughWaterford Harbour.) From 918 to 925 the vikings at Waterford are not mentioned in the chronicles, and then in 926 the earlier name, Port Láirge, resumesin the chronicles and is employed thenceforth. This switch could represent a change

43 AFM 951[=953] and note by O’Donovan, Annala, II.698.44 I should like to thank Martin Syrett for his comments on this name. While the second element

of the name could be Old-Scandinavian -ríkr, it does not conform to any known Old-Scandinavianname. Alternatively, as Breandán Ó Ciobháin suggested after my lecture, it could represent thesecond element of Port Láirge (Laraic) used as a shorthand for the place. If so, O’Donovan’stranslation should be amended to state that St Mullins was raided ‘from [Port] Láirge’ rather than‘by Laraic’.

45 Dictionary of the Irish Language based mainly on Old and Middle Irish Materials. CompactEdition, gen. ed. E.G. Quin (Dublin 1983), p. 422, L, col. 54, line 80 – col. 55, line 51; Mac GiollaEaspaig, ‘L’influence’, p. 476; T.F. O’Rahilly, ‘Iarann, Lárag, etc.’, Ériu 13 (1940–2) 119–27, at p.123 and n. 5; W.J. Watson, The History of the Celtic Place-names of Scotland (Edinburgh 1926), pp.483–5. In Scottish place-names this element means ‘a pass’.

46 Táin bó Cúalnge from The Book of Leinster, ed. & transl. Cecile O’Rahilly (Dublin 1967), pp.136, 272.

47 The Metrical Dindshenchas, ed. & transl. Edward J. Gwynn (5 vols, Dublin 1903–35),III.190–3.

48 Cogadh, ed. & transl. Todd, pp. 26/7 (§26).


in chronicling practice, or it could be that a new viking-camp was founded in914, which was identified with Loch dá Chaech.49

Is the viking-settlement at Loch dá Chaech to be thought distinct from bothWaterford and Woodstown? The chronology of finds revealed so far at Woodstownmakes it unlikely that it can be identified with the viking-base called Loch dáChaech. Alternatively, is it possible that Loch dá Chaech is the name first givento what is now Waterford? Woodstown lies over three miles upstream from thecentre of Waterford. The vikings who came to Ireland in 914 may have placed astronghold nearer the sea-lough.50 Perhaps this fort was initially identified by thename Loch dá Chaech but by 927 received the more specific name of Port Láirgeborrowed from the neighbouring site which it superseded. According to thistentative argument, Woodstown was put out of use in the early tenth century.From an archaeological perspective this would mean that the site has beenrelatively undisturbed for over a thousand years.

It is of course possible that the two viking-sites of Woodstown and Waterfordcoëxisted for a longer period before Woodstown fell out of use and Waterford tookover as the main viking-port on the River Suir. The political situation in the earlytenth century nevertheless provides a possible context for that development. Theearliest archaeological levels so far identified at Waterford are from the earlyeleventh century, and so the secrets of any earlier history are still tantalisinglyburied underground.51

It is clear that, at some point in the Viking-Age, Woodstown declined whileWaterford flourished. Various possible reasons for this come to mind. One is thatthe present site of Waterford may have been more easily defended, so thatWaterford became more successful for military reasons. Waterford was also closerto the estuaries of the Barrow and Nore: it was therefore better sited to mediatetraffic passing through Waterford Harbour than was Woodstown. Another causeof Waterford’s success could be linked to river-morphology: Waterford may haveprovided a better location for a quay than Woodstown as traffic to the settlementincreased, bringing many deep-berthed cargo-ships as well as the more

49 That Loch dá Chaech is not Port Láirge seems clear when Loch dá Chaech is named (in arecord of the death of the poet Flann mac Lonáin) as being in the territory of Déisi Muman in 896,and thus distinct from the viking-settlement of Port Láirge: AFM 891[=896]; AI [896].3; AU895[=896].10; CS [896]. Cf. AFM 918[=920] for a doublet. There are works attributed to Flannmac Lonáin but written after his death: M. Dobbs, ‘A poem ascribed to Flann mac Lonáin’ Ériu 17(1955) 18–34; C. O Lochlainn, ‘Poets on the battle of Clontarf ’, Éigse 3 (1942/3) 208–18 and 4(1944/5) 33–47.

50 AClon 910[=914]; AFM 910[=914], 912[=914]; AU 913[=914].5; CS [913][=914].51 J. Bradley & A. Halpin, ‘The topographical development of Scandinavian and Anglo-Norman

Waterford City’, in Waterford – History and Society, edd. William Nolan & T.P. Power (Dublin 1992),pp. 105–30.


143

manoeuvrable longships which could be beached on the sands at Woodstown.This is speculation, but the circumstances and timing of the abandonment of

Woodstown constitute a crucial issue which needs to be addressed and which couldfurther highlight the importance of the site.

RELATIONS WITH OTHER VIKING-SETTLEMENTS IN IRELAND

Another question to be raised is how Waterford related to other viking-settlements in Ireland. As the distribution of Scandinavian place-namesdemonstrates, trading posts and way-stations were used by vikings around theIrish coastline.52 Waterford is comparatively well documented in written sources.It seems to have been one of the most prominent viking-settlements in Ireland,on a par with Limerick and Dublin.

There was competition, for control of the rivers leading into WaterfordHarbour, between rival viking-groups during the 860s. The Dubgaill led by Ímarand his associates won the upper hand, and this group may have controlled theport until Ímar’s family was expelled from Ireland in 902. When they returned toIreland in 914, Waterford provided their headquarters until the re-conquest ofDublin in 917.53

Coöperation between Waterford and Dublin continued in the 920s and 930swhen they made common cause against vikings based at Limerick. The prolongedwarfare, between the vikings of Limerick and those based in the east, encouragedthe establishment of viking-bases across Ireland. Each side sought to secure itseconomic and political interest in different regions, but sometimes deliberatelywithin the sphere of influence of its rivals. We can see in particular the rivalrybetween Limerick and Waterford over economic interests in Ossory and theGolden Vale of Counties Limerick and Tipperary. In 927 a battle was foughtbetween the two sides at Kilmallock (Co. Limerick) where the vikings of Waterfordwere defeated.54 Four years later Gofraid (Old Scandinavian Guðrøðr), king ofDublin, expelled the troops of Limerick from Ossory.55 Then in 937 the army ofLimerick was decisively beaten by the Dubliners who captured its king, Amlaíb(Old Scandinavian Óláfr), and broke up his fleet.56

52 Mac Giolla Easpaig, ‘L’influence scandinave’, pp. 448–9.53 AClon 917; AU 916[=917].4; CS [916][=917].54 AI [927].2.55 AClon 926[=931]; AFM 929[=931].56 AFM 935[=937].

VIKING-AGE WATERFORD

Waterford remained on friendly terms with Dublin for some years after thisvictory. For example, in 939 the Waterford-vikings sacked the church of Killeigh(Co. Offaly) and captured its abbot, Coibdenach.57 He was evidently handed overor sold on to the vikings of Dublin, for he drowned the following year by DalkeyIsland while trying to flee his jailers.58 Nevertheless, while coöperation betweenDublin and Waterford is well evidenced in the 930s, relations soured in mid-century. This may have been provoked by the decline of Dublin’s power overseas.The viking-kingdom of Northumbria, which had been ruled by the dynasty ofÍmar intermittently from the later ninth century, was lost in the 950s.59 This mayhave heightened competition between vikings in Ireland to control diminishingresources.

In 953 Amlaíb Cuarán, king of Dublin, attacked Great Island in WaterfordHarbour, in alliance with Tuathal, overking of Leinster.60 This may be interpretedas a hostile intrusion into Waterford’s sphere of influence. The island was sackednine years later by members of the dynasty of Ímar who were based in theHebrides.61 There is evidence for an independent line of kings at Waterford,ultimately related to the kings of Dublin but acting in hostility to them duringthe 960s. As Dublin was drawn more closely into the political affairs of Leinster,so Waterford became closely involved with the overkings of Munster. Thus in 969we find Ímar, king of Waterford, allied with the Dál Cais overking of Munster,Mathgamain mac Cennétig, to defend Ossory against Leinster.62 In the same year,Sitric son of Amlaíb of Dublin and Murchad mac Finn, overking of Leinster, alliedto attack the church of Kells (Co. Meath).63

In the years 967 to 977 Dál Cais first weakened Waterford and then conqueredit. Brian Bóruma mac Cennétig then promoted Waterford as a counterweight toDublin, as Dublin resisted his efforts to gain recognition as overking of Ireland.Dál Cais negotiated a pact with vikings from Waterford and the kingdom of Manand the Isles in 984 to oppose Leinster and Dublin.64 During the 990s, Ímar of Waterford and his sons fought for control of Dublin. They ruled Dublin

57 AFM 937[=939].58 AFM 938[=940]; CS [939][=940].59 C. Downham, ‘The chronology of the last Scandinavian kings of York’, Northern History 40

(2003) 25–51.60 AFM 951[=953].61 AFM 960[=962]; C. Etchingham, ‘North Wales, Ireland and the Isles: the Insular viking zone’,

Peritia 15 (2001) 145–87, at p. 170.62 AFM 967[=969].63 AFM 967[=969].64 AI [984].2.



briefly in 993 and 995, vying for control with Sitric son of Amlaíb.65 This feudlasted for a decade until Ímar’s death in A.D. 1000.66 It is not clear whether vikingsof Waterford supported Brian in his great battle against Dublin and Leinster atClontarf (Co. Dublin) in 1014. Certainly, Ímar’s descendants at Waterfordcontinued to oppose Dublin after that conflict. These rivalries came to a head in1035 when Ragnall, king of Waterford, was treacherously killed by Sitric, king ofDublin.67 This proved to be a major setback for the viking-rulers of Waterford,and from that time the port became increasingly vulnerable to the conflictingambitions of Uí Briain in Munster and Uí Cheinnselaig in Leinster.

Prior to 1035 Waterford’s relations with other viking-settlements show that itwas a major player in Irish politics. At different times it rivalled Limerick andDublin as one of the most powerful towns in Ireland. I shall briefly considerWaterford’s relations with Irish polities before discussing its economic significanceand its links outside Ireland.

LINKS WITH IRISH POLITIES

Mediaeval chronicles tend to chart discord rather than coöperation, a featurewhich they share with modern news-reports. Our image of vikings’ relations withthe Irish is seen through the record of numerous violent acts from which lessaggressive interaction such as alliance, intermarriage, or trade can sometimes beinferred. The vikings of Waterford are most frequently mentioned as raiders ofchurches. The chronicles identify such prestigious targets for the armies ofWaterford as Kildare (Co. Kildare) and Clonenagh (Co. Laois) in Leinster, andCork (Co. Cork) and Lismore (Co. Waterford) in Munster.68 Initially, economicmotives may have been behind these attacks. Significant amounts of ecclesiasticalmetalwork made their way to Scandinavia, as attested by heathen graves there;ecclesiastical-metalwork fragments also adorn balance-weights found atWoodstown.69 Irish chronicles mention the plundering of shrines and the takingof captives for ransom or for the slave-market. Slaves seem to have been exported from Ireland at an early date, as graphically illustrated by the ninth-century Life of

65 AFM 992[=993], 994[=995]; AI [993].6.66 AClon 993[=1000]; AFM 999[=1000]; AU 999[=1000].3; CS [998][=1000].67 AFM 1031[=1035], 1035; AU 1035.5.68 AFM 913[=915], 937[=939]; AU 981[=982].4; CS [914][=915], [980].69 E. Wamers, ‘Insular finds in Viking Age Scandinavia and the state formation of Norway’, in

Ireland and Scandinavia in the Early Viking Age, edd. Howard B. Clarke et al. (Dublin 1998), pp. 37–72; E. Wamers, ‘Some ecclesiastical and secular Insular metalwork found in Norwegianviking graves’, Peritia 2 (1983) 277–306.

St Findán of Rheinau, a Leinsterman who was captured by vikings.70 Some of the raids on churches could also have served as strikes against rival trading-centres, as vikings sought to develop their bases as permanent trading-sites.

These raids on churches may have been inspired by political as much aseconomic considerations. Kildare was raided by Waterford-vikings in 926 and byan army from Dublin in the same year.71 The Waterford-vikings returned in 928,and many captives and treasures were removed.72 Then in 929 vikings travelledfrom Dublin to sack Kildare on the festival of St Brigit (1 February).73 Whilewealth was removed on each occasion, the frequency of the attacks would hardlyenable the church of St Brigit at Kildare to replace stolen goods between each raid. The principal motive for these assaults appears to have been the closeassociation between Kildare and Uí Dúnchada. The head of this dynasty, Faelánmac Muiredaig, was overking of Leinster from 917 to 942 and outlived a briefperiod of imprisonment by the Dublin-vikings in 925.74 These attacks on Kildarecan be interpreted as symbolic political acts, demonstrating the failure of itspatrons to provide protection. The church was also a major asset. Raiders not only took rich items from the church, but they brought adverse publicity. There mayhave been a decline in profits from visiting pilgrims and traders who subsequentlyfeared to go to Kildare. While the attacks on Kildare are the most noticeable ofthose led from Waterford, others also dovetail with political rivalries, thus mixing war and profit.

Irish chronicles also demonstrate alliances between the rulers of Waterford andIrish kings. In 916 unspecified vikings aided Diarmait mac Cerbaill, overking ofthe Osraige, against Uí Cheinnselaig; it is likely that they came from Waterford.75

In the late 930s Waterford also made common cause with Cellachán, overking ofMunster, against Uí Néill.76 A pact was subsequently made between the kings ofWaterford and Dál Cais. The alliances seem to have spawned marriages between

70 R.T. Christiansen, ‘The people of the North’, Lochlann 2 (1961) 137–64; M. Ní Mhaonaigh,‘Friend and foe: vikings in ninth- and tenth-century Irish literature’, in Ireland and Scandinavia, edd.Clarke et al., pp. 381–402, at 392–4.

71 AFM 924[=926]; CS [925][=926].72 AFM 926[=928]; CS [927][=928].73 AFM 927[=929].74 AClon 920[=925]; AFM 923[=925]; CS [924][=925]. 75 AFM 914[=916].76 AFM 937[=939]. The son of Acond (Old Scandinavian Hákonn) – macc Acuind (O’Donovan,

Annala, II.638/9, misread this as Macca cuind) – who led the Waterford-contingent on this occasion mayhave been a jarl rather than a king of Waterford. According to Cocad Gaedel re Gallaib an Acond (perhapshis father) arrived at Waterford with Cossanara in 914, followed by Ragnall ua Ímair and Ottir:Cogadh, ed. & transl. Todd, pp. 26–31 (§§26, 28).



royal dynasties. A son of Ímar, king of Waterford, in the late tenth century borethe name Gillaphátraic (†983).77 This name was common in the royal dynasty ofthe Osraige. The name is indicative of Irish affinities as well as christian identity.An alliance between Waterford and the Osraige is evidenced for 998, and it maybe that Ímar enjoyed a long-term alliance with the Osraige which was strengthened by marriage.78 We have two pieces of evidence that in the eleventhcentury Uí Briain intermarried with descendants of Ragnall. In each case thismight be one of three Ragnalls: Ragnall son of Ímar of Waterford; Ragnallgrandson of Ímar of Waterford; or Ragnall, king of Man and the Isles, who diedin Munster in 1005.79 A son of Brian Bóruma married a daughter of Ragnall, anda great-grandson – Tadc mac Toirrdelbaig – married Mór daughter of Echmarcach son of Ragnall. These marriages bound these families closely together in alliance against Dublin.80

As regards the fate of peoples and churches within the immediate vicinity ofWaterford, it is notable that the recorded ecclesiastical sites within a day’s marchof this viking-port survived during the Viking-Age. These include Fiddown (Co.Kilkenny) and Great Island (Co. Wexford).81 Such survival reflects the tenacity ofecclesiastics at these sites but may also say something of the circumstances whichprevailed in the immediate vicinity of Waterford after the initial viking-onslaughts,once the viking-settlement became established. Studies of urban hinterlands inthe last thirty years have tended to stress the necessity for interaction and stablerelations between towns and surrounding areas to ensure regular supplies ofagricultural produce and to encourage trade.82 The viking-towns have come to beseen less as parasitic entrepôts and more as part of a supply-network dependent

77 AFM 982[=983]; AU 982[=983].2; CS [981][=983].78 AFM 997[=998].79 AI [1018].5; AU 994[=995].7, 1004[=1005].1; AFM 1031[=1035].80 AI [1032].6; M.E. Dobbs, ‘The Ban-shenchus’, Revue celtique 48 (1931) 163–234, at p. 196. Cf. S.

Duffy, ‘Irishmen and Islesmen in the kingdoms of Dublin and Man, 1052–1171’, Ériu 43 (1992) 93–133.

81 AFM 828[=830], 873[=875], 948[=950], 980[=981]; Ní Dhonnchadha, ‘Inis Teimle’, p. 458. It is possible that these sites were temporarily abandoned in the wake of viking-attacks and that they soon came back into use.

82 J. Bradley, ‘The interpretation of Scandinavian settlement in Ireland’, in Settlement and Society in Medieval Ireland. Studies presented to F.X. Martin, O.S.A., ed. John Bradley (Kilkenny 1988), pp. 49–78; P. Holm, ‘Viking Dublin and the city-state concept: parameters and significance of the Hiberno-Norsesettlement’, in A Comparative Study of Thirty City-state Cultures, ed. Mogens Herman Hansen(København 2000), pp. 251–62; M. Valante, ‘Dublin’s economic relations with hinterland andperiphery in the later Viking Age’, Medieval Dublin 1 (1999) 69–83; P.F. Wallace, ‘The economy and commerce of Viking Age Dublin’, in Untersuchungen zu Handel und Verkehr der vor- und frühgeschichtlichen Zeit in Mittel- und Nordeuropa, IV,Der Handel der Karolinger- und Wikingerzeit, edd. Klaus Düwel et al. (Göttingen 1987), pp. 200–45.

on production-surpluses inland and a market for imported raw materials andexotic goods. According to this model, as Waterford developed an urban characterthe incentive for positive contacts in the immediate locality outstripped thenegative ones. Raids brought upheaval and disrupted the transport of goods aswell as forcing the movement of people as refugees, thus inhibiting agriculture.While this raiding may in the short term have boosted the supply of slaves andbooty, in the longer term it was inimical to trade and security and therefore lessdesirable for the people of Waterford. The vikings may have reached someaccommodation with churches and polities in the immediate vicinity of Waterfordafter an initial phase of bloodshed.

Waterford may have controlled a hinterland prior to 1035, although its extentis very unclear. According to John Bradley, this may have comprised the barony of Gaultier and parts of the barony of Middlethird.83 Nevertheless, Kenneth Nichollshas pointed out that the name Gaultier, originating as Gall-tír or ‘land of the Foreigners’, may have been coined in response to English rather thanScandinavian immigration.84 Therefore it should not be assumed that this area was under the control of the vikings of Waterford. The suggestion has been made thatWaterford could have dominated a coastal area stretching to the Blackwater-estuary;85 however, this is based on argument from later circumstances. Henry IIseized territory stretching from Waterford to Dungarvan (Co. Waterford) in 1175,according to the Treaty of Windsor; however, this may not have been a territorialunit in the tenth and eleventh centuries. There were people of Hiberno-Scandinavian identity living in Dungarvan in the thirteenth century, but this neednot indicate that Dungarvan was part of the Viking-Age hinterland of Waterford.Rather, one might envisage that small communities of Hiberno-Scandinavianmerchants and craftsmen became established in various Irish trading settlementsby the thirteenth century, such as that convincingly identified at Killaloe by JohnBradley.86 Thus it is uncertain how big Waterford’s hinterland was before 1035.Relations between vikings and Irish in the Waterford-area are also hard to define,although it is likely that each culture borrowed elements from the other. Theterritory ruled from Waterford may have been fairly compact, although ineconomic terms the influence of Waterford was felt across a wider area. Thedistribution of Viking-Age silver-hoards may be used as evidence of Waterford’seconomic power during these years.

83 Bradley, ‘The interpretation’, pp. 64–5.84 I should like to thank Kenneth Nicholls for making this comment after my lecture.85 Bradley, ‘The interpretation’, pp. 64–5.86 J. Bradley, ‘Killaloe: a pre-Norman borough?’, Peritia 8 (1994) 170–9.



ECONOMIC SIGNIFICANCE

This leads me to the penultimate topic of this paper, the economic significance of Waterford. The distribution of Viking-Age silver-hoards near Waterford impliesthat links existed between the port and the valleys of the Nore and the Suir(Appendix 2). Nevertheless, clear boundaries cannot be drawn between theeconomic sphere of one viking-port and those of its competitors. The rivalry which developed between Limerick and Waterford in the 920s and 930s may haveinvolved competition for trade in the fertile valleys lying between these ports.

The silver-hoards found near Waterford represent a significant amount ofwealth, accumulated as the profits of trade, plunder, or gift-exchange. In theircomposition and chronological distribution they fit within hoarding patternsnoted across south-east Ireland, although the hoard deposited at Dunmore Cave(Co. Kilkenny) about 970 is unusual by virtue of its inclusion of elaborateornaments which have been identified as buttons.87 It is evident that manycontents of the silver-hoards passed though Irish hands, even though theyoriginated in a Scandinavianised cultural milieu. The hoards at Rathmooley (Co.Tipperary) and Kilmacomma (Co. Waterford) were found in ringforts, and that at Knockmaon (Co. Waterford) was discovered at a natural fortress.88 The finds showthat exchange happened across cultural boundaries and thus hint at the long-term impact which vikings had on the Irish economy.

The hoards show Ireland’s participation in trading networks across Europe.That at Knockmaon includes coins from Dublin, Anglo-Saxon England, andFrance, as well as arm-ring fragments of a type produced in North Britain.89

Other finds indicate conflict or trade with vikings, notably the spearheads ofviking-type found at Borris-in-Ossory (Co. Laois) and Clashnamuck (Co. Laois),while a harness-mount similar to one found in a viking-grave on Colonsay in theHebrides was found at Freestone Hill in County Kilkenny.90 Written sources alsoshed light on trading links, referring to exports from Ireland including slaves,

87 M. Blackburn & H. Heathen, ‘A revised checklist of coin hoards from the British Isles’(http://www.fitzmuseum.cam.ac.uk/coins/hoards/index.html), no. 151b.

88 D.H. Scott-O’Connell, ‘Viking period silver ornaments from Rathmooley, County Tipperary’, Journal of The Cork Historical and Archaeological Society 43 (1938) 125–6; J. Sheehan, ‘The Rathmooley hoard and other Viking Age silver from Co. Tipperary’, Tipperary Historical Journal (1992) 210–16; J. Sheehan, ‘Viking Age hoards from Munster: a regional tradition?’, in Early Medieval Munster.Archaeology, History and Society, edd. Michael A. Monk & J. Sheehan (Cork 1998), pp. 147–63.

89 J.R.B. Jennings, ‘On some ancient coins found in west Waterford’, Journal of The Waterford and South-east of Ireland Archæological Society 15 (1912) 163–7; Sheehan, ‘Viking Age hoards’, p. 154.

90 E.P. Kelly & J. Maas, ‘The vikings and the kingdom of Laois’, in Laois – History and Society, edd. Pádraig G. Lane & W. Nolan (Dublin 1999), pp. 123–59, at 137, 147–8.

hides, woollens, and fine metalwork, while imports included pottery, wine, andhoney.91 Thus Waterford brought parts of Ireland into closer contact with aScandinavian trading-network which stretched from Iceland to the Mediterraneanand from Ireland to the Near East.

EXTERNAL CONTACTS

The foreign contacts of Waterford during the Viking-Age provide the final topicof this paper. Trade must have been a significant feature of those contacts, andone can imagine various ships from different lands anchored at Waterford whiletheir wares were sold in the bustling port. Political links abroad are also describedin chronicles. The dynasty of Ímar which ruled Waterford also ruled, at varioustimes, Northumbria, East Anglia, the Hebrides and Man, Anglesey, Dublin, andLimerick.92 When the dynasty of Ímar was temporarily expelled from Ireland in902, its external involvements were extended and reïnforced.93 A band of vikingsled by one Ingimundr travelled first to Anglesey and then settled near Chester in902.94 Others sought help at York. The massive hoard deposited at Cuerdale(Lancashire) around 905 combines Irish hacksilver, freshly minted coins fromNorthumbria, and Continental and Kufic coins.95 This hoard may result fromcoöperation between vikings from York and Ireland, and it has been interpretedas the pay-chest for an army intending to reconquer Dublin.96 Other members ofthe dynasty of Ímar fought in Alba in 904.97

91 Wallace, ‘The economy’, pp. 215–25.92 Alfred P. Smyth, Scandinavian Kings in the British Isles, 850–880 (Oxford 1977); Alfred P. Smyth,

Scandinavian York and Dublin. The History and Archaeology of Two related Viking Kingdoms (2 vols, Dublin 1975/9); Etchingham, ‘North Wales’; Downham, ‘Britain’.

93 AFM 897[=902]; AU 901[=902].2; CS [902]; Downham, ‘Britain’, pp. 77–81.94 Fragmentary Annals, ed. & transl. Radner, pp. 166–9 (§429); J. Jesch, ‘Scandinavian Wirral’, in

Wirral and its Viking Heritage, edd. Paul Cavill et al. (Nottingham 2000), pp. 1–10.95 J. Graham-Campbell, ‘Some archaeological reflections on the Cuerdale hoard’, in Coinage in

Ninth-century Northumbria. The Tenth Oxford Symposium on Coinage and Monetary History, ed. D.M. Metcalf (Oxford 1987), pp. 329–54.

96 Ibid., pp. 343–4.97 AU 903[=904].4; CS [904].



Place-names indicate that a wider-ranging sequence of population-movements took place in these years. Individuals of mixed Gaelic-Scandinavianidentity settled a wide arc of territory including the Faeroe Islands, Cumbria,Lancashire, and the Cotentin-peninsula in what was then Brittany but is nowNormandy.98

These upheavals were influenced by political circumstances following theexpulsion of viking-leaders from Ireland in 902. In 914 some of the vikings whoreturned to Waterford travelled from Brittany and came via the Severn-estuaryand South Wales.99 After they secured control of Waterford and Dublin, vikingsfrom Waterford under the leadership of Ragnall ua Ímair crossed Britain andfought Constantine, king of Alba, and the English of Northumbria at Corbridgein 918. An understanding of events at Corbridge has been rather muddled bydependence on the eleventh-century Historia de Sancto Cuthberto which states thattwo battles were fought; but the recent editor of that text, Ted Johnson-South,has pointed out that ‘on the strength of the single word iterum the majority ofscholars have supposed that there must have been two battles of Corbridge’.100

98 G. Fellows-Jensen, ‘Common Gaelic áirge, Old Scandinavian ærgi or erg?’, Nomina 4 (1980) 67–84; G. Fellows-Jensen, ‘Scandinavian settlement in the Isle of Man and north-west England: the place-name evidence’, in The Viking Age in the Isle of Man. Select Papers from the Ninth Viking Congress, edd.Christine Fell et al. (London 1983), pp. 37–52; G. Fellows-Jensen, ‘Scandinavian place-names of theIrish Sea Province’, in Viking Treasure from the North West. The Cuerdale Hoard in its Context, ed. James Graham-Campbell (Liverpool 1992), pp. 31–42; A. Grant, ‘A new approach to the inversioncompounds of north-west England’, Nomina 25 (2002) 65–90; Lucien Musset, Nordica et Normannica. Recueil d’études sur la Scandinavie ancienne et médiévale, les expéditions des vikings et la fondation de la Normandie(Paris 1997), pp. 279–96. The theory that names with the element -ærgi / -erg represent settlementfrom the Hebrides alone, and not Ireland, has been challenged by Breandán Ó Ciobháin (as Iunderstand from him in correspondence).

99 The Anglo-Saxon Chronicle. A Collaborative Edition, III, MS A, ed. Janet M. Bately (Cambridge1986), p. 65; N.S. Price, ‘The vikings in Brittany’, Saga-book 22 (1986–9) 323–440, at p. 360.

100 T. Johnson-South, ‘The “Historia de Sancto Cuthberto”: a New Edition and Translation, with Discussions of the Surviving Manuscripts, the Text, and Northumbrian Estate Structure’(unpublished Ph.D. dissertation, Cornell University 1990), p. 159; cf. Historia de Sancto Cuthberto. A History of Saint Cuthbert and a Record of his Patrimony, ed. & transl. Ted Johnson South(Cambridge 2002).

More reliable Insular sources, notably ‘The Annals of Ulster’ and ‘The Chronicleof the Kings of Alba’, record only one event.101 This battle was messy andindecisive (leading to both sides claiming victory), but it paved the way for thecapture of York by the Waterford-vikings in the following year.102

When York was seized by King Æthelstan in 927, troops from Waterford mayhave participated in efforts to win it back. A leader called Maelmuire son ofCossanara is named among the fallen at the battle of Brunanburh in 937 in ‘TheAnnals of Clonmacnoise’. Although it is not clear how reliable this list is, it isstriking that a Cossanara is identified in Cocad Gaedel re Gallaib as one of thevikings who came to Waterford in 914.103 The tenth-century evidencedemonstrates Waterford’s political involvement in the affairs of Britain andnorthwestern France.

CONCLUSIONS

How should we assess the historical significance of Viking-Age Waterford?Excavations at Woodstown can tell us more about the origins of viking-settlementin Ireland, and about links between Woodstown and Waterford. Historically,Waterford was strategically significant because it could be used to control accessto three navigable rivers which together provided access to the heartlands ofMunster and Leinster. The border-location of Waterford was an important factorin the early life of the port. The evidence of Irish chronicles for the mid-ninthcentury suggests that this area may have been fought over by rival viking-groups.Initially the fortunes of Waterford were closely tied to those of its sister-settlement at Dublin. From the mid-tenth century, however, the challenges facingthe dynasty of Ímar heightened rivalries between viking-towns. Dublin was drawnmore closely into the sphere of Leinster-politics, and Waterford’s history becameclosely linked to that of the overkings of Munster. Indeed Waterford’s alliancewith Dál Cais may have strengthened Brian Bóruma in his bid to dominate Irelandduring the late tenth and early eleventh centuries.


101 AU 917[=918].4; Historia Regum Anglorum, Part I s.a. $$$$ (Symeonis Monachi Opera Omnia,ed. Thomas Arnold [2 vols, London 1882/5], II.93); B. T. Hudson, ‘The Scottish Chronicle’, ScottishHistorical Review 77 (1998) 129–61, at pp. 150, 157.

102 W.S. Angus, ‘The annals for the tenth century in Symeon of Durham’s Historia Regum’,Durham University Journal 32 [new series,1](1939/40) 213–29, at p. 221.

103 Cogadh, ed. & transl. Todd, pp. 26/27 (§26).


The economic significance of Viking-Age Waterford is suggested by silver-hoards recovered from neighbouring counties. Waterford’s economic hinterlandseems to have been focused west of the River Barrow, and this may be associablewith the fact that hostility with Leinster was a fairly constant feature of Waterford’s history during the period under discussion. Excavation can add moreto our understanding of urban development in Ireland by revealing goodsimported into, produced at, and exported from the settlement. External contacts,with the empire of the dynasty of Ímar and a wider network of trading routes,highlight Waterford’s importance on a broader playing field. Discoveries at Woodstown will undoubtedly add to this picture.104

104 Since this article was written, Cathy Swift and Colmán Etchingham have set up a website,http://www.vikingwaterford.com/svwag.html, as part of a campaign to save Woodstown. Thisprovides up-to-date information on the site and opportunities for involvement.

APPENDIX 1: WATERFORD IN IRISH CHRONICLES, A.D. 914–926914AFM 910 Guill do thecht i ndErinn go ro ghabhsat hi Port Lairghe.

CS 913 Gaill do tiachtain a nErinn a bPort Lairge.

FAI §458 Coblach lánmhór Lochlannac[h] [do] ghabhail ag Port Lairge, 7 fochla Osraighe, .i. tuaisgeart Osraige, d’ionnradh dhóibh brad mór 7 iomad bó 7 eallaigh do bhreith dhóibh gonuige i longa.

AClon 910 There came new supplyes of Danes this year and landed at Waterford.

AFM 912 Nocobhlach mór do Ghallaibh do thocht go Loch Da Chaoch, go ro gabhsat longport and.

AU 914.5 Nocoblach mar di gentibh oc Loch Da Caech.

915AU 915.7 Tórmach mar meinic do ghentibh do thichtain oc Loch Da Chaech beos, 7 indred tuath 7 ceall Muman h-uadhibh.

AFM 913 Tormach mór meinic do Gallaibh do thiachtain co Loch Dá Chaoch bheos, 7 indradh tuath 7 ceall Mumhan leo do ghrés.

916AFM 914 Ainle mac Catháin, tighearna Uaithne Cliach, do bhásughadh la Gallaibh Locha Dá Chaoch.

AU 916.3 Annle m. Cathan, ri Uathne Cliach, do bhas o Gallaibh Locha Da Chaech.

AU 916.6 Gaill Locha Da Caech beos do innriuth Muman 7 Laighen.

AFM 914 Goill Locha Da Chaoch beos do indradh Mumhan 7 Laighen.

917CS 916 Sluaiccedh ffer nErenn la Niall mac Aodha, go Galloibh Locha da caech, gur marbta Gaill ocus Gaoidhil ann, um righ Cairrge Brachaide, ocus um taisech. H. Cernaigh .i. Maolfinnén mac Donnagáin, ocus um taoisech .H. Cremtainne .i. Fergus mac Muirigén, et alii multi occisi sunt.

AU 917.2 Sitriuc h. Imair cona chobluch do ghabail oc Cinn Fhuait i n-airiur Laigen. Ragnall h. hImair cona chobluch ailiu co Gallu Locha Da Chaech. Ár nGall oc Neimlid la Mumain. Ár n-aile la Eoganacht 7 Ciaraidhe.

AFM 915 Sitrioc ua h-Iomhair cona chobhlach do ghabháil oc Cind Fuaith i n-airer Laighen. Raghnall, ua h-Iomhair, co cobhlach oile go Gulla Locha Dá Chaoch.

918AU 918.4 Gaill Locha Da Caech do dergiu Erenn, .i. Ragnall rí Dubgall, 7 na da iarla, .i. Ottir 7 Graggabai 7 sagaith dóoib iar sin co firu Alban ….

AFM 916 Oitir 7 na Goill do dhul o Loch Dá Chaoch i nAlbain, 7 Constantin, mac Aedha do thabhairt catha dóibh, 7 Oitir do mharbhadh co n-ár Gall immaille friss.926AFM 924 Orgain Cille Dhara do Ghallaibh Puirt Loairge. A h-orgain doridhisi ó Ath Cliath isin m-bliadhain chedna.

CS 925 Orgain Cille dara do Galloibh Puirt Lairge. A hargain aridhisi ó Ath Cliath isin bliadain cedna.



APPENDIX 2: SILVER-HOARDS IN COUNTIES KILKENNY, TIPPERARY, AND WATERFORD BEFORE A.D. 1035

Coinless

Derrynahinch (Co. Kilkenny) Rings and hacksilver

Dysart (Co. Kilkenny) Ingot-fragment, arm-ring fragment

unlocalised (Co. Kilkenny) Rings

Bog of Cullen (Co. Tipperary) Arm-rings

Rathmooley (Co. Tipperary) 2 arm-rings

Kilmacomma (Co. Waterford) Arm-ring fragment, 10 ingot-fragments, rod-fragment

MixedDunmore Cave (Co. Kilkenny) ca 930 10 coins (Anglo-Saxon, Anglo-

Scandinavian, Kufic), ingot-fragment

Dunmore Cave (Co. Kilkenny) ca 970 14 coins (Anglo-Saxon, 6 in fragments), 4 ingots, buckle, 16 buttons (?)

Kilkenny West (Co. Kilkenny) ca 970 60+coins (Anglo-Saxon), ingot-rings

Knockmaon (Co. Waterford) ca 1000 14+ coins (Anglo-Saxon, Hiberno-Scandinavian, Continental), 3 arm-ring fragments

Coinunlocalised (Co. Tipperary) ca 942 19 coins (Anglo-Saxon)

near Kilkenny (Co. Kilkenny) ca 1035 ‘A large number’ (Anglo-Saxon, Hiberno-Scandinavian)

SOURCES

J. Sheehan, ‘Early Viking Age silver hoards from Ireland’, in Ireland and Scandinavia in the Early VikingAge, edd. Howard B. Clarke et al. (Dublin 1998), pp. 166–202

M. Blackburn & H. Pagan, ‘A revised checklist of coin hoards from the British Isles’, available online: http://www.fitzmuseum.cam.ac.uk/coins/hoards/index.html

8Living on the edge:

Scandinavian Dublin in the twelfth century

THE term ‘Scandinavian Dublin’ applies to the years before Pentecost 1171, thatis, prior to the seizure of Dublin by Earl Richard ‘Strongbow’ and the subsequentexecution of its last viking-king, Ascall mac Torcaill.1 My purpose in this chapteris to examine Dublin’s status as a frontier-polity. The port had a border-positionin Ireland. It was identified as a kingdom ruled and inhabited by foreigners (Gaillwas the term most commonly employed in Irish chronicles). Dublin also borderedthe sea, and therefore its neighbours may be counted as the kingdoms closest toIreland across the water, namely, Man and the Isles, England, Wales, and Scotland.

It can be argued that the history of Dublin during the twelfth century has beenlargely construed from the perspective of those neighbours who were most eagerto possess it, namely the English and the Irish. From an English historiographicalperspective, events there before the 1170s tend to be judged as a prelude to the

West over Sea. Studies in Scandinavian Sea-borne Expansion and Settlement before 1300. A Festschriftin Honour of Dr Barbara E. Crawford,edd. Beverley Ballin Smith et al. (Leiden: Brill 2007)

ISBN 978-90-04-15893-1

1Ascall’s death as execution is the version provided by Gerald of Wales, Expugnatio hibernica,I.21: Expugnatio Hibernica, The Conquest of Ireland, by Giraldus Cambrensis, edd. & transl. A.B.Scott & F.X. Martin (Dublin 1978), pp. 76–9. Ascall had escaped from Earl Richard’s and KingDiarmait’s successful assault on Dublin (ibid., I.16–17, pp. 66–9), probably to the Orkneys. Hereturned thence around Whitsun 1171, was defeated and captured for ransom, but was subsequentlyexecuted on the orders of Milo de Cogan, the Norman governor of Dublin. Irish chronicles simply note his death in the context of a battle for Dublin. (1) For ‘The Annals of The Four Masters’, seebelow, nn. 32 and 66. (2) Annala Uladh, Annals of Ulster, otherwise, Annala Senait, Annals of Senat;a Chronicle of Irish Affairs from A.D. 431, to A.D. 1541, edd. & transl. William M. Hennessy &B. MacCarthy (4 vols, Dublin 1887–1901; 2nd edn, by Nollaig Ó Muraíle, 1998), II.166–9: Ascallmac Torcaill, rí Átha Cliath, 7 Eoan a hInnsibh Orc ‘killed by the same Foreigners’ (but nonepreviously mentioned in annal 1171). (3) The Annals of Loch Cé. A Chronicle of Irish Affairs fromA.D. 1014 to A.D. 1590, ed. & transl. William M. Hennessy (2 vols, London 1871), I.144/5(1171.2): ‘Axoll mac Torcaill, rí Átha Cliath, ocus Heoin a hInnsibh Orc, do mharbad dona Gallaibcédna’ (essentially the same as [2], above; see the editor’s comments, ibid., I.144[–5], n. 3). (4) TheAnnals of Tigernach, ed. & transl. Whitley Stokes (2nd edn, 2 vols, Felinfach 1993), II.428 (1171.7):‘Cath Átha Cliath etir Miligh Gogan 7 Asgall mac Raghnaill, ríg Gall, co-ndorchair iarom andAschall 7 Eoan Lochlandach et alii multi’. (5) ‘Mac Carthaigh’s Book’, in Miscellaneous Irish Annals(A.D. 1114–1437), ed. & transl. Séamus Ó hInnse (Dublin 1947), pp. vii–xiii, 2–115, at 54/5(1171.2), records the deaths in battle of Iohannes Orcach 7 Turcaill (the latter’s appearance in thegenitive case encouraged the editor to supply in his translation ‘[Haskulf son of Raghnaill son of ]’between the last two words).

157

Angevin invasion.2 From an Irish perspective, the contest between principaloverkings to control Dublin has sometimes been regarded as part of a widerstruggle for national unity, a process which is deemed to have been cut short byHenry II’s arrival.3 These hindsight-approaches are by no means invalid.However, there may be scope for closer examination of Dublin’s quest for self-determination and power in relation to its neighbours before 1170.4

Three subjects will be briefly explored in this chapter. The first is the identityof the Dubliners in the twelfth century, the second is the town’s political dealingswith its Irish neighbours, and the last is Dublin’s connections across the Irish Sea.

IDENTITY

The subject of identity is undoubtedly complex, but some generalisations may bemade by dividing our evidence into the broad categories of assigned and self-expressed identity. Here it is relevant to mention R.R. Davies’s observationthat ‘the medieval construction of the world … was one which defined it as acollection of peoples; whatever modern historians care to believe, contemporarieswere not in doubt about the reality and solidity of such communities’.5 In otherwords the mediaeval European world defined people by ethnicity, no matter howfragile and arbitrary its labels might be.To some degree we can evaluate the identityof twelfth-century Dubliners from the opinions of their neighbours. Dublinduring the twelfth century comprised an urban settlement but also a ruralhinterland which was defined in Irish sources as Fine Gall or crích Gall (kindredor territory of [the] Foreigners).6 The use of the term Gaill in Irish-languagesources to refer to the inhabitants of Dublin in the twelfth century indicates that they were regarded as non-Irish, foreign, despite the presence ofviking-communities in Ireland since the ninth century.7 Further evidence thatthe Dubliners remained culturally distinct can be found in the writings of Geraldof Wales. He stated that the inhabitants of the viking-towns were different from

2C. Downham, ‘England and the Irish-Sea zone in the eleventh century’, Anglo-Norman Studies26 (2003) 55–73, at p. 55.

3Dáibhí Ó Cróinín, Early Medieval Ireland, 400–1200 (Harlow 1995), pp. 290–2; S. Duffy,‘Ireland’s Hastings: the Anglo-Norman conquest of Dublin’, Anglo-Norman Studies 20 (1997) 69–85, at p. 81.

4The approach in this paper has been stimulated by work, undertaken since the early 1990s bya range of scholars (for example, Ian Beuermann, Seán Duffy, and Kari Maund), which has beensignificant in broadening debate about Dublin’s history in the eleventh and twelfth centuries.

5R.R. Davies, ‘The peoples of Britain and Ireland, 1100–1400. I. Identities’, Transactions ofThe Royal Historical Society, 6th series, 4 (1994) 1–20, at p. 7.

6C. Downham, ‘Fine Gall’, in Medieval Ireland. An Encyclopedia, ed. Seán Duffy (New York2005), pp. 170–1.

7See above, chapter 3, especially p. $75$.


the Irish and that they were called Ostmen.8 This was a Germanic term meaning‘easterners’ (perhaps reflecting a self-perception among the people of Dublin thattheir ancestors came from the east).9 The Scandinavian customs and dress of thoseinhabiting Ireland’s major ports are also referred to by an Arab geographer, Tarsi‘al-Akhbar, who wrote around 1080.10 He assumed that the people whom heencountered reflected the character of Ireland as a whole and therefore identifiedit as a viking-land. While these accounts provide strong evidence that theinhabitants of Dublin retained Scandinavian cultural traits, they were notuniformly regarded as a Scandinavian people.

Old-West-Scandinavian literature displays an awareness that the people ofDublin had a claim to Scandinavian roots, but (as John Hines has pointed out)they were often labelled írskr, ‘Irish’.11 Therefore they were seen as different fromthe people of Iceland or Norway who wrote these texts, and the strange ways ofvikings from the Gaelic west are sometimes commented on in sagas.12 Thisambiguity is further highlighted in Welsh sources which sometimes refer toDubliners as being of Germanic stock and sometimes as being Irish. An exampleis the record of Dubliners being present on Anglesey in 1144. One version ofAnnales Cambriae and one version of Brenhinoedd y Saeson identify them asGermanici or Germanwyr.13 The other versions of these texts identify them as Irish(Hybernienses or Gwydil).14 In terms of their assigned identity, the inhabitants ofthe viking-towns of Ireland could be deemed a people apart: they were neitherpurely Irish nor purely Scandinavian. But how was identity defined in their owncommunity?

8J.J. O’Meara (ed.), ‘Giraldus Cambrensis in Topographia Hibernie: text of the first recension’,Proceedings of The Royal Irish Academy 52 C (1948–50) 113–78, at p. 175; Gerald of Wales, TheHistory and Topography of Ireland, transl. John J. O’Meara (2nd edn, Portlaoise/Harmondsworth1982), p. 122 (§117).

9E. Curtis, ‘The English and Ostmen in Ireland’, English Historical Review 23 (1908) 209–19,at p. 209, n. 1.

10D. James, ‘Two medieval Arabic accounts of Ireland’, Journal of The Royal Society of Antiquariesof Ireland 108 (1978) 5–9.

11John Hines, Old-Norse Sources for Gaelic History (Cambridge 2002), p. 21.12Gísli Sigurðsson, Gaelic Influence in Iceland. Historical and Literary Contacts. A Survey of

Research (Reykjavík 1988), pp. 86–102.13Annales Cambriæ, ed. John Williams (Ab Ithel) (London 1860), p. 42 (s.a. 1144), text B. For

Brenhinoedd y Saeson, text R, see Brut y Tywysogyon or The Chronicle of the Princes. Red Book ofHergest Version, ed. & transl. Thomas Jones (2nd edn, Cardiff 1973), pp. 118/19 (s.a. 1142).

14Annales Cambriæ, ed. Williams, p. 42, n. 13 (s.a. 1144), text C. For Brenhinoedd y Saeson,text P, see Brut y Tywysogyon Peniarth MS. 20, ed. Thomas Jones (Cardiff 1941), p. 90 (s.a. [1142]),and Brut y Tywysogyon or The Chronicle of the Princes. Peniarth MS. 20 Version, transl. Thomas Jones(Cardiff 1952), p. 53 (s.a. 1143); for text S, see Brenhinedd y Saesson or The Kings of the Saxons. BMCotton MS. Cleopatra B.V. and The Black Book of Basingwerk NLW MS. 7006, ed. & transl. ThomasJones (Cardiff 1971), pp. 148/9 (s.a. 1143).

SCANDINAVIAN DUBLIN IN THE TWELFTH CENTURY 159

The mediaeval Dubliners might be judged to express their identity in twobroad fashions. On the one hand, there was ceremonial expression throughsymbolic gestures on special occasions or among the political elite as a way ofenforcing a sense of common citizenship. On the other hand, there was mundaneexpression, acted out in everyday life through the media of language and materialculture. Of course, the two categories are interlinked; but they may be separatedhere for convenient analysis. The symbolic life of mediaeval Dublin was partlyexpressed through monuments. There was a Scandinavian-style thing-site,destroyed in 1685 but which (as Charles Haliday remarked) remained a focus forcivic ceremony for military or legal purposes long after the Viking-Age: forexample, civic dignitaries gathered on it to oversee the muster of the city’s forcesin 1593, and it is recorded as a place of capital punishment in 1613.15 Apillar-stone (called the Long Stone) also marked in symbolic fashion the entranceof Dublin-harbour, from the Viking-Age until the seventeenth century.16 Urbanceremonies may have been held around the assembly-hall mentioned by Geraldof Wales.17 The walls of Dublin may have also had more than a purely functionalrole. A substantial circuit was built around the town at the turn of the twelfthcentury.18 This perhaps served as a statement of civic pride and exclusivity, creatinga boundary between the town and the outside world.

Awareness of Dublin’s distinct political identity may have been promoted byorigin-legends about the port, which circulated in the twelfth century. Versionsof these legends are recorded in Historia Norwegie, composed in the mid-to-latetwelfth century, and in the twelfth-century Vita Griffini filii Conani and thewritings of Gerald of Wales. According to Gerald, the viking-towns of Dublin,Waterford, and Limerick were each founded by one of three brothers: Olaf, Sitric,and Ivar.19 According to Vita Griffini and its thirteenth-century Welsh translation,the legendary Norwegian king Harald Finehair built Dublin and his brother ruled

15Charles Haliday, The Scandinavian Kingdom of Dublin (2nd edn, Dublin 1884), pp. 156–99.16Ibid., pp. 143–55; P. Holm, ‘Viking Dublin and the city-state concept. Parameters and

significance of the Hiberno-Norse settlement’, in A Comparative Study of Thirty City-state Cultures,ed. Mogens Herman Hansen (København 2000), pp. 251–62, at p. 257.

17Expugnatio Hibernica, edd. & transl. Scott & Martin, pp. 66/7 (I.17).18P.F. Wallace, ‘The archaeological identity of the Hiberno-Norse town’, Journal of The Royal

Society of Antiquaries of Ireland 122 (1992) 35–66, at p. 45.19See above, n. 8.


Waterford.20 Although Historia Norwegie dates the foundation of the towns tothe time of Harald Finehair, credit is given there to the descendants of Røgnvaldof Møre, first earl of Orkney.21 Although these legends exhibit some variation,there is striking correspondence in that the viking-towns in Ireland were thoughtto have been founded at the same time by members of the same family. As theauthors of these accounts from Norway, Wales, and England were linked throughsome familiarity with Irish affairs, one might conclude that the legend circulated(and perhaps originated) among the people of Dublin. Comparison may be madewith the legend of the Rurikids, the dominant viking-dynasty of Russia. The storyis recorded in the twelfth century that three brothers – Rurik, Sineus, and Truvor– founded respectively the towns of Novgorod, Belozero, and Izborsk in themid-850s, and that Rurik was the ancestor of future viking-kings in Russia.22 Theexistence of a distinct origin-legend for the viking-towns of Ireland is one indicatorof their separate political identity.

What about the more everyday expressions of identity in Viking-Age Dublin?In summary, these highlight diverse cultural influences on life in the town anddemonstrate that the population was neither wholly Irish nor wholly Scandinavianin character. In terms of language, inscriptions which have been recovered fromDublin-excavations date from the late tenth to the early twelfth century.23 Theseare in a range of three scripts – ogham, Latin, and runes – and they exhibitknowledge of four tongues, Gaelic, Scandinavian, Latin, and English. As someof these inscriptions were found on mundane objects such as scraps of wood oranimal-bones, they may provide an insight into languages used by a semi-literateclass of urban dweller, and not just the elite. Other features of town-life emphasisethe cosmopolitan character of Dublin. This can be seen in two key-elements ofeveryday life, namely the homes in which people lived and the way in which theyprepared food.

20Vita Griffini filii Conani. The Medieval Latin Life of Gruffudd ap Cynan, ed. & transl. PaulRussell (Cardiff 2005), pp. 54–7 (§5); Historia Gruffud vab Kenan, ed. D. Simon Evans (Cardiff1977), pp. 3–4, 49–53; A Mediaeval Prince of Wales. The Life of Gruffudd ap Cynan, ed. & transl.D. Simon Evans (Felinfach 1990), pp. 25, 56. Cf. K.L. Maund, Ireland, Wales, and England in theEleventh Century (Woodbridge 1991), pp. 171–82; J. Jesch, ‘Norse historical traditions and HistoriaGruffud vab Kenan: Magnús berfœttr and Haraldr hárfagri’, in Gruffudd ap Cynan. A CollaborativeBiography, ed. K.L. Maund (Woodbridge 1996), pp. 117–47.

21Historia Norwegie, edd. & transl. Inger Ekrem et al. (København 2003), pp. 64–9 (§VI); seealso below, chapter 9, pp. 195~9.

22Wladyslaw Duczko, Viking Rus. Studies on the Presence of Scandinavians in Eastern Europe(Leiden 2004), pp. 80–1.

23Michael P. Barnes et al. (edd.), The Runic Inscriptions of Viking Age Dublin (Dublin 1997).On bilingualism more generally, see D. Ó Corráin, ‘Old Norse and medieval Irish: bilingualism inViking-Age Dublin’, in Dublin in the Medieval World. Studies in Honour of Howard B. Clarke, edd.John Bradley et al. (Dublin 2009), pp. 63–72.


In terms of domestic architecture, the so called ‘Type-1’ houses identified byPatrick Wallace predominate in the Dublin-excavations. These are characterisedby their rectangular shape, with a door at each end, central hearth, side-benches,and wattle-construction, and they have been found in archaeological levels datingfrom the ninth to the twelfth century.24 Type-1 houses became less popular duringthe twelfth century, but they persisted as a form distinct from contemporary Irishand Scandinavian dwellings. The twelfth century was also marked by increasinguse of sunken-floored dwellings and sill-beam construction.25 The house-typesfound in Viking-Age Dublin can be seen to represent distinctive urban styles whichdeveloped in Ireland through a combination of influences.

Dubliners differed from many of their Irish neighbours in their use of pottery.Large quantities were imported to Dublin in the twelfth century from Englandand Northern France.26 Apart from the souterrain-ware of Ulster (dating fromthe ninth century onwards) and Leinster Cooking Ware (dating from themid-twelfth century onwards) there was no tradition of pottery-making in Irelandbefore the Angevin invasion.27 The use of ceramics in Ireland seems to reflect theinfluence of Viking-Age trading networks rather than direct Scandinavian links.28

Pottery became a significant part of the material culture of the viking-towns dueto external trading links, and this in turn influenced customs of food-preparationand consumption.

The impression of eclecticism in Dublin’s material culture extends to otherexcavated finds which highlight the town’s contacts abroad.29 It is also exhibitedin the saints’ cults observed by the twelfth-century inhabitants of Dublin. Thistopic has been studied in detail by Howard Clarke, who has highlighted links with

24Wallace, ‘The archaeological identity’, pp. 47–8.25Ibid., pp. 55–60. Interesting further discussion has been provided by N.S. Price, ‘Acts of

identity? Ethnicity and architecture in the Viking Age’, Medeltids-arkeologisk tidskrift 94 (1994),nos 3–4, pp. 49–71.

26J.G. Hurst, ‘Medieval pottery imported into Ireland’, in Keimelia. Studies in MedievalArchaeology and History in Memory of Tom Delaney, edd. Gearóid Mac Niocaill & P.F. Wallace(Galway 1988), pp. 229–53.

27C. McCutcheon, ‘Brought to book – the potter and the pot’, Archaeology Ireland 20 (2006),no. 1, pp. 22–5.

28A.M. Lane, ‘Viking-Age and Norse pottery in the Hebrides’, in The Viking Age, Ireland andthe West – Papers from the Proceedings of the Fifteenth Viking Congress, Cork, 18–27 August 2005, edd.John Sheehan & D. Ó Corráin (Dublin 2010), pp. 204–16.

29P.F. Wallace, ‘The economy and commerce of Viking Age Dublin’, in Untersuchungen zuHandel und Verkehr der vor- und frühgeschichtlichen Zeit in Mittel- und Nordeuropa, IV, Der Handelder Karolinger- und Wikingerzeit, edd. Klaus Düwel et al. (Göttingen 1987), pp. 200–45.


Norway, through the cult of Olaf; with Gaelic Ireland, through veneration ofSS. Brigit, Patrick, and Columba, among others; with England, through theveneration of St Werburgh; and with the Continent through a range ofinternational saints.30

Of course, within Dublin many different identities may have been expressed,but in generalised terms the identity of the townspeople appears both eclectic anddivergent from that of their neighbours. Nevertheless, the boundaries betweenIrish and Hiberno-Scandinavian urban identity were permeable and blurred bycenturies of interaction and intermarriage. Viking-leaders spoke Irish andfrequently adopted Gaelic names.31 Conversely, Old-Scandinavian names foundtheir way into Irish families, as is shown by chronicle-records of the twelfthcentury.32 Dublin was economically dependent on the exchange of perishable and

30H.B. Clarke, ‘Christian cults and cult-centres in Hiberno-Norse Dublin and its hinterland’,in The Island of St Patrick. Church and Ruling Dynasties in Fingal and Meath, 400–1148, ed. AilbheS. Mac Shamhráin (Dublin 2004), pp. 140–58.

31B. Ó Cuív, ‘Personal names as an indicator of relations between native Irish and settlers in theviking period’, in Settlement and Society in Medieval Ireland. Studies presented to F.X. Martin, O.S.A.,ed. John Bradley (Kilkenny 1988), pp. 79–88; Holm, ‘Viking Dublin and the city-state concept’,p. 260; E. Bhreathnach, ‘Documentary evidence for pre-Norman Skreen, County Midhe’, Riochtna Midhe 9 (1994–8), no. 2, pp. 37–45, at 37–41; D. Ó hAodha, ‘The first Middle Irish metricaltract’, in Metrik und Medienwechsel. Metrics and Media, ed. Hildegard L.C. Tristram (Tùbingen1991), pp. 207–44, at p. 235; Stephen N. Tranter, Clavis Metrica: Háttatal, Háttalykill and the IrishMetrical Tracts (Basel 1997), especially pp. 40–51; D.N. Dumville, ‘What is mediaeval Gaelicpoetry?’, in Explorations in Cultural History: Essays for Peter Gabriel McCaffery, edd. David F. Smith& H. Philsooph (Aberdeen 2010), pp. 81–153, at p. 94 (and references given there).

32Annala Rioghachta Eireann, Annals of the Kingdom of Ireland, by The Four Masters, from theEarliest Times to the Year 1616, ed. & transl. John O’Donovan (2nd edn, Dublin 1856), II.1022–5(1126.6), Iomhar Ua Aedhacáin; II.1036/7 (1130.9), Amhlaoibh mac Mic Seanáin; II.1038/9,Raghnall Ua Heochadha (1131.5); II.1058/9, Raghnall mac Iomhair Ua Catháin (1138.6); II.1086/7,Sitriog Ua Braonáin (1148.8); II.1112/13, Gofraidh Ua Raghallaigh (1154.12). These borrowedScandinavian names occur again and again, from one end of Ireland to the other. Another indicatorof the extent of cultural penetration is provided by the name and status of the monk Amhlaeibh,‘son of the successor of Finnén’ (ibid., II.1178/9 [1170.31]). In addition to these relativelystraightforward borrowings of ninth-century antecedents of Old-Scandinavian Guðrøðr (asOld-Gaelic Gothfrith, Middle-Gaelic Gofraid) – see David N. Dumville, Celtic Essays, 2001–2007(2 vols, Aberdeen 2007), I.114–15, on these and related forms –, Ívarr (Ímar), Óláfr (Amlaíb),Røgnvaldr (Ragnall), and Sigtryggr (Sitriuc, etc.), there is a range of different types of influence ofthe Scandinavian name-stock on the Gaelic. As personal names (and, in time, family-names), GaelicAn Gall, Dubgall, and Lochlann indicate the acceptability and status of Scandinavian heritage, butnone is a lexical borrowing from Old Scandinavian. Old-Gaelic Glúniarann may be aloan-translation of Scandinavian Járnkné (presumably a nickname in origin), but there are olderknee-nicknames attested in Celtic (Old-Brittonic Glinmaur, for example). Gaelic Magnusmay havea long history as a borrowing from Latin, but its popularity in and after the Viking-Age must betestimony to the impact of Scandinavian usage.


manufactured goods inland.33 Art-styles crossed political boundaries.34 Thus thedifferences between the vikings of Dublin and their Irish neighbours were notnecessarily self-evident. They were reïnforced on the one hand by a consciousdecision to remain separate, and on the other hand by contacts abroad which drew Dublin within the cultural sphere of other trading communities around theIrish Sea.

POLITICAL RELATIONS WITHIN IRELAND

Dublin’s relations with her neighbours in Ireland were coloured by the attemptsof various kings to bring the wealthy port under their control. The political powerof Dublin was in decline from the late tenth century.35 While Dublin’s northerlyneighbours, the overkings of Mide, also suffered a series of setbacks from the earlyeleventh century, the opportunity arose for the overkings of Leinster, located tothe south, to impose their control over the town. In 1052 the Leinster-overkingDiarmait mac Maíl na mBó succeeded in expelling the viking-king and assumeddirect authority over the port.36 This set a precedent for his descendants in thetwelfth century who sought to bring Dublin under their control.37 Other Irishoverkings from more distant provinces were not content to be left out of thisstruggle. Dublin was by far the wealthiest settlement in Ireland, and its fiscal andmilitary resources offered overkings the opportunity to outdo their rivals who viedfor recognition as the most powerful ruler in Ireland.38 During the early twelfthcentury the Ua Briain dynasty of Munster acted as overkings of Dublin too. Thiscontrol slipped from their hands by 1118; the overking of Connaught, ToirdelbachUa Conchobair, took control.39 Rather than seeking to rule the port directly,

33M. Valante, ‘Dublin’s economic relations with hinterland and periphery in the later VikingAge’, Medieval Dublin 1 (1999) 69–83.

34H.B. Clarke, ‘Conversion, Church and cathedral: the diocese of Dublin to 1152’, in Historyof the Catholic Diocese of Dublin, edd. James Kelly & D. Keogh (Dublin 2000), pp. 19–50, at p. 38;Ruth Johnson, Viking Age Dublin (Dublin 2004), pp. 81–7; Lloyd Laing & J. Laing, Art of the Celts(London 1992), pp. 186–90.

35A. Mac Shamhráin, ‘The battle of Glenn Máma, Dublin and the high-kingship of Ireland: amillennial commemoration’, Medieval Dublin 2 (2000) 53–64.

36The Annals of Ulster (to A.D. 1131), I, edd. & transl. Seán Mac Airt & G. Mac Niocaill (Dublin1983), pp. 488/9 (AU2 1052.8, an addition to the text). See the seminal paper by S. Duffy, ‘Irishmenand Islesmen in the kingdoms of Dublin and Man, 1052–1171’, Ériu 43 (1992) 93–133, at p. 94.Cf. S. Duffy, ‘The royal dynasties of Dublin and the Isles in the eleventh century’, Medieval Dublin7 (2005) 51~65.

37Duffy, ‘Ireland’s Hastings’, p. 77.38D. Ó Corráin, ‘The vikings in Ireland’, in The Vikings in Ireland, ed. Anne-Christine Larsen

(Roskilde 2001), pp. 17–27, at p. 26.39The Annals of Tigernach, ed. & transl. Stokes, II.343 (1118.4–5).


Toirdelbach allowed the overking of Leinster, Énna mac Diarmata Mac Murchada,to hold it for him. When Énna died in 1126, Toirdelbach attempted to passDublin to his own son Conchobar, but the latter was ousted by the people ofDublin in 1127.40

It is not entirely clear who ruled the port after this coup, but by 1133 powerseems to have passed into the hands of a local king named Torcall (Old-ScandinavianÞorkell).41 It is not known whether Torcall was descended from the dynasty of Ívarr(whose members had ruled Dublin until 1094),42 but later records suggest that hebelonged to a family of major landowners in the Dublin-area.43 A close allianceensued between Dublin and the Leinster-overking Diarmait Mac Murchada.Diarmait used Dublin’s military might to wage war against Waterford in 1137 (andperhaps in 1140), thus demonstrating that no love was lost between theseviking-towns which had a long history of rivalry.44 Diarmait’s alliance with Dublinwas briefly disrupted in 1141 when Conchobar Ua Briain, overking of Munster,took control of the port.45 He was soon ousted, and the townspeople invited Ottar

40The Annals of Inisfallen (MS. Rawlinson B.503), ed. & transl. Seán Mac Airt (Dublin 1951),pp. 286/7 (1126.7–8); The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.572/3(1127.5); The Annals of Loch Cé, ed. & transl. Hennessy, I.120–3 (1127.5).

41Ibid., I.132/3 (1133.2).42This affiliation of King Gofraid is made clear in the notice of his accession in The Annals of

Tigernach, ed. & transl. Stokes, II.317 (1091.5), Goffraigh mac maic Arailt, rí Átha Cliath, and ofhis death in The Annals of Inisfallen, ed. & transl. Mac Airt, pp. 250/1 (1095.13), ro-marb ...Gobraith rí(g) Átha Cliath ocus Inse Gall. In The Annals of Ulster, edd. & transl. Mac Airt & MacNiocaill, I.528/9 (1094.2) and 530/1 (1095.11), he is Goffraigh méránach, rí Gall, expelled fromDublin in 1094. For Torcall’s descent-group, see below, p. 178.

43Clarke, ‘Conversion’, pp. 36–7; Duffy, ‘Irishmen’, p. 122, n. 46; Emmett O’Byrne, War,Politics and the Irish of Leinster, 1156–1606 (Dublin 2003), p. 25.

44The Annals of Loch Cé, ed. & transl. Hennessy, I.136–9 (1137.4); The Annals of Clonmacnoise,being Annals of Ireland from the Earliest Period to A.D. 1408, translated into English A.D. 1627 byConell Mageoghagan, ed. Denis Murphy (Dublin 1896), p. 196 (1137.3); Annala, ed. & transl.O’Donovan, II.1056/7 (1137.11), 1064/5 (1140.12), but in this last entry no mention is made ofKing Diarmait. On Waterford, see above, chapter 7.

45Ibid., II.1066/7 (1141.8). According to ‘Mac Carthaigh’s Book’ – Miscellaneous Irish Annals,ed. & transl. Ó hInnse, pp. vii–xiii, 2–115 –, Conchobar Ua Briain died at Killaloe in 1142, but in1143 Conchobar Ua Conchobar ‘was killed in treachery by the men of Mide ... after he had takenthe full kingship of Mide and Dublin’, ibid., pp. 28/9 (1142.1 and 1143.2).


mac meic Ottair, from the kingdom of the Isles, to become their king.46 Ottar(Old-Scandinavian Óttarr) initially coöperated with the sons of Torcall, but thelatter brought about his death in 1148.47 Thereafter the sons of Torcall controlledDublin, at first under the supervision of Diarmait Mac Murchada.

During the 1150s Diarmait was subject to Muirchertach Mac Lochlainn,overking of the Northern Uí Neill, and this presumably gave Muirchertach somemeasure of control over Dublin. Despite Muirchertach’s attempt to buy thesupport of the Dubliners in 1154 with a stipend of 1,200 cows,48 in 1156/7 theDubliners sought to obtain Godred son of Olaf (Guðrøðr Óláfsson), king of Man,as their leader.49 After Brodar son of Torcall, king of Dublin, was killed in 1160,50

in 1161 the Dubliners and Diarmait Mac Murchada submitted to MuirchertachMac Lochlainn,51 but this cannot have been sustained, for in 1162 Mac Lochlainnbrought an army to Dublin ‘to take vengeance ... for his wife and her violation’.52

46On the years 1142–8, cf. Duffy, ‘Irishmen’, pp. 121–3. The chronicle-sources are various.Annala, ed. & transl. O’Donovan, II.1068/9 (1142.13): Mac mic Ottir, .i. Ottir do lucht Insi Gall,do gabháil cheannais 7 forlámhais Átha Cliath. But the gloss creates an ambiguity: was the grandsoncalled Ottar, or does the gloss identify the grandfather? If the king created in 1142 was he who wasassassinated in 1148, the former option is correct. But a further complication (apparentlyforeshortening the pedigree) is created by The Annals of Clonmacnoise, ed. Murphy, p. 193(1134[=1142].6): ‘Mc. Otyr, prince of Inis Gall, was chosen to be prince of the Danes of Dublin’.However, Mac Ottair may by now have become a surname. If the new king in 1142 was he whodied in 1148, his reign was interrupted at some point by that of Ragnall mac Turcaill who died in1146 (Annala, ed. & transl. O’Donovan, II.1081/2 [1146.3]; The Annals of Tigernach, ed. & transl.Stokes, II.382 [1146.9]; Chronicum Scotorum. A Chronicle of Irish Affairs, from the Earliest Times toA.D. 1135; with a Supplement, containing the Events from 1141 to 1150, ed. & transl. William M.Hennessy (London 1866), pp. 342/3 [1146.2]; ‘Mac Carthaigh’s Book’, 1145.1 [in MiscellaneousIrish Annals, ed. & transl. Ó hInnse, pp. 28/9]). The Welsh chronicles describe coöperation in Walesin 1144 by Octer (PS)/Otter (R) son of Octer/Otter, the son of Twrkyll (P)/Turkyll (RS), and theson of Yscherwlf (P)/Cherwlf (R)/Cherulf (S): see above, nn. 13–14. Ragnall’s reign presumablytherefore began no earlier than 1144.

47See the references given in nn. 13–14, above, to Welsh chronicles’ accounts of the year 1144.For the Irish evidence, see Chronicum Scotorum, ed. & transl. Hennessy, pp. 342–5 (1146.2, 6: deathof King Raghnall mac Turcaill), 345/6 (1148.5); The Annals of Tigernach, ed. & transl. Stokes, II.384(1148.3).

48The Annals of Clonmacnoise, ed. Murphy, p. 205 (1161[=1154].2); Annala, ed. & transl.O’Donovan, II.1112/13 (1154.12).

49Cronica Regum Mannie & Insularum, Chronicles of the Kings of Man and the Isles, ed. & transl.George Broderick (2nd edn, Douglas 1996), folios 36v–37r (s.a. 1144[=1154]).

50‘The Annals of Boyle’, 1160.2: A.M. Freeman (ed.) & R. Flower, ‘The Annals in Cotton MS.Titus A.xxv’, Revue celtique 41 (1924) 301–30, 42 (1925) 283–305, 43 (1926) 358–84, 44 (1927)336–61, at p. 288; Annala, ed. & transl. O’Donovan, II.1138/9 (1160.12); Annala Uladh, edd. &transl. Hennessy & MacCarthy, II.136/7 (1160.5); The Annals of Tigernach, ed. & transl. Stokes,II.404 (1160.5).

51Ibid., II.406 (1161.6).52Ibid., II.407 (1162.1), where [Lochlainn] has had to be supplied by the editor.


Diarmait Mac Murchada reïmposed his authority in 1162.53 He may have soughtto consolidate it by disenfranchising the descendants of Torcall (Meic Thorcaill):Diarmait granted lands of Meic Thorcaill in Baldoyle (Co. Dublin) to ChristChurch Cathedral in 1162; and Diarmait’s brother-in-law, Lorcán Ua Tuathail,was consecrated to the see of Dublin in the same year.54 Diarmait failed to securethe loyalty of the Dubliners through these actions. When Ruaidrí Ua Conchobair,overking of Connaught, challenged Diarmait in 1166, the Dubliners were drawnto his side; this coup was presumably assisted by an advance-promise of Ruaidrí’spayment of 4,000 cows to the town.55 Ruaidrí’s success caused Diarmait to fleeIreland; his exile led ultimately to the Angevin invasion of 1171.56

One factor which emerges from this narrative is how difficult it was for anIrish overking to dominate Dublin without the coöperation of leadingtownspeople. Alien rulers who imposed their authority, Conchobar UaConchobair in 1126 and Conchobar Ua Briain in 1141, lasted only a year beforethey were ejected.57 Uí Chennselaig overkings of Leinster had enjoyed moresuccess. Members of that dynasty dominated Dublin without any significantchallenges to their power for up to six years at a time.58 Presumably theintermittent hold of Uí Chennselaig over the town since the mid-eleventh centurygave them some legitimacy in the eyes of the Dubliners, and they may have madeefforts to win the support of the wealthy burgesses of the town. It seems that Irishoverkings enjoyed greater success in controlling Dublin when they had the supportof indigenous leaders. The emergence of the lineage of Torcall during the twelfth

53Annala Uladh, edd. & transl. Hennessy & MacCarthy, II.142/3 (1162.5, 7).54Registrum Prioratus Omnium Sanctorum juxta Dublin, ed. Richard Butler (Dublin 1845),

pp. 50–1 (no. XLIX); Irish Royal Charters. Texts and Contexts, ed. & transl. Marie Therese Flanagan(Oxford 2005), pp. 64–85, 269–81 (no. 3).

55Annala Uladh, edd. & transl. Hennessy & MacCarthy, II.152/3 (1166.9); The Annals ofTigernach, ed. & transl. Stokes, II.413–15, 416, 416–17, 417–18 (1166.1, 6, 11, 17–18); Annala,ed. & transl. O’Donovan, II.1160/1 (1166.11), reporting that at Dublin ro-ríghadh ann RuaidhriUa Concobhair febh as onóraighe ro-ríghadh rí riamh do Ghaoidhealaibh. Cf. ‘The Annals of Boyle’,edd. Freeman & Flower, p. 289 (1166.2), for an even more elaborate account.

56Duffy, ‘Ireland’s Hastings’, p. 78.57Cf. n. 40, above, for 1126/7, and The Annals of Ulster, edd. & transl. Mac Airt & MacNiocaill,

I.572/3 (1127.5); The Annals of Loch Cé, ed. & transl. Hennessy, I.120/1 (1126.6) and 120–3(1127.5). For 1141/2, see above, n. 46.

58Duffy, ‘Irishmen’, p. 96. Cf. Donncha Ó Corráin, Ireland before the Normans (Dublin 1972),pp. 105 (‘Dublin came very definitely into Irish hands in the eleventh century, and became sodefenceless that it was hard put to defend itself against the petty kings of Brega’), 130 (‘long beforeClontarf [1014] the Norse had become a minor political force in Irish affairs’), 133–7 (on the careerof King Diarmait mac Maíl na mbó, 1032–72), 137–50 (on the Ua Briain overkings of Munster,1072–1119), for summary judgments on the changing place of Dublin in Irish politics.


century may belong to this context.59 When a local leader was not forthcoming,the people of Dublin might appeal to the kingdom of Man and the Isles for aleader rather than be ruled by a non-Leinster overking.60 The Dubliners mighthave preferred leaders of mixed Gaelic-Scandinavian descent with whom theycould more readily identify.

Dublin’s independence was also expressed in the ecclesiastical sphere. Duringthe late eleventh and early twelfth centuries its bishops were consecrated byarchbishops of Canterbury, thus avoiding subjection to an ecclesiastical authoritywithin Ireland.61 This was a source of controversy in 1121 when Cellach,archbishop of Armagh, sought to bring the town under his jurisdiction.62 Dublin’ssegregation from the Irish ecclesiastical hierarchy was ended when Diarmait MacMurchada controlled the town. In 1152 the Synod of Kells-Mellifont-Droghedamade Dublin the seat of an archbishopric of Leinster.63 When Diarmait broughtDublin under his direct control in 1162, he secured this Church-dignity for hisbrother-in-law Lorcán and then used a series of ecclesiastical endowments andalliances to try to bring the port more tightly within his grasp.64 In particular,Diarmait’s foundation of Augustinian houses in Dublin – at All Hallows near thesymbolic Long Stone, and at St Mary del Hogges near the thing-site – seemintended to mark his authority on ceremonial space within the city.65 Indeed, in

59For a genealogical chart showing the recoverable filiation of Torcall and his descendants in themiddle-decades of the twelfth century, see p. 168, below. It could be that Turcall mac Éola, killedin southwestern Wales by Normans in 1093 (The Annals of Inisfallen, ed. & transl. Mac Airt,pp. 244/5 [1093.5]), represented an earlier filiation of this family.

60Cf. n. 49, above: The Annals of Clonmacnoise, ed. Murphy, p. 193 (1134[=1142].6); Cronica,ed. & transl. Broderick, folios 36v–37r (s.a. 1144[=1154]).

61Marie Therese Flanagan, Irish Society, Anglo-Norman Settlers, Angevin Kingship – Interactionsin Ireland in the Late Twelfth Century (Oxford 1989), pp. 30–1. Cf. n. 98, below.

62The Annals of Ulster, edd. & transl. Mac Airt & Mac Niocaill, I.566/7 (1121.7); The Annalsof Loch Cé, ed. & transl. Hennessy, I.115/16 (1121.5); Annala, ed. & transl. O’Donovan, II.1010–13 (1121.1); see also the text presented and discussed by David N. Dumville [et al.], Saint Patrick,A.D. 493–1993 (Woodbridge 1993), pp. 259–64.

63The Annals of Tigernach, ed. & transl. Stokes, II.389 (1152.2), which places the meeting atDrogheda; cf. David N. Dumville, Councils and Synods of the Gaelic Early and Central Middle Ages(Cambridge 1997), p. 44.

64M.T. Flanagan, ‘Strategies of lordship in pre-Norman and post-Norman Leinster’,Anglo-Norman Studies 20 (1997) 107–26, at pp. 119, 125.

65Ibid., pp. 118–19.


the years 1162–6 Diarmait, it could be argued, was seeking to make Dublin thecapital of his overkingdom, and indeed capital of Ireland.66

This brings me to the question whether Dublin could indeed be regarded asthe capital of Ireland before 1170. Some historians have called it so, pointing outthat control of Dublin was essential for overkings who sought to impose their ruleacross the island.67 This was undoubtedly the case in economic terms because ofthe resources which Dublin represented. Irish chronicles also imply that a levelof pomp was involved when great rulers imposed their authority over the town.The large stipends paid by Muirchertach Mac Lochlainn and Ruaidrí UaConchobair certainly convey that impression.68 Although there was growingawareness of Dublin’s importance, the port-town (it could be argued) remainedtoo foreign in its associations in the twelfth century to be successfully establishedas capital of Ireland in any symbolic sense.

Dublin was clearly regarded as significant in the literature of this period, itsdún being celebrated in one poem of the Laídshenchas of Leinster.69 Nevertheless,the alien nature of its inhabitants and the validation of Irish claims to rule it arerecurrent themes. As Donnchadh Ó Corráin has pointed out, Irish royalpropaganda focused on the evil deeds of the vikings’ ancestors as a means of

66Duffy, ‘Irishmen’, pp. 128–9, and ‘Ireland’s Hastings’, p. 77. The coronation of Ruaidrí atDublin in 1166 (n. 55, above) might seem to be the end of its own royalty. It is striking that thekings and kingdom of Dublin did not disappear, however. In 1167 Raghnall mac Raghnaill is ‘theking of the Dubliners’ (Annala, ed. & transl. O’Donovan, II.1162–5 [1167.5]). Ruaidrí must havedelegated the immediate overkingship of Dublin to Diarmait Ua Maíl Shechlainn, overking of Mide,for at the latter’s death in 1169 (The Four Masters have their entries deranged, thereby making himlead a military campaign after his death: ibid., II.1172/3 [1169.4, 9]) he is ‘king of Mide/Tara andof the Foreigners of Dublin [and of Uí Fhailge and Uí Fhaeláin]’; for interpretation of that situation,see Ó Corráin, Ireland, pp. 166–7. In 1170, his brother Ascall was in office when the Dublinersdeserted Ruaidrí Ua Conchobair and went over to but then double-crossed Diarmait Mac Murchada:Ascall mac Raghnaill mic Thurcaill had the grand title airdrí Gall Átha Cliath, according to TheFour Masters, when he fled (Annala, ed. & transl. O’Donovan, II.1176/7 [1170.13]); and in thefollowing year when, with Eoan Lochlandach a hInsibh hOrc, he was killed by the Normans, hisobituary calls him Asgall mac Racchnaill, rí Gall nÉireann fecht riamh (ibid., II.1184/5 [1171.16]).

67Duffy, ‘Ireland’s Hastings’, pp. 80–1; F.J. Byrne, ‘Senchas: the nature of Gaelic historicaltradition’, Historical Studies (Irish Conference of Historians) 9 (1971) 137–59, at p. 140; Ó Corráin,‘The vikings in Ireland’, pp. 26–7. Cf. Dumville, Celtic Essays, 2001–2007, I.165 and n. 26; AubreyGwynn, The Irish Church in the Eleventh and Twelfth Centuries (Dublin 1992), pp. 69, 75–7.

68Annala, ed. & transl. O’Donovan, II.1112/13 (1154.12), 1160/1 (1166.11); The Annals ofClonmacnoise, ed. Murphy, p. 205 (1161.2); The Annals of Tigernach, ed. & transl. Stokes, II.413–15, 417–18 (1166.1, 18).

69The Book of Leinster, formerly Lebar na Núachongbála, edd. R.I. Best et al. (6 vols, Dublin1954–83), I.206–12, at p. 212, line 6362; M.E. Dobbs (ed. & transl.), ‘On the graves of Leinstermen’, Zeitschrift für celtische Philologie 24 (1953/4) 139–53. This text was drawn into the discussionby H.B. Clarke, ‘The topographical development of early medieval Dublin’, Journal of The RoyalSociety of Antiquaries of Ireland 107 (1977) 29–51, at p. 38.


justifying Dublin’s subjection to native Irish hegemony.70 Rather than beingcentral to Irish identity, the port symbolised a foreign people whose character wasnegatively contrasted with that of the Irish.

In twelfth-century literature Tara remained prominent as a symbol for Ireland’sunity.71 Nevertheless, its links were with the distant past, and it was not promotedas a potential site for a new capital. The gathering of troops by Ruaidrí UaConchobhair at Tara in 1169, before they marched against Diarmait MacMurchada and his allies from Britain, paid homage to Tara’s legendarysignificance.72 Uí Chonchobair overkings celebrated also the oenach or fair ofTailtiu in 1120 and 1167.73 This was another ancient site rich with symbolismconcerning the concept of Irish unity.74 Various synods and royal gatheringschaired by Irish overkings during the twelfth century sought likewise to foster asense of national identity, but these events were not concentrated on Dublin.75

It can be argued that Dublin avoided being the capital of Ireland before 1170through its own struggle for self-determination. Dublin’s persistent efforts toremain distinct, expressed in its origin-legends and symbols of statehood, mayhave created an ideological obstacle for overkings who sought to rule all Ireland.Nevertheless, the Dubliners ultimately failed in their quest for independence, asthe port and its hinterland were annexed in 1171 by King Henry II.

POLITICAL RELATIONS ABROAD

The last theme of this chapter is Dublin’s political links across the Irish Sea. Thekingdom of Man and the Isles I have already mentioned as an important potentialsource of alternative leaders for Dublin. The brief unity of the kingdoms undera single leadership in the second half of the eleventh century, attested by ‘TheAnnals of Inisfallen’ (and alleged by the non-contemporary Manx Chronicle),76

70D. Ó Corráin, ‘Nationality and kingship in pre-Norman Ireland’, Historical Studies (IrishConference of Historians) 11 (1975) 1–35, at pp. 31–2.

71Edel Bhreathnach, Tara: a Select Bibliography (Dublin 1995), nos 17, 18, 59, 62, 71, 84–6,98, 154, 155, 163, 168, 170, 179.

72Annala, ed. & transl. O’Donovan, II.1172/3 (1169.9).73The Annals of Tigernach, ed. & transl. Stokes, II.345 (1120.2), 421 (1167.7).74Ó Corráin, ‘Nationality’, pp. 20–1; Proinsias Mac Cana,

75Dumville, Councils, pp. 38–9. See, for example, The Annals of Tigernach, ed. & transl. Stokes,II.326 (1101.8), 334–5 (1111.6), 345 (1120.2), 364 (1134.2), 376 (1143.5), 378–9 (1144.7–8),383–4 (1148.1), 398 (1157.3), 407 (1162.4), 417–18 (1166.18), 421 (1167.7), 421 (1168.3), 432(1172.12).

76See n. 42, above, for the Irish evidence. For problematic testimony, see Cronica, ed. & transl.Broderick, folios 32v–33r (s.a. 1056[=1072?]); the implied dating there of Godred Crouan’s reignis 1072–88.


would have given kings of Man and the Isles precedent for claiming power inDublin. Irish leaders sought to capitalise on these links by intermarrying withthe Manx royal dynasty.

The Ua Briain dynasty of Munster seems to have intermarried with the familyof Echmarcach mac Ragnaill, who in the mid-eleventh century ruled Dublin andthen (in exile from 1052) in the Isles and perhaps Galloway.77 This led to UaBriain dynasts ruling Man and the Isles briefly in the 1090s and again in thesecond decade of the twelfth century.78 Intrusion by Uí Briain into this strategicsphere of Irish-Sea politics seems to have prompted kings from Norway andEngland to engage with Manx politics during the twelfth century to ensure thatthe area did not fall under Irish control.79

Magnús Barelegs, king of Norway, led campaigns into the Irish Sea from 1098to 1103.80 He took control of the Isle of Man and seems to have issued threats to Irish kings. However, the end of his campaigning came when he was killed in Ulster.

According to the non-contemporary Manx Chronicle, Henry I, king ofEngland 1100–35, brought Olaf, son of Godfrey/Godred Crouan, future king ofMan and the Isles, to be raised at the Anglo-Norman court.81 Olaf’s nephews(sons of his brother Harold) were, according to the same source, raised in Dublin.82

The close relationship between Man and Dublin erupted into a complex series ofpower-struggles in the 1140s and 1150s. In 1142, Ottar grandson of Ottar, aruler from the Isles, went to be king of Dublin:83 this presumably posed a challengeto the status of Olaf, king of Man. In 1152/3 Ragnall son of Harold (RøgnvaldrHaraldsson) then came from Dublin to contest the kingdom of Man and the Isles,killing Olaf in the process; Ragnall’s reign was nevertheless short-lived as he waskilled by Olaf ’s son Godred who returned from Norway soon afterwards.84

Godred, once established as king of Man and the Isles, was in 1156/7 invited to

77Cf. Duffy, ‘Irishmen’, pp. 96–101. The marriage of Donnchad mac Briain with a daughterof Ragnall mac Ragnaill, king of Waterford, is not at all well attested: The Annals of Inisfallen, ed.& transl. Mac Airt, pp. 198/9 (1032.6) is a twelfth-century interpolation and is in any case hardlyunambiguous. That Echmarcach mac Ragnaill was probably the brother-in-law of Donnchad is avery uncertain proposition.

78Duffy, ‘Irishmen’, pp. 108–16.79Ian Beuermann, Man amongst Kings and Bishops. What was the Reason for Godred Olafsson’s

Journey to Norway in 1152/3? (Oslo 2002), pp. 133–46.80For discussion, see R. Power, ‘Magnus Barelegs’ expeditions to the West’, Scottish Historical

Review 65 (1986) 107–32.81Cronica, ed. & transl. Broderick, folio 35r (s.a. 1098).82Ibid., folio 36r/v (s.a. 1142[=1152]).83Annala, ed. & transl. O’Donovan, II.1068/9 (1142.13); cf. n. 46, above.84Cronica, ed. & transl. Broderick, folio 36v (1143[=1153].3). A narrative account of the

assassination of Olaf stands in the previous annal: folio 36r/v (1142[=1152]).


rule Dublin, but Muirchertach Mac Lochlainn prevented this.85 Muirchertachlater gave his daughter Finnguala (Phingola) in marriage to Godred, perhaps inan attempt to heal relations.86 The feud between Ottar’s family and Godred’sfamily continued during this time (1155–8) and Thorfinn son of Ottar (ÞorfinnrÓttarsson) attempted to raise Dubgall, son of Somerled (king of Argyll, Godred’sbrother-in-law), to the kingship of Man.87 The series of difficulties faced by Olafand his son Godred, as kings of Man, caused them to turn increasingly to Norwayfor support. Godred visited the Norwegian royal court in 1152/3 and in 1160(after he had been ousted from the throne by Somerled).88

It seems that the impact of these developments was to provoke efforts by KingHenry II (1154–89) to control Irish-Sea politics. Some of Henry’s most loyaladvisers may have encouraged this. During the preceding civil war (1135×1154),the geographical base of Henry’s support was focused along the southwestern coastof England and drawn from men whose livelihoods depended in part on Irish-Seatrade.89

As for Wales, Dublin had involvements there too.90 The participation ofMuirchertach Ua Briain in Irish-Sea politics led him to support Gruffudd apCynan, a claimant to the kingship of Gwynedd who was raised in Dublin.91

Gruffudd benefited from the hospitality of Muirchertach during periods of exile,as did his ally Owain who fled to Ireland in 1109.92 Muirchertach’s support ofthese Welsh royals conflicted with the interests of Henry I, who sought to extendAnglo-Norman rule in Wales. After Muirchertach’s death, Irish participation in

85Ibid., folios 36v–37v (1144[=1154]).86Ibid., folios 39v–40r (1176.2); we learn this because, a quarter-century later, the visiting papal

legate, Cardinal Vivian, deemed the marriage not to have been lawfully completed and arranged forthis to be done.

87Ibid., folios 35v (1102), 36v–37v (1144[=1154]), 37v (1156), 37v–38r (1158.1).88Beuermann, Man, pp. 233–53. The chronology given in Cronica regum Mannie et Insularum

attributes these visits to 1152(–?) and 1158–64/5: Cronica, ed. & transl. Broderick, folios 36r(1152.1) and 39r/v (1164.2).

89Austin Lane Poole, From Domesday Book to Magna Carta, 1087–1216 (2nd edn, Oxford 1955),pp. 138–40.

90For an extended discussion of interaction between Wales and Ireland (both Scandinavian andIrish), see Historia Gruffud, ed. Evans, pp. lix–cxvii.

91Vita Griffini, ed. & transl. Russell; Historia Gruffud, ed. Evans; A Mediaeval Prince, ed. &transl. Evans.

92The evidence comes from the three versions of Brenhinoedd y Saeson: for the P-text, see Bruty Tywysogyon Peniarth MS. 20, ed. Jones, p. 44, and Brut ... Chronicle ... Peniarth, transl. Jones,pp. 20–1 (1096[=1098] and 1097[=1099]), 31 (1106[=1109]), 33 and 34–5 (1107[=1110]), 35–6 (1108[=1111]); for the R-text, see Brut ... Red Book, ed. & transl. Jones, pp. 36–9 (1094=[1098]and 1095[=1099]), 58–63 (1105[=1109]), 70–5 (1106[=1110] and 1107[=1111]); for the S-text,see Brenhinedd, ed. & transl. Jones, pp. 88–93 (1096[=1098] and 1097[=1099]), 108–11(1106[=1109]), 114–19 (1107[=1110] and 1108[=1111]).


Welsh politics declined, partly through Henry’s own efforts.93 However, theselinks resurfaced during the English ‘Anarchy’. The sons of Gruffudd ap Cynanfought against Norman leaders in Wales in 1138, and a viking-fleet from Irelandwas recruited to attack St Dogmaels.94 This fleet was no doubt encouraged by theopportunity to capture booty. In 1144 the sons of Gruffudd fell out: Cadwaladrraised a fleet in Dublin to attack his brother Owain; but the brothers arranged atruce. The Dublin-fleet was denied the opportunity of a lucrative civil war andsufficient payment for their services in coming to Wales; so its crews capturedCadwaladr and held him to ransom for 2,000 captives. Owain attacked theDubliners’ ships, and they were forced to flee home.95 The ransom-demand maygive an insight into the continuity of the slave-trade between Ireland and Walesduring these years. After 1144 there is no further record of Dublin’s war-fleetsbeing used in Wales, until 1165 when King Henry II briefly hired a Dublin-fleetto attack the Welsh.96 The interim silence in the records may – in respect of itssecond decade – be due to Henry’s diplomatic efforts to prevent Dublin’s resourcesbeing used against his interests in Wales.

Turning to England, I reïterate that Dublin’s involvement in Irish-Sea politicsled successive English kings to seek greater control there. The potential for royalenemies to find assistance in Ireland was demonstrated in 1102 when the

93Poole, From Domesday Book, p. 290.94Annales Cambriæ, ed. Williams, p. 41 (text C, annal 1138), gives a rather different view from

that in the vernacular texts. The sons (from where is not stated) ‘cum .xv. nauibus gentilibus plenisuenerunt ad Aberteyui et treugas fecerunt usque ad festiuitatem Sancti Martini. Gentiles uerospoliauerunt uillam et ecclesiam de Landedoch, id est de Sancto Dogmaelo, et predam maximamad naues portauerunt.’

95Ibid., p. 42 (text B, annal 1144); text C has a radically summary account, printed ibid., n. 13.For the evidence of the versions of Brenhinoedd y Saeson, see the references to the annal for 1144given in my n. 13, above. The texts differ as to whether bondmen, captives, cattle, or marks (cash)were demanded and paid, but all agree on the two thousand: Brut ... Chronicle ... Peniarth, transl.Jones, p. 175 (ad 53/23).

96Annales Cambriae have no information on this matter; that derives instead from the versions ofBrenhinoedd y Saeson. For text P, see Brut y Tywysogyon Peniarth MS. 20, ed. Jones, pp. 110–12, andBrut ... Chronicle ... Peniarth, transl. Jones, pp. 63–4 (1164[=1165].1); for text R, see Brut ... RedBook, ed. & transl. Jones, pp. 144–7 (1163[=1165].1); for text S, see Brenhinedd, ed. & transl. Jones,pp. 164–7 (1164[=1165].1). All versions stress the uselessly small size of the fleet which came fromIreland. For later involvement of Hiberno-Scandinavians in Wales, see S. Duffy, ‘The 1169 invasionas a turning-point in Irish-Welsh relations’, in Britain and Ireland 900–1300. Insular Responses toMedieval European Change, ed. Brendan Smith (Cambridge 1999), pp. 98–113, at 106–8.


Pembrokeshire rebel Arnulf de Montgomery married the daughter ofMuirchertach Ua Briain in return for a promise of military support.97

To some extent, rulers of England sought to influence Irish affairs in the sphereof ecclesiastical politics. Successive archbishops of Canterbury pursued their claimto authority over the Irish Church from the late eleventh century. Dublin wasused as a foothold for these ambitions. Its bishops were consecrated by successivearchbishops of Canterbury in 1074 and 1085, 1096, and 1121.98 King Stephen(1135–54) tried to obstruct Irish efforts to gain papal recognition for anecclesiastical hierarchy which made Dublin free from English control.99 Despitethis, in 1152 Dublin became the seat of an archbishopric which was fullyincorporated within the Irish diocesan system. The subsequent efforts of theEnglish Church to gain papal sanction for military intervention in Ireland havebeen well documented and debated by historians.100 However, ecclesiasticalmatters alone cannot explain the invasion.

A key-debate concerning Henry II is whether he had always nurturedambitions to conquer Ireland, or whether the invasion of 1171 was a response toa particular chain of events. A case can be made that Henry sought to stabiliserelations with Ireland through good connexions with Diarmait Mac Murchadaprior to Ruaidrí’s coup in 1166.101 Diarmait had controlled Dublin and thesoutheastern coasts of Ireland intermittently from the 1140s. Both men sharedinterests in Bristol and Irish-Sea trade. Bristol was the major trading partner of

97The Annals of Inisfallen, ed. & transl. Mac Airt, pp. 258/9 (1102.6); A. Candon, ‘MuirchertachUa Briain, politics and naval activity in the Irish Sea, 1075 to 1119’, in Keimelia, edd. Mac Niocaill& Wallace, pp. 397–415; E. Curtis, ‘Murchertach O’Brien, high king of Ireland, and his Normanson-in-law, Arnulf de Montgomery, circa 1100’, Journal of The Royal Society of Antiquaries of Ireland51 (1921) 116–24.

98Dumville, Councils, pp. 36–46; Flanagan, Irish Society, pp. 21–3, 29–31. For the documentaryevidence, see Canterbury Professions, edd. Michael Richter & T.J. Brown ([London] 1973), pp. 29(no. 36), 31 (no. 42), 34 (no. 51), 39 (no. 69). For a recent contextual discussion, seeD.N. Dumville, ‘Frivolity and reform in the Church: the Irish experience, 1066–1166’, Studies inChurch History 48 (2012) 47–64.

99Beuermann, Man, p. 214.100Maurice Sheehy, When the Normans came to Ireland (Cork 1975), pp. 10–16, 23; Flanagan,

Irish Society, pp. 7–55; C. Kostick, ‘Laudabiliter: a new interpretation by Professor Anne Duggan’,History Ireland 13 (2005), no. 3, pp. 7–8. The importance of trade in Anglo-Irish relations isillustrated by the events of 1102 and 1170: William of Malmesbury, Gesta Regum Anglorum, TheHistory of the English Kings, edd. & transl. R.A.B. Mynors et al. (2 vols, Oxford 1998/9), I.738–9(V.409); Expugnatio Hibernica, edd. & transl. Scott & Martin, pp. 68–71 (I.18). Cf. L. Hays &E.D. Jones, ‘Policy on the run: Henry II and Irish Sea diplomacy’, Journal of British Studies 29 (1990)293–316, at pp. 300–1.

101F.X. Martin, ‘Diarmait Mac Murchada and the coming of the Anglo-Normans’, in A NewHistory of Ireland, II, Medieval Ireland, 1169–1534, ed. Art Cosgrove (Oxford 1987), pp. 43–66, atp. 62. On the coronation of Ruaidrí Ua Conchobair in Dublin in 1166, see above, n. 55.


Dublin and the focus of Henry’s support during the English civil war. Diarmaitwas also friendly with Robert fitz Harding, a wealthy merchant of that town whoseems to have financed Henry during the darkest days of that conflict.102 Roberthad acquired lands near Bristol from Henry’s uncle, Robert, earl of Gloucester.The earl was a major supporter of Henry’s cause from 1139 until his death in1147.103

This network of connexions was demonstrated in 1141 when King Stephenwas held captive by Mabel, wife of Robert earl of Gloucester. At the same time,the earl was in the hands of Stephen’s supporters. While Stephen’s allies threatenedto send Robert to Boulogne (Stephen had married the countess of Boulogne),Mabel threatened to send Stephen to Ireland.104 It may be guessed that Diarmaitwould have been expected to provide suitable accommodation.

Good relations between Henry and Dublin may have soured in 1154 whenthe Irish overking Muirchertach Mac Lochlainn imposed his authority on theport.105 Muirchertach’s determination to have a say in Dublin’s affairs is suggestedby his payment of 1,200 cows to the citizens as a stipend to secure their allegiance.This threatened Henry’s policy of stabilising relations with Dublin throughcommon interests with Diarmait. It was in the following year that the council ofWinchester was held. According to Robert of Torigny, its purpose was to discussplans for an invasion of Ireland, but because the idea ‘was not pleasing to theEmpress, Henry’s mother, the expedition was put off to another time’.106 It seemsthat the papal bull Laudabiliter was issued shortly after the Council; that documentgave some support for a projected invasion of Ireland.107 These events may be

102R.B. Patterson, ‘Robert fitz Harding of Bristol: profile of an early Angevin burgess-baronpatrician and his family’s urban involvement’, The Haskins Society Journal 1 (1989) 109–22; Hayes& Jones, ‘Policy’, p. 294.

103Chris Given-Wilson & A. Curteis, The Royal Bastards of Medieval England (London 1984),pp. 74–93.

104William of Malmesbury, Historia Novella, The Contemporary History, edd. & transl. EdmundKing & K.R. Potter (Oxford 1998), pp. 118/19 (III.65).

105Annala, ed. & transl. O’Donovan, II.1112/13 (1154.11); The Annals of Clonmacnoise, ed.Murphy, p. 205 (1161.2).

106Chronicles of the Reigns of Stephen, Henry II., and Richard I., ed. Richard Howlett (4 vols,London 1884–9), IV.186 (Chronica, s.a. 1155.19).

107Pontificia Hibernica. Medieval Papal Chancery Documents concerning Ireland, 640–1261, ed.Maurice P. Sheehy (2 vols, Dublin 1962/5), I.15–16 (no. 4). Ioannis Saresberiensis episcopi CarnotensisMetalogicon Libri IIII, ed. Clement C.J. Webb (Oxford 1929), pp. 217–18 (IV.42); The Metalogiconof John of Salisbury. A Twelfth-century Defense of the Verbal and Logical Arts of the Trivium, transl.Daniel D. McGarry (Berkeley, CA 1955), pp. 274–5. For discussion, see Sheehy, When the Normanscame, pp. 13–15; Kostick, ‘Laudabiliter’; M. Haren, ‘Laudabiliter: text and context’, in Charters andCharter Scholarship in Britain and Ireland, edd. Marie Therese Flanagan & J.A. Green (Basingstoke2005), pp. 140–63.


interpreted as a direct response to the subjection of Dublin to the Irish overkingMuirchertach Mac Lochlainn in 1154. I suggest that Muirchertach’s action wasmore significant than the ecclesiastical independence of Dublin from Canterburyin determining Angevin royal policy. While it is possible that Henry would haveinvaded Ireland in 1155 had not his mother intervened, I think it more likely thathe was using the papal bull and the council as threats to induce Dublin not to actagainst his interests in Britain.

THE END OF SCANDINAVIAN DUBLIN

Taking these events into account, we may understand why Henry II resumed goodrelations with Diarmait Mac Murchada once Dublin was firmly under the latter’scontrol, after 1162. In 1165, as we have seen, Henry hired a Dublin-fleet to attackWales. But in 1166 Ruaidrí Ua Conchobair seized Dublin and had himselfcrowned there. In the immediate aftermath, it was to Henry that Diarmait wentto seek help following his expulsion from Ireland.108 Diarmait’s expulsion anduncertainty over the fate of Dublin may have led Henry to invade Ireland in 1171.It seems that his first concern was to bring the major ports under Angevin control,and Dublin was swiftly annexed.109 Trading privileges in the port were thengranted to the loyal citizens of Bristol.110

But what was the fate of the inhabitants of Dublin after 1170? Initially itappears that Ostmen received preferential treatment when compared with theirIrish neighbours.111 They were placed under the special protection of the Crownand Hamond mac Torcaill, who must have been a close relation of the last Ostmanking and who had his lands at Kinsealy near Portmarnock confirmed in 1174.112

These concessions were probably inspired by economic and political expediencyrather than by Henry’s magnanimity. Over time, the Ostmen found themselves

108The Annals of Inisfallen, ed. & transl. Mac Airt, pp. 300/1 (1166.7). Annales Cambriæ, ed.Williams, p. 51 (texts BC, annal 1167.3): ‘Diermit filius Murcath Hiberniam a suis expulsus regemAnglie addit’ (B); ‘Dyermict rex filius Marchat adiit regem Anglie querens auxilium contra inimicossuos’ (C). The versions of Brenhinoedd y Saeson are more specific: for text P, see Brut y TywysogyonPeniarth MS. 20, ed. Jones, p. 113, and Brut ... Chronicle ... Peniarth, transl. Jones, p. 64(1165[=1166].3); for text R, see Brut ... Red Book, ed. & transl. Jones, pp. 148/9 (1164[=1166].3);for text S, see Brenhinedd, ed. & transl. Jones, pp. 168/9 (1165[=1166].3). Cf. Flanagan, IrishSociety, pp. 56–76; Hays & Jones, ‘Policy’, p. 295.

109The Annals of Inisfallen, ed. & transl. Mac Airt, pp. 302–5 (1170.2–4, 1171.5); ExpugnatioHibernica, edd. & transl. Scott & Martin, pp. 88/9 (I.28). Cf. Duffy, ‘Ireland’s Hastings’, p. 80.

110Calendar of Ancient Records of Dublin, edd. John T. Gilbert et al. (19 vols, Dublin 1899–1944), I.1.

111E. O’Byrne, ‘Cultures in contact in the Leinster and Dublin marches, 1170–1400’, MedievalDublin 5 (2003) 111–48, at p. 115.

112O’Byrne, War, p. 25. Cf. Duffy, ‘The 1169 invasion’, pp. 104–8.


excluded from gilds and burgess-rights. Although they were entitled to Englishlaw, they found it increasingly difficult to get authorities to recognise them asbeing anything other than Irish. Edward I received a petition to this effect in thelate thirteenth century.113 It seems that the Ostmen came to inhabit an ethnicno-man’s land, a harsh fate in a culture where identity and ethnicity were generallysynonymous.114 The cession of the Hebrides from Norwegian to Scottish controlin 1266 gradually eroded any trading advantage to be had from maintainingelements of Scandinavian identity or language.115 It is around the end of thirteenthcentury that the Ostmen cease to be identifiable as an ethnic category in Ireland.Their descendants assumed either an Irish or an English persona.116

In conclusion it can be said that the inhabitants of Dublin in the twelfthcentury had a liminal or hybrid identity. They were recognised as being neitherfully Irish nor fully Scandinavian. However, it would be wrong to see them asvictims assigned a marginal identity by their neighbours. Rather they consciouslyplayed their frontier-role to some advantage. Particularly in the sphere of trade,the cosmopolitan identity of the Dubliners made them ideal go-betweens. It alsosuited their role as mercenaries in Irish-Sea politics. By maintaining a separateidentity, the people of Dublin also sought to retain their political independencefrom Irish overkings. Only after the Angevin conquest of Dublin, and the steady erosion of the Ostmen’s rights, was this distinctive Hiberno-Scandinavianidentity dissolved.117

113Calendar of Documents relating to Ireland, preserved in Her Majesty’s Public Record Office,London, 1171–1307, edd. H.S. Sweetman & G.F. Handcock (5 vols, London 1875–86), III.305–20 (no. 622: 23 April, 1290), at p. 306 (petition of Maurice MacOtere); Curtis, ‘The English andOstmen’, pp. 213–15.

114Davies, ‘The peoples of Britain and Ireland, 1100–1400, I’, pp. 17–18.115A.A.M. Duncan, Scotland. The Making of the Kingdom (Edinburgh 1975; rev. imp., 1978),

pp. 577–82.116Curtis, ‘The English and Ostmen’, p. 219.117This chapter had its first outing at the conference ‘The March in the Medieval West’, held at

University College, Dublin, 15–17 September, 2005, at the invitation of the organisers, EmmettO’Byrne and Edel Bhreathnach. An abridged version was presented at the conference ‘Celtic-NorseRelationships in the Irish Sea in the Middle Ages’, held in the University of Oslo, 5 November,2005, for which invitation I owe thanks to Jón Viðar Sigurðsson and Gro Steinsland. When thischapter was originally published, its documentation had to be severely curtailed for reasons of space.I have taken the present opportunity of republication to reïnstate that documentation in footnotes,as well as to add a few references to more recent work.


The lineage of Torcall, attested 1124–71 in Irish chronicles

Torcallfl. 1133ALC

Torfinnob. 1124

(immatura aetate)AU

Ragnallk. Dublinob. 1146

AFM, AT, CS, McC

Brodark. Dublinob. 1160

AB, AFM, AT, AU

mac meicThorcaillob. 1138AFM

Ascallk. Dublinfl. 1170AFMob. 1171

AFM, ALC,AT, AU, McC

Ragnallfl. 1167AFM


V

MYSTERIESOF

THE IRISH-SEA ZONE

9Eric Bloodaxe – axed?

The mystery of the last Scandinavian king of York

THE identity of Eiríkr of York was a matter of debate among historians in thenineteenth century. Mediaeval English sources were rather vague on this issue,although they did report that he was a son of someone called Haraldr.1 Thisenabled two rival theories of Eiríkr’s identity to be developed. One view –espoused by Sharon Turner, J.M. Lappenberg, Charles Plummer, and J.H. Todd– was that Eiríkr was the same person as Hringr son of Haraldr blátönn (HaroldBluetooth) of Denmark.2 The other opinion – advanced by Charles Haliday,Joseph Stevenson, and W.G. Collingwood – was that Eiríkr of York was Eiríkrblóðøx (Eric Bloodaxe), erstwhile king of Norway and a son of Haraldr hárfagri(Harold Finehair).3 Collingwood persuasively argued this view in 1900; sincethen, Eiríkr of York has been regarded as synonymous with Eiríkr blóðøx.

Collingwood’s case was developed within the context of late Victorian methodsof analysing saga as historical evidence. There is also a certain imaginative appeal,in the epithet ‘Bloodaxe’ and in the portrayal of this figure in saga, which hasencouraged acceptance of this identification.4 It is, however, striking that theearliest source used to identify Eiríkr of York with Eiríkr blóðøx belongs to theend of the twelfth century. As this seems rather late, and as the evidence itselfmay be challenged, I propose to review the sources for Eiríkr’s identity, guided by

Mediaeval Scandinavia 14 (2004) ISSN 0076-5864

1The Anglo-Saxon Chronicle. A Collaborative Edition, gen. edd. David Dumville & S. Keynes,VII, MS. E, ed. Susan Irvine (Cambridge 2004), p. 55 (s.a. 952): ‘Her Norðhymbre fordrifan Anlafcyning 7 underfengon Yric Haroldes sunu’; cf. ibid., VIII, MS F, ed. Peter S. Baker (Cambridge2000), p. 81 (s.a. 952); The Anglo-Saxon Chronicle. A Revised Translation, transl. Dorothy Whitelocket al. (London 1961; rev. imp., 1965), p. 73; Henry, Archdeacon of Huntingdon, Historia Anglorum,The History of the English People, ed. & transl. Diana Greenway (Oxford 1996), pp. 316/17 (V.22).

2J.M. Lappenberg, A History of England under the Anglo-Saxon Kings (2 vols, London 1845),II.124: Lappenberg cited the opinion of Sharon Turner on this matter. Two of the Saxon ChroniclesParallel, ed. Charles Plummer (2 vols, Oxford 1892/9), II.148. Cogadh Gaedhel re Gallaibh, TheWar of the Gaedhil with the Gaill, or, The Invasions of Ireland by the Danes and Other Norsemen, ed.& transl. James Henthorn Todd (London 1867), p. 267.

3Charles Haliday, The Scandinavian Kingdom of Dublin (Dublin 1881), pp. 68, 96; The ChurchHistorians of England, transl. Joseph Stevenson (5 vols in 9, London 1854–8), III.1, p. 128, n. 1;W.G. Collingwood, ‘King Eirík of York’, Saga-book 2 (1897–1900) 313–27.

4Alfred P. Smyth, Scandinavian York and Dublin. The History and Archaeology of Two relatedViking Kingdoms (2 vols, Dublin 1975/9), II.155–90.

181

more recent developments in source-criticism.5 For the purpose of this chapter Ishall adopt the guise of Devil’s advocate and suggest the existence of anotherpossible identity for Eiríkr of York, which has never yet been considered. I hopethat this will provoke further debate either to disprove or to support this theory,and to promote further analysis of the evidence from England, Germany,Scandinavia, and Ireland on this issue.

In making a case against the identification of Eiríkr of York with Eiríkr blóðøx,I shall first discuss the English evidence for Eiríkr’s reign. This includes chronicleswhose authors drew on tenth-century annals, namely ‘The Anglo-SaxonChronicle’, Historia regum Anglorum (Part I, section 6), and Roger of Wendover’sFlores historiarum (although in the last of these the tenth-century evidence is muchsupplemented).6 There are also twelfth-century historians who referred to Eiríkr,including Henry of Huntingdon, John of Worcester, William of Malmesbury, andRoger of Howden.7 None of these identified Eiríkr of York as Eiríkr blóðøx ormentioned that Eiríkr of York had been king of Norway. Nor do later sourcesfrom England acknowledge this connexion. For example, when in the latethirteenth century the spin-doctors of King Edward I were recasting history tomake a case for his claim to rule Scotland, Eiríkr was cited as an example of a

5R.W. McTurk, ‘Ragnarr Loðbrók in the Irish annals?’, in Proceedings of the Seventh VikingCongress, Dublin, 15–21 August 1973, edd. Bo Almqvist & D. Greene (Dublin 1976), pp. 93–123;P.H. Sawyer, ‘Harold Finehair and the British Isles’, in Les Vikings et leur civilisation. Problèmesactuels, ed. Régis Boyer (Paris 1976), pp. 105–9; D. Ó Corráin, ‘High-kings, vikings and otherkings’, Irish Historical Studies 21 (1978/9) 283–323; J. Lindow, ‘Íslendingabók and myth’,Scandinavian Studies 69 (1997) 454–64.

6Two of the Saxon Chronicles, ed. Plummer, II.113; The Anglo-Saxon Chronicle. A CollaborativeEdition, gen. edd. Dumville & Keynes, VI, MS D, ed. G.P. Cubbin (Cambridge 1996), pp. 44–5;The Anglo-Saxon Chronicle, transl. Whitelock et al., pp. 72–3, Flores historiarum, s.aa. 947, 948,950, in Rogeri de Wendover Chronica, sive Flores Historiarum, ed. Henry O. Coxe (5 vols, London1841–4), I.402, and Roger of Wendover’s Flowers of History, comprising the History of England fromthe Descent of the Saxons to A.D. 1235, formerly ascribed to Matthew Paris, transl. J.A. Giles (2 vols,London 1849), I.255–6, Historia regum Anglorum, Part I, s.a. 948, in Symeonis Monachi OperaOmnia, ed. Thomas Arnold (2 vols, London 1882/5), II.94; The Church Historians of England,transl. Stevenson, III.2, p. 483.

7Chronicon, s.aa. 949, 950, in The Chronicle of John of Worcester, edd. & transl. R.R. Darlingtonet al., II (Oxford 1995), pp. 400/1; William of Malmesbury, Gesta Regum Anglorum, The History ofthe English Kings, edd. & transl. R.A.B. Mynors et al. (2 vols, Oxford 1998/9), I.236/7 (II.146);Historia Anglorum, V.22, in Henry Archdeacon, ed. & transl. Greenway, pp. 316/17; ChronicaMagistri Rogeri de Houedene, ed. William Stubbs (4 vols, London 1868–71), I.56–7 (s.aa. 947, 948,952), and The Annals of Roger de Hoveden comprising the History of England and of Other Countriesof Europe from A.D. 732 to A.D. 1201, transl. Henry T. Riley (2 vols, London 1853), I.66–7;Collingwood, ‘King Eirík’, p. 320.


Scottish king who had been installed by a royal English overlord.8 The absencefrom English sources of any mention of Eiríkr’s previous rule over Norway isperhaps surprising, if it had been historical.

The most reliable clue to Eiríkr’s identification in English sources is that hewas a son of Haraldr. This is stated in the E-text of ‘The Anglo-Saxon Chronicle’and was repeated by Henry of Huntingdon.9 The early thirteenth-century authorRoger of Wendover supplied further details of family-connexions. He stated thatEiríkr died with his son Henry (filio suo Henrico) and with his brother Reginald(fratre Reginaldo).10 It is possible that these details are a later inventionincorporated into the text and are not historically accurate. Here it is enough tonote that they do not match Scandinavian accounts of Eiríkr blóðøx: he was saidto have had six sons whose names are listed in Norwegian sources, but none ofthese names is in any way equivalent to the name Henry.11 However, Collingwoodattempted to identify these names, Henry with that of Hárekr son of Guthormrand Reginald with that of an unidentified King Røgnvaldr who are listed amongthose killed in the final battle of Eiríkr blóðøx in the thirteenth-centurysaga-compendium known as Heimskringla.12 This is a rather tenuous similaritywhich could be coïncidental. Even if it is not coïncidental, we must note thatAlan Binns, Axel Seeberg, and others have suggested that the author ofHeimskringla drew these names from an English source.13 This could thereforebe a late addition to the legend of Eiríkr blóðøx in Heimskringla, influenced byEnglish accounts of the Northumbrian king. These reports – in Heimskringla andin Roger’s Flores historiarum – seem too imprecise and too doubtful to show thatEiríkr blóðøx was Eiríkr of York.

8Ibid., p. 325; Anglo-Scottish Relations, 1174–1328. Some Selected Documents, ed. & transl.E.L.G. Stones (2nd edn, Oxford 1970), pp. 198/9 (no. 30). This is found in a letter of 1301, fromKing Edward I to Pope Boniface VIII: quodam Yricio rege super ipsos Scotos statuto (‘a certain Ericinstalled as king over the Scots’).

9The Anglo-Saxon Chronicle. A Collaborative Edition, gen. edd. Dumville & Keynes, VII, MS.E, ed. Susan Irvine, p. 55 (s.a. 952); The Anglo-Saxon Chronicle, transl. Whitelock et al., p. 73; Henryof Huntingdon, Historia Anglorum, V.22 (Henry Archdeacon, ed. & transl. Greenway, pp. 316/17).

10Flores historiarum, s.a. 950: Rogeri de Wendover Chronica, ed. Coxe, I.402; Roger of Wendover’sFlowers, transl. Giles, I.256.

11Ágrip af Nóregskonungasögum. A Twelfth-century Synoptic History of the Kings of Norway, ed.& transl. M.J. Driscoll (London 1995), pp. 8/9 (§5); Historia Norwegie, edd. & transl. Inger Ekremet al. (København 2003), pp. 80–3 (§XI).

12Hákonar saga góða, §4, in Snorri Sturluson, Heimskringla: Nóregs Konunga sögur, ed. FinnurJónsson (København 1911), pp. 72–3; Heimskringla, History of the Kings of Norway, by SnorriSturluson, transl. Lee M. Hollander (Austin, TX 1964), p. 99; Collingwood, ‘King Eirík’, p. 314.

13A. Binns, ‘The York viking kingdom: relations between Old English and Old Norse culture’,in The Fourth Viking Congress, York, August 1961, ed. Alan Small (Edinburgh 1965), pp. 179–89,at p. 188; A. Seeberg, ‘Five kings’, Saga-book 20 (1978–81) 106–13, at p. 107, n. 4. Seeberg citedwork by Ove Moberg, Jan de Vries, and Bjarni Einarsson in support of this idea.

ERIC BOODAXE – AXED? 183

A king called Eiríkr is recorded in the Anglo-Norman Liber uitae of Durhamas rex Danorum.14 In some Scandinavian sources Eiríkr blóðøx is presented as aconvert to Christianity, and this has been regarded as a reference to the sameking.15 Although it is possible that Eiríkr of York was christian, Lesley Abrams hasrecently identified the king in the Liber uitae as Eric Eiegod, who ruled Denmarkfrom 1095 to 1103. This is because his wife, Botild regina, is mentioned alongsidehim.16 It seems therefore that English sources provide little guidance concerningthe identification of Eiríkr of York. They provide no proof that he can beidentified with Eiríkr blóðøx.

Evidence from Germany has been used to support the idea that Eiríkr of Yorkwas the son of Haraldr blátönn, king of Denmark, rather than a Norwegian king.The earliest surviving source for this idea is Gesta Hammaburgensis ecclesiaepontificum by Adam of Bremen. This was originally composed around 1075,although some notes seem to have been added soon after.17 Adam reported thatHaraldr blátönn ‘sent his son Hringr to England with an army. When the latterhad subjugated the whole island, he was in the end betrayed and killed by theNorthumbrians’ (Hring filium cum exercitu misit in Angliam, qui subacta insulatandem proditus et occisus est a Nordhumbris).18 We know that Adam used Danishinformants as a source when writing his Gesta ~ , including the Danish king, SvenEstrithsen.19 Thus we may have the evidence of a Danish witness in this passagewhich eulogises the deeds of Haraldr blátönn and his successors of the Danishroyal line. Adam’s record of Hringr was quoted in later collections of Icelandicliterature, namely Flateyjarbók and Fornmannasögur.20

14Liber Vitæ Ecclesiæ Dunelmensis; nec non Obituaria Duo ejusdem Ecclesiæ, ed. [Joseph] Stevenson(Durham 1841), p. 78; Liber Vitae Ecclesiae Dunelmensis, facs. ed. Alexander Hamilton Thompson(Durham 1923), folio 51v.

15E.V. Gordon, An Introduction to Old Norse (2nd edn, rev. A.R. Taylor, Oxford 1957), p. 245;Smyth, Scandinavian York, II.180.

16L. Abrams, ‘The conversion of the Danelaw’, in Vikings and the Danelaw. Select Papers fromthe Proceedings of the Thirteenth Viking Congress, Nottingham and York, 21–30 August 1997, edd.James Graham-Campbell et al. (Oxford 2001), pp. 31–44, at p. 41(–2), n. 16.

17Martin Syrett, Scandinavian History in the Viking Age. A Select Bibliography (Cambridge 2001),p. 11.

18Gesta Hammaburgensis ecclesiae pontificum, II.25 (22): Magistri Adami Bremensis GestaHammaburgensis Ecclesiae Pontificum, ed. Bernhard Schmeidler (Hannover 1917), p. 84; Adam ofBremen, History of the Archbishops of Hamburg-Bremen, transl. Francis J. Tschan (New York 1959),p. 71.

19Syrett, Scandinavian History, p. 11; Adam, transl. Tschan, p. xx. Cf. T. Bolton, ‘Englsig politicalrefugees at the court of Sveinn Ástríðarson, king of Denmark (1042~76)’, Mediaeval Scandinavia15 (2005) 17~36.

20Fornmanna sögur, edd. Sveinbjörn Egilsson et al. (12 vols, København 1825–37), XI.418;Flateyjarbók. En Samlinge af Norske Konge-sagaer, edd. Guðbrand Vigfússon & C.R. Unger (3 vols,Oslo 1860–8), I.17; Collingwood, ‘King Eirík’, p. 316.


There are reasons to doubt that this record of Hringr is an accurate referenceto Eiríkr of York. First, the name, Hringr, is manifestly not Eiríkr. J.R. Greenand Charles Plummer resorted to calling Eiríkr of York ‘Eric Hiring’, but there isno evidence that the names should be connected.21 The second problem is withAdam’s chronology. According to Adam, this event happened when Hákon theGreat was ruling Norway (ca 971×995), which is too late to fit Eiríkr’s career inNorthumbria.22 Furthermore, Adam asserted that Hringr subjugated all Britain,which is manifestly not true. Even if this is an exaggerated account of the seizureof Northumbria, it does not fit with other sources. Adam of Bremen was perhapsrelying on oral witnesses who gave an inaccurate account exalting the deeds ofHaraldr blátönn and his family. The account seems very unreliable, even thoughit was written barely a century after the events were said to have happened. It ispossible that the story of Hringr developed from the name of Culén Hringr, kingof Alba, who died in 971. This fits the time-frame and the geographical contextsuggested by Adam. However, no one has argued that Culén was a son of Haraldrblátönn.23 It is possible that Adam’s story developed from a desire to glorify theroyal line of Denmark, and it can be argued that a similar motive comes into playin the saga-portrayals of Eiríkr blóðøx.

Scandinavian histories and sagas, dating from the end of the twelfth centuryand later, assert a connexion between the Northumbrian Eiríkr and Eiríkr blóðøx.It is clear from these texts that several different stories circulated about the fate ofEiríkr blóðøx once he left Norway (these may be compared with thesurvival-legends of other mediaeval kings).24 It is my hypothesis that, in theScandinavian sources, legends concerning Eiríkr blóðøx became influenced by,and were combined with, the legends of Eiríkr of York. This may simply resultfrom confusion of two kings of the same name. Alternatively, such a conflationcould have been motivated by a narrative desire to fill the gap in the story of the

21John Richard Green, The Conquest of England (3rd edn, 2 vols, London 1899), I.319; Two ofthe Saxon Chronicles, ed. Plummer, II.148.

22Adam, transl. Tschan, p. 70, n. 84.23B.T. Hudson (ed. & transl.), ‘The Scottish Chronicle’, Scottish Historical Review 77 (1998)

129–61, at p. 151 and n. 34; D.N. Dumville, ‘The Chronicle of the Kings of Alba’, in Kings, Clericsand Chronicles in Scotland, 500–1297. Essays in Honour of Marjorie Ogilvie Anderson on the Occasionof her Ninetieth Birthday, ed. Simon Taylor (Dublin 2000), pp. 73–86, at p. 81. A Hringr is alsonamed in Egils saga as a ruler of part of Britain – see Egils saga Skalla-Grímssonar, ed. Sigurður Nordal(Reykjavík 1933), pp. 129–30, 136–8, 142 (§§51, 53,55 [stanza 18]); ‘Egils Saga’, transl. BernardScudder, in The Sagas of the Icelanders, gen. ed. Örnólfur Thorsson (5 vols, Reykjavík 1997), I.33–177, at pp. 93, 96–7, 99).

24M. Cohen, ‘From Throndheim to Waltham to Chester: Viking- and post-Viking-Age attitudesin the survival legends of Óláfr Tryggvason and Harold Godwinson’, in The Middle Ages in theNorth-west, edd. Tom Scott & P. Starkey (Oxford 1995), pp. 143–54; A. Thacker, ‘The cult of KingHarold at Chester’, ibid., pp. 155–76.


life of Eiríkr blóðøx once he had left Scandinavia. The association of Eiríkr ofYork with Eiríkr blóðøx also fits with the legends evolving around the persona ofHaraldr hárfagri (father of Eiríkr blóðøx) and the royal line of Norway at thattime.25

The Scandinavian evidence is best discussed in two parts. I shall treat the prosetexts separately from the skaldic poems embedded in them. This is becausescholars have often assigned the poems a separate genesis from and earlier datethan the surrounding prose. For the purposes of historical analysis the prose maytherefore sometimes cloud rather than clarify the poems’ interpretation. TheNorwegian Latin Historia de antiquitate regum Norwagiensium of Theodoric theMonk appears to be the earliest source linking Eiríkr blóðøx with England. Thistext was composed in 1177×1188.26 Theodoric stated that his account was basedon the oral testimony of Icelanders.27 He made the apology that, although theIcelanders were well versed in historical lore, such oral evidence is weaker thanwritten testimony. In his Historia, Theodoric reported that Eiríkr, whom he calledfratrum interfector (‘brother-killer’), succeeded Haraldr hárfagri on the Norwegianthrone. His brother drove him out after three years on account of his cruelty andparticularly that of his wife Gunnhildr. Eiríkr then sailed to England, where theking received him honourably, and he died immediately afterwards.28 No referenceis found to Eiríkr having ruled Northumbria.

The link between Eiríkr blóðøx and York is first attested in the slightly laterhistory known as Ágrip. This has been dated 1188×1200.29 According to Ágrip,when Eiríkr fled Norway and came to England he was granted the earldom ofNorthumbria by King Æthelstan (927–39). However, Eiríkr soon lost hold ofNorthumbria (again this is blamed on Gunnhildr). Eiríkr finally resorted to apirate’s life and died while fighting in Spain.30 The origins of the other ‘synoptic’history of Norway, Historia Norwegie, are open to question. Although its exactdate has been much disputed, it has until recently seemed to be a later text thanHistoria de antiquitate and Ágrip. Now, however, we are offered a dating in thethird quarter of the twelfth century.31 Historia Norwegie agrees with Ágrip that

25This is discussed below, p. 193~4.26The opinion of Peter Foote, in Theodoricus Monachus, Historia de Antiquitate Regum

Norwagiensium, An Account of the Ancient History of the Norwegian Kings, transl. David McDougallet al. (London 1998), p. xii.

27Theodoricus, Historia, preface: Monumenta Historicae Norwegiae latine conscripta. LatinskeKildeskrifter til Norges Historie i Middelalderen, ed. Gustav Storm (Oslo 1888), p. 1; Theodoricus,transl. McDougall et al., p. 1.

28Theodoricus, Historia, §2 (Monumenta, ed. Storm, p. 7; Theodoricus, transl. McDougall et al.,p. 6).

29Ágrip, ed. & transl. Driscoll, p. xii.30Ibid., p. 17.31Ibid., p. xiv; cf. now Historia Norwegie, edd. & transl. Ekrem et al., pp. 11–24.


Eiríkr died in Spain. But Historia Norwegie also states that Eiríkr was baptised in Northumbria.32

These accounts diverge in detail. They also contradict English sources aboutEiríkr of York. The reports of Eiríkr’s violent death in England – given in Historiaregum Anglorum, Part II, and by Roger of Wendover – do not match any of thethree Norwegian ‘synoptic’ histories mentioned above. Theodoric implied thatEiríkr died peacefully in England, while Ágrip and Historia Norwegie record hisdeath in Spain. Finnur Jónsson suggested that there was confusion in theNorwegian sources which turned Stan from Stainmore, where Eiríkr diedaccording to Roger of Wendover, to Span, Spain.33 I do not find this convincing.Moreover, none of the Scandinavian accounts names Stainmore.34 In the Life ofSt Cathroe, written at Metz in the late tenth century, Eiríkr had (not later than946) a wife who was a member of the royal dynasty of Alba and/or Strathclyde.35

According to the Scandinavian accounts, Gunnhildr, who is named as wife ofEiríkr blóðøx throughout his time in England, was of Danish or Norwegiandescent.36 It seems that records of the details of Eiríkr’s life diverge between Insularand Scandinavian sources.

Another difficulty in reconciling Scandinavian and Insular evidence resides intheir chronology of events. Scandinavian sources relate that Eiríkr blóðøx wasking of Norway after Haraldr hárfagri. Haraldr is said to have begun his reign in

32Ibid., pp. 80–3 (§XII).33Finnur Jónsson, Den Oldnorske og Oldislandske Litteraturs Historie (2nd edn, 3 vols, København

1920–4), II.614, n. 2.34Hákonar saga góða, §4: Snorri, ed. Finnur Jónsson, pp. 72–3; Heimskringla, transl. Hollander,

p. 99. The text refers to an inland location in England. Collingwood made much of this in relationto Eiríkr’s identity: W.G. Collingwood, ‘The battle of Stainmoor in legend and history’, Transactionsof The Cumberland and Westmorland Antiquarian and Archaeological Society, new series, 2 (1902)231–41. As I mentioned previously (see p. 183, above), the author of Heimskringla may have hadaccess to an English source. This obviates Collingwood’s idea that Heimskringla reports anindependent Scandinavian tradition concerning the location of Eiríkr’s demise. Furthermore, thisinland spot could have been anywhere in England away from the sea. The description fits thecomment in Heimskringla concerning Eiríkr’s bravery, that he was willing to raid far inland. Thebattle-location may therefore be a literary device rather than a historical location.

35Vita Sancti Kaddroe, §17: Acta Sanctorum Veteris et Maioris Scotiæ, seu Hiberniæ, SanctorumInsulæ, ed. Ioannes Colganus (Leuven 1645; 2nd edn, by B. Jennings, Dublin 1948), p. 497; EarlySources of Scottish History, A.D. 500 to 1286, transl. Alan Orr Anderson (2 vols, Edinburgh 1922; rev.imp., Stamford 1990), I.441. Cf. n. 130, below. For the authorship and circumstances of compositionof this Life, see David N. Dumville, ‘St Cathróe of Metz and the hagiography of exoticism’, in Studiesin Irish Hagiography: Saints and Scholars, edd. John Carey et al. (Dublin 2001), pp. 172–88, revisedin his Celtic Essays, 2001–2007 (2 vols, Aberdeen 2007), I.158–82; C. Downham, ‘The chronologyof the last Scandinavian kings of York’, Northern History 40 (2003) 25–51.

36Ágrip, §5 (ed. & transl. Driscoll, pp. 8/9 and 87–8 [n. 15]); Historia Norwegie, §XII (edd. &transl. Ekrem et al., pp. 80–3).


858 or 863, and he is reported to have held power for as long as seventy orseventy-three years.37 On this basis Haraldr completed his reign in 928×936.39

Eiríkr blóðøx then ruled for a period of one to three years independently of hisfather.39 This leaves a significant gap in time before Eiríkr is first recorded inNorthumbria in the English sources. The gap is not satisfactorily filled inScandinavian accounts. Ágrip, Historia Norwegie, and Theodoric’s Historia allstate that Eiríkr went directly to England.40 There are, however, two reports inÁgrip: one says that he went to Denmark first.41 In Heimskringla and Egils sagawe read that Eiríkr blóðøx stopped at Orkney first, but this seems to have beenfor a short amount of time before he went raiding in North Britain and England.42

This sequence conflicts with Orkneyinga saga which states that Eiríkr came toOrkney after he had been expelled from Northumbria.43 The Scandinavianchronology does not seem to fit with Insular sources, and the evidence is somewhatcontradictory.

Two Scandinavian sources, Ágrip and Historia Norwegie, link Eiríkr to KingÆthelstan, an unlikely relationship in terms of the English historical record.44 Theidea may be due to the fame of the brother of Eiríkr blóðøx, Hákon inn góði, whowas said to have been fostered by Æthelstan (I shall discuss this in more detailbelow45). Details of the life of Eiríkr blóðøx and his chronology in Scandinaviansources do not therefore match up with English accounts of Eiríkr of York. We

37Theodoricus, transl. McDougall et al., pp. 57–8 (n. 18). Íslendingabók and Ágrip have seventyyears. Historia Norwegie, Fagrskinna (Nóregs konungatal), and others have seventy-three years.

38Jesse L. Byock, Viking Age Iceland (Harmondsworth 2001), p. 83, has suggested that Harold’sreign ran from about 885 to 930.

39Theodoricus, transl. McDougall et al., p. 59 (n. 26).40Theodoricus, Historia, §2 (Monumenta, ed. Storm, p. 7; Theodoricus, transl. McDougall et al.,

p. 5); Historia Norwegie, §XII (edd. & transl. Ekrem et al., pp. 80–3); Ágrip, §7 (ed. & transl.Driscoll, pp. 16/17 and 88 [n. 18]).

41Ágrip, §5 (ed. & transl. Driscoll, pp. 8/9).42Hákonar saga góða, §4 (Snorri, ed. Finnur Jónsson, p. 71; Heimskringla, transl. Hollander,

pp. 98–9); Egils saga (ed. Nordal, p. 176, §59; transl. Scudder, pp. 115–16, §60).43Orkneyinga saga, §8, in Orkneyinga saga, Legenda de Sancto Magno, Magnúss saga skemmri,

Magnúss saga lengri, Helga þáttr ok Úlfs, ed. Finnbogi Guðmundsson (Reykjavík 1965), p. 18;Orkneyinga Saga, The History of the Earls of Orkney, transl. Hermann Pálsson & P. Edwards(Harmondsworth 1981), p. 32.

44Ágrip, §7 (ed. & transl. Driscoll, pp. 16/17); Historia Norwegie, §XII (edd. & transl. Ekremet al., pp. 80–3).

45R.I. Page, ‘The audience of Beowulf and the vikings’, in The Dating of Beowulf, ed. Colin Chase(Toronto 1981), pp. 113–22, at 114–15; L. Abrams, ‘The Anglo-Saxons and the christianization ofScandinavia’, Anglo-Saxon England 24 (1995) 213–50, at p. 217; G. Williams, ‘Hákon Aðalsteinsfóstri: aspects of Anglo-Saxon kingship in tenth-century Norway’, in The North Sea World in theMiddle Ages. Studies in the Cultural History of North-western Europe, edd. Thomas R. Liszka &L.E.M. Walker (Dublin 2001), pp. 108–26. See below, pp. 199–200.


might permit the Scandinavian sources some margin of error on the assumptionthat they depend on oral tradition, but their very imprecision calls other featuresof the story into doubt. Furthermore, their factual correctness may be called intoquestion.

The saga-accounts of Eiríkr which postdate the synoptic histories, includingEgils saga, Heimskingla, and Orkneyinga saga, provide further elaboration of thestory of Eiríkr blóðøx. Some of this seems to have originated in an earlier writtenversion, but the accounts are contradictory in nature, as Margaret Cormack haspointed out.46 It is necessary to employ a sceptical approach.47 Ideally, theIcelandic evidence should be analysed on a case-by-case basis and compared withother available evidence. More investigation of the Scandinavian accounts of thereign of Eiríkr blóðøx is therefore necessary. In terms of their reliability concerningthe association of Eiríkr blóðøx with Eiríkr of York, there are difficulties in trustingthe accurate transmission of information over time (from the tenth century, whenevents happened, to the late twelfth and thirteenth centuries when, on the evidenceof the dates of the available texts, they were committed to writing) and distance(how deeds in Northumbria were reported in Iceland). It is also necessary toanalyse this evidence (as with all written sources) for potential bias, ormisrepresentation of the past to suit contemporary political agenda or literarypredilection. On all three counts, I suggest that there is potential for error in thesaga-accounts of Eiríkr.

It seems that Icelandic lore knew of a Northumbrian ruler Eiríkr: he wasperhaps mistakenly associated with Eiríkr blóðøx, given that the Icelandic sourcesshow some uncertainty about that king’s fate once he left Norway. Eiríkr blóðøxaccording to some accounts died in England, while according to others he died inSpain. The reign of Eiríkr of York was also mistakenly associated with the time ofÆthelstan.48 Sagas, based on oral accounts over a century after Adam of Bremenwas writing, cannot be accepted as historical evidence without question. A caseof mistaken identity is possible. The frequency with which the names Haraldrand Eiríkr were used in Scandinavian ruling families makes it plausible that therewere two (or more) rulers with the same name, active in western Europe at muchthe same date.49

46M. Cormack, ‘Egils saga,Heimskringla, and the daughter of Eiríkr blóðøx’, Alvíssmál 10 (2001)61–8.

47J. Hines, ‘Egill’s Höfuðlausn in time and place’, Saga-book 24 (1994–7) 83–104, at p. 84.48Downham, ‘The chronology’; cf. A. Woolf, ‘Erik Bloodaxe revisited’, Northern History 34

(1998) 189–93, and Dumville, Celtic Essays, 2001–2007, I.166.49See, for example, the indices of Heimskringla, transl. Hollander, pp. 831 and 837. Eiríkr

blóðøx had a grandson called Eiríkr Haraldsson, which illustrates this point: Collingwood, ‘KingEirík’, p. 317.


The next type of Scandinavian evidence, though closely related, is skaldicpoetry. Some of this has been thought to have been composed in the tenth centuryand to have survived in the oral culture of the skalds, who passed down theirlearning from generation to generation. These poems were brought into writingin the thirteenth century.50 The skaldic poems connected with Eiríkr blóðøx orEiríkr of York are mainly preserved in Egils saga, Fagrskinna, and Heimskringla.51

It should be said at the outset that – once isolated from the untrustworthy prosesurroundings – none of these skaldic poems explicitly identifies Eiríkr blóðøx asEiríkr of York.

There are various opinions as to the historical validity of skaldic verse: this isa complex issue. Russell Poole has stated that it is ‘mostly authentic’, but at theother extreme Sigrún Daviðsdóttir has argued that there is ‘no reason ... to putmuch faith in the historical value of the court poetry’.52 Most scholars have beenwilling to admit some failings in the skaldic corpus as a historical witness, whilerecognising it as a valuable resource. Attention has been drawn to frequentdivergence in the record of the same poem in different manuscripts,53 unreliableauthorial attribution,54 deliberate archaisms in poetic language which inhibitlinguistic dating,55 and the observed plasticity of poetry composed in oral cultures,when studied over time.56

Specific criticisms have been made concerning the poems in Egils saga, whichis the main repository of verses relating to Eiríkr of York and Eiríkr blóðøx. Thesaga-writer asserted that they were tenth-century compositions of the poet Egill

50Sigrún Daviðsdóttir, ‘Old Norse court poetry – some notes on its purpose, transmission andhistorical value’, Gripla 3 (1979) 186–203, at p. 192.

51Egils saga, stanzas 29 and 31, 33, Höfuðlausn (twenty stanzas), Arinbjarnarkviða (twenty fivestanzas) /ed. Nordal, pp. 165, 170, 180, 185–92, 258–67 [§§57, 59, 60, 78]; transl. Scudder,pp. 110, 113, 118, 120-4, 157–62); Hákonar saga góða, §§19, 27, and Haralds saga gráfeldar, §§1,6 (Snorri, ed. Finnur Jónsson, pp. 82, 86, 94, 97; Heimskringla, transl. Hollander, pp. 113, 117,129, 135). These refer to Eiríkr blóðøx through the deeds of his sons. See also Corpus PoeticumBoreale. The Poetry of the Old Northern Tongue from the Earliest Times to the Thirteenth Century, edd.& transl. Gudbrand Vigfusson & F. York Powell (2 vols, Oxford 1883), I.259–61, 266–80, andII.31, 40; Chronicles of the Vikings. Records, Memorials and Myths, transl. R.I. Page (London 1995),pp. 110–11.

52Sigrún Daviðsdóttir, ‘Old Norse court poetry’, p. 202; Viking Poems on War and Peace. AStudy in Skaldic Narrative, ed. & transl. R.G. Poole (Toronto 1991), p. vii.

53R.G. Poole, ‘Variants and variability in the text of Egill’s Höfuðlausn’, in The Politics of editingMedieval Texts, ed. Roberta Frank (New York 1991), pp. 65–106; Sigrún Daviðsdóttir, ‘Old Norsecourt poetry’, p. 200; H. Kuhn, ‘Extended elements in the kenning’, Parergon 22 (1978) 13–22, atp. 13.

54Judith Jesch, Ships and Men in the Late Viking Age. The Vocabulary of Runic Inscriptions andSkaldic Verse (Woodbridge 2001), p. 15; Viking Poems, ed. & transl. Poole, p. vii.

55Sigrún Daviðsdóttir, ‘Old Norse court poetry’, p. 193; Jesch, Ships, p. 10.56Sigrún Daviðsdóttir, ‘Old Norse court poetry’, pp. 194–200; cf. Kuhn, ‘Extended elements’,


Skalla-grímsson. Russell Poole has demonstrated that verses (relating to the slayingof Ljótr) in §65 of the saga must be dated later than the tenth century on stylisticand linguistic grounds.57 Another verse, in §73, bears signs of alteration whencompared with independent and datable record elsewhere.58 It therefore seemslikely that verses in Egils saga had diverse origins. They may have been gatheredtogether (and some perhaps even composed) by the saga-writer to fit within thestructure of his prose creation.59

A general approach to verses in Egils saga has been to regard loose stanzas asless authentic but to deem the longer poems – Arinbjarnarkviða, Höfuðlausn, andSonnatorrek – to be a distinct group composed by one author, even if that authorcannot be identified as Egill.60 Two of these poems are particularly relevant tothis study, namely Arinbjarnarkviða and Höfuðlausn. These too have beencriticised. Carolyne Larrington has suggested that the tenses of verbal forms inArinbjarnarkviða may have been altered.61 More fundamentally, a tenth-centurydate for the composition of Höfuðlausn has been doubted on linguistic and stylisticgrounds.62 It is thus debated how reliable these poems are in respect oftenth-century political history where specific details (names, relationships, androyal deeds) are essential. These details may have fallen victim to elaboration,alteration, or ‘improvement’ over such a long span of time.

Despite the reservations expressed by various commentators, skaldic poetry isgenerally still deemed a significant source for early Scandinavian political history,albeit one where each example needs to be judged on its own merits. In relationto Eiríkr of York and Eiríkr blóðøx, there are four relevant poems in Egils saga.63

Poems which are linked to these characters by the prose context alone I have notincluded. An isolated stanza (no. 31) refers to a warlike son of Eiríkr blóðøx andGunnhildr; this may be associated with the poems about the sons of Eiríkr blóðøx

57Viking Poems, ed. & transl. Poole, pp. 178–9.58J.E. Knirk, ‘Runes from Trondheim and a stanza by Egill Skalla-Grímsson’, in Studien zum

Altgermanischen. Festschrift für Heinrich Beck, ed. Heiko Uecker (Berlin 1994), pp. 411–21;M. Clunies Ross, ‘The art of poetry and the figure of the poet in Egils saga’, Parergon 22 (1978) 3–12, at pp. 8–9, has suggested that other poems in Egils saga were reworked or augmented.

59Smyth, Scandinavian York, II.167.60C. Larrington, ‘Egill’s longer poems: Arinbjarnarkviða and Sonatorrek’, in Introductory Essays

on Egils Saga and Njáls Saga, edd. John Hines & D. Slay (London 1992), pp. 49–63, at p. 49.61Ibid., p. 50.62Collingwood, ‘King Eirík’, p. 141; D. Hofmann, ‘Das Reimwort giör in Egill Skallagrímssons

Höfuðlausn’, Mediaeval Scandinavia 6 (1973) 93–101; Poole, ‘Variants’, p. 3; Smyth, ScandinavianYork, II.166–7. Cf. Hines, ‘Egill’s Höfuðlausn’, pp. 84, 87.

63Egils saga, stanzas 11, 29, 31, 33 (ed. Nordal, pp. 113, 165, 170, 180 [§§45, 57, 59]; transl.Scudder, pp. 84, 110, 113, 118, [§§45, 58, 60]). These poems refer to Gunnhildr and obliquely toher husband Eiríkr blóðøx.


found in Heimskringla.64 Another loose stanza (no. 33) refers to a king of Haraldr’sline ruling in England.65 This is rather ambiguous: which Haraldr is not clear,and no Eiríkr is named. No reliable clue as to the stanza’s date of compositionhas yet been found.

Most poetic information about Eiríkr in Egils saga is found in Höfuðlausn andArinbjarnarkviða.66 These clearly relate to Eiríkr, king of York, with no apparentreference to Eiríkr blóðøx. Höfuðlausn is a eulogy of Eiríkr. According to theprose, the poem was composed under duress, a statement which negates thepoem’s praise-content to fit the bad image of Eiríkr in the saga as a savage rulerwith a sorceress for a wife.67 Alfred Smyth and John Hines have suggested thatthe saga may not present an accurate account of the circumstances in whichHöfuðlausn was composed.68 This poem describes Eiríkr as a scourge of the Scots,and his fame is said to have travelled to the eastern shore (presumably Norway).

The other lengthy poem, Arinbjarnarkviða, is said to have been recited by Egillfor a friend. It begins with an account of Egill’s escapades in England, and Eiríkris named as a member of the Yngling line. From this royal lineage were descendedthe kings of Norway. This genealogical statement has been used to support theidentification of Eiríkr of York with Eiríkr blóðøx.69 However, it was not onlykings of Norway who claimed descent from the Yngling dynasty. Certainly byabout 1030, Uí Ímair of Ireland claimed such a link, which we see in ‘TheFragmentary Annals of Ireland’.70 Members of this Hiberno-Scandinavian dynastyhad ruled in Northumbria before Eiríkr. The skaldic poems in Egils saga do nottherefore conflict with the theory that Eiríkr of York and Eiríkr blóðøx wereseparate individuals who were confused in Scandinavian literature of the end ofthe twelfth century and later.

64Egils saga, §59(ed. Nordal, p. 170; transl. Scudder, pp. 113); Hákonar saga góða, §§19, 27,and Haralds saga gráfeldar, §§1, 6 (Snorri, ed. Finnur Jónsson, pp. 82, 86, 94, 97; Heimskringla,transl. Hollander, pp. 113, 117, 129, 135).

65Egils saga, §60(ed. Nordal, p. 180; transl. Scudder, pp. 118).66Egils saga, §??(ed. Nordal, pp. 185–92, 258–67 [§§60, 78]; transl. Scudder, pp. 120–4, 157–

62 [§§61, 80]).67G. Jones, ‘Egill Skallagrímsson in England’, Proceedings of The British Academy 37 (1952) 127–44.68Hines, ‘Egill’s Höfuðlausn’, p. 84. Cf. Smyth, Scandinavian York, II.166–8: at p. 168, he has

stated that ‘Egill may never have stood before Eiríkr at York, but medieval Icelanders believed thathe did’; he has also suggested that Egill’s adventures at York and their associated poetry were laterinventions, thus denying any near-contemporary evidence in Scandinavian sources for the career ofEiríkr of York.

69P. Sawyer, ‘The last Scandinavian kings of York’, Northern History 31 (1995) 39–44, at p. 42.70Fragmentary Annals of Ireland, ed. & transl. Joan Newlon Radner (Dublin 1978), pp. 94–5,

144–5 (§§239, 400, and especially 401); J. Steffensen, ‘A fragment of viking history’, Saga-book 18(1970–3) 59–78. Cf. D. Ó Corráin, ‘The vikings in Scotland and Ireland in the ninth century’,Peritia 12 (1998) 296–339, at pp. 297–9. On the date of the relevant source of ‘The FragmentaryAnnals of Ireland’, see above, chapter 6.


There are other skaldic poems relating to Eiríkr blóðøx and Eiríkr of York,which are not found in Egils saga. There are stanzas in Heimskringla which referobliquely to Eiríkr blóðøx, through the deeds of his sons.71 A fragmentary poemattributed to Glúmr Geirason in praise of an Eiríkr also survives, but it provideslittle help to the historian.72 Finally, there is Eiríksmál whose language indicatesa Northumbrian origin.73 It is likely to refer to Eiríkr of York. The author of thislay is anonymous and the text seems (frustratingly) to be incomplete. It survivesonly in Fagrskinna, whose prose asserts that Gunnhildr had the poem composedsoon after the death of her husband, Eiríkr blóðøx, during her stay on the OrkneyIslands.74 The historicity of this statement is uncertain. The poem tells of Eiríkr’swelcome into Valhalla, following his death in battle with five kings. These kingsare not named. This may be because the poem is incomplete or it may be, as AxelSeeberg has suggested, that they were not historical kings but were rathermentioned as a literary embellishment intended to raise the profile of Eiríkr.75

Eiríksmál merely provides a witness that Eiríkr died in battle and that those whowished to commemorate him felt comfortable using overtly heathen imagery,whether or not Eiríkr himself was a christian. This does not contradict theargument advanced so far in this paper that Eiríkr blóðøx and Eiríkr of York areseparate individuals.

The conclusion of this discussion is that we have no identification of Eiríkr ofYork with Eiríkr blóðøx from a date earlier than the end of the twelfth century.What was written at that time seems to come from oral tradition, which over sucha span (250 years) is likely to have been subject to embellishment or invented.There is also conflict between accounts of the life of Eiríkr blóðøx in Scandinaviansources and the evidence concerning Eiríkr of York in Insular sources.Nevertheless, the small body of skaldic poetry which clearly relates to Eiríkr ofYork does not conflict with Insular records. One possible solution to theseproblems is to suggest that Eiríkr of York and Eiríkr blóðøx were separateindividuals who were confused in later Scandinavian accounts.

This would not be the only case of confused identities in mediaeval Icelandicsaga. Alistair Campbell noted in a lecture published in 1971 ‘that prose may be

71Hákonar saga góða, §§19, 27, and Haralds saga gráfeldar, §§1, 6 (Snorri, ed. Finnur Jónsson,pp. 82, 86, 94, 97; Heimskringla, transl. Hollander, pp. 113, 117, 129, 135).

72Corpus, edd. & transl. Gudbrand Vigfusson & York Powell, II.40: ‘Brandr fær logs ok landalandz Eiriki banda’ (‘His sword wins Eiríkr lands and gold’). If this fragment is correctly attributedto Glúmr Geirason, it is likely to refer to Eiríkr blóðøx and his activities in Norway.

73Seeberg, ‘Five kings’, p. 106; Fagrskinna. Nóregs kononga tal, ed. Finnur Jónsson (København1902/3), pp. 28–30; Chronicles, transl. Page, pp. 110–11.

74Fagrskinna, ed. Finnur Jónsson, p. 27; Chronicles, transl. Page, p. 110; Fagrskinna. A Catalogueof the Kings of Norway, transl. Alison Finlay (Leiden 2004), pp. 58–9.

75Seeberg, ‘Five kings’, p. 110.


wrong on its attribution of the verse to particular persons or events’, and he wenton to cite some examples.76 Matthew Townend has more recently discussed theconscious or accidental recontextualisation of memories of Scandinavian kings ofYork into a setting which was more meaningful to Icelandic saga-writers. The twocases which he presented were Óláfr Guðrøðsson and Sigtryggr caech, whobelonged to the dynasty of Ívarr (Uí Ímair) whose members ruled Dublin andYork intermittently in the tenth century.77 In developing his argument Townenddrew on the work of earlier scholars. Campbell had argued that Óláfr Guðrøðsson,leader of the Scandinavian troops at Brunanburh, was reïnvented as Óláfr the Red,king of Scotland, by later saga-writers. Campbell suggested that this was becauseat the time when the sagas were written ‘the likeliest nation to invade Englandwere the Scots’.78 It is therefore possible that oral memories from Northumbriaregarding Eiríkr of York were recontextualised to fit with legends of the NorwegianEiríkr blóðøx. This could be because Eiríkr of Norway was more famous thanEiríkr of York at the time when the sagas were written. The fame of the formermay be due to the importance of the sons of Eiríkr blóðøx, who ruled Norway,and in turn because of the greater fame of his father, Haraldr hárfagri.79

However, one may ask why Eiríkr of York and Eiríkr son of Haraldr hárfagriwere regarded as synonymous at the end of the twelfth century if that was not ahistorical reality in the tenth. Other features of the Scandinavian literary recordraise such questions, notably the association claimed between Eiríkr blóðøx andÆthelstan. I suggest that the answer to these questions may lie in the legendarysignificance which Haraldr hárfagri and Æthelstan had after their deaths.80 Duringthe Middle Ages both were perceived as founding figures of national unity, to theextent that in legend they became emblematic of their peoples’ identities. I suggestthat this status was influential in the way in which both kings became involved inthe story of Eiríkr of York.

I shall deal first with Haraldr hárfagri and his alleged paternity of Eiríkr ofYork. This should be assessed in the context of Haraldr’s links to Scandinaviansettlements across northwestern Europe, and it is worth taking some time todiscuss this issue. Haraldr, his sons, and his associate Røgnvaldr, jarl of Møre,were linked to stories of Scandinavian settlement in Iceland, in the Scottish isles

76Alistair Campbell, Skaldic Verse and Anglo-Saxon History (London 1971), p. 3.77M. Townend, ‘Whatever happened to York viking poetry? Memory, tradition and the

transmission of skaldic verse’, Saga-book 27 (2003) 48–90.78Campbell, Skaldic Verse, p. 6.79Hines, ‘Egill’s Höfuðlausn’, p. 88. Hines has added that ‘the personal history of this temporary

king [Eiríkr] of Northumbria did not provide an obviously well-suited character for historically falseadoption as the father of kings of Norway’. However, the Norwegian king Eiríkr blóðøx wasdeliberately portrayed as evil in Icelandic prose, perhaps as a foil to his good brother Hákon, and toprovide a moral tale around Eiríkr’s loss of power in Norway.

80Smyth, Scandinavian York, II.169.


and Man, in Ireland, and in Normandy. Some of these stories have been regardedas mythical rather than historical. They relate to Haraldr’s significance as foundingfigure of a unified Norwegian kingdom and progenitor of the Norwegian royalline.

Haraldr is a major figure in mediaeval Icelandic literature and its ideology ofnationhood: Iceland was settled as a consequence of the tyranny of Haraldrhárfagri, which caused many to leave Norway and start a new life in Iceland. Somecommentators have doubted the story. For example, Jesse Byock has suggestedthat Haraldr may have had some influence in the settlement of Iceland, but thathis role has probably been exaggerated.81 Haraldr seems eventually to symbolisethe Norwegian cultural origins of the Icelanders but also the royal oppression fromwhich the first settlers sought liberty. Both issues (origins, and freedom from royalrule) were a matter of cultural pride and identity among mediaeval Icelanders.82

Furthermore, it is through the medium of Icelandic sources that we gain moreinformation concerning Haraldr’s alleged involvement in other Scandinaviancolonies.

Peter Sawyer has argued that the association of Haraldr hárfagri withScandinavian settlements in the Scottish isles and in Man was influenced by hisstatus in Icelandic legend.83 Sawyer has stated that ‘It is not surprising that theIcelanders were ready to believe that Haraldr extended his power to the BritishIsles for it enabled them to reconcile a conviction that their ancestors had fled toIceland from the tyranny of Haraldr with the knowledge that many of them cameto Iceland not from Norway, but from the British Isles’.84 Sawyer has alsosuggested that the saga-accounts of Haraldr harrying the Northern Isles andHebrides may have been inspired by the expeditions of Magnús berfœttr (MagnúsBarelegs) to the same region in 1098 and 1102.85 Thus later events may haveaffected Scandinavians’ interpretations of their past.

The story of Haraldr’s involvement in the Scottish isles is most fully expressedin §§8–9 of Orkneyinga saga,86 where we read that Haraldr plundered theNorthern Isles, the Hebrides, and Man as a punishment for the piracy of theirScandinavian inhabitants. The Northern Isles were then granted to Røgnvaldr,jarl of Møre. However, Røgnvaldr’s position was usurped by Hálfdan son of

81Byock, Viking Age Iceland, pp. 82–4.82Íslendingabók, Landnámabók, ed. Jakob Benediktsson (Reykjavík 1968), p. cii; Orkneyinga

saga, transl. Hermann Pálsson & Edwards, p. 19.83Sawyer, ‘Harold Finehair’; P.H. Sawyer, Kings and Vikings, Scandinavia and Europe, A.D. 700–

1100 (London 1982), p. 13.84Sawyer, ‘Harald Finehair’, pp. 107–8.85Ibid., p. 107.86Orkneyinga saga, §8 (ed. Finnbogi Guðmundsson, pp. 12–20; transl. Hermann Pálsson &

Edwards, pp. 29–33).


Haraldr hárfagri who established himself as a king in Orkney. Hálfdan was slainby Røgnvaldr’s son Einarr. This provoked Haraldr hárfagri to undertake anotherexpedition to the west, but eventually Einarr and Haraldr came to terms. PeterSawyer has presented a strong case for rejecting this narrative as legend derivativeof a range of Insular source-material.87 In §9 of Orkneyinga saga, we read thatEiríkr blóðøx visited Orkney after being exiled from York and that his daughterRagnhildr married Arnfinnr, son of Jarl Þórfinnr.88 Ragnhildr is portrayed verymuch like a fickle sovereignty-goddess of the sort found in Gaelic legends.89 Shedisposes of husbands, but only the men who marry her attain the earldom. Isuggest that these details in §§8–9 of Orkneyinga saga are unreliable as a historicalsource.

Haraldr hárfagri was deemed to have been a founder-figure in the Scandinaviancolonies in Ireland too. This assertion is found in the thirteenth-century Welshtext Historia Gruffud vab Kenan and in its twelfth-century Cambro-Latin sourceVita Griffini filii Conani;90 we read that Haraldr built the city of Dublin, alongwith many other strongholds, and that his progeny still ruled there.91 This legendmay have been derived from ‘the oral accounts of twelfth-century Dubliners awareof their distant Norse heritage’ or it may bear the stamp of Icelandic influence.92

Any notion that Haraldr hárfagri ruled any part of Ireland has been entirelyrejected on the basis of analysis of Irish records.93 A conflicting and equallyspurious account is found in the Topographia Hiberniae of Gerald of Wales (1146–1223). He asserted that three brothers Amelauus, Sitaracus, et Yuorus (Óláfr,

87Sawyer, ‘Harald Finehair’; Sawyer’s discussion included Irish and Welsh sources.88The story of Ragnhildr’s marriage is also described variously in Fagrskinna, the separate Saga

of St Óláfr, and Egils saga and Heimskringla, and has been the subject of a recent study: Cormack,‘Egils saga’.

89Orkneyinga saga, §9 (ed. Finnbogi Guðmundsson, pp. 20–2; transl. Hermann Pálsson &Edwards, pp. 33–5). See, for example, M. Herbert, ‘Goddess and king: the sacred marriage in earlyIreland’, in Women and Sovereignty, ed. Louise Olga Fradenburg (Edinburgh 1992), pp. 264–75;P. Lysaght, ‘Aspects of the earth-goddess in the traditions of the banshee in Ireland’, in The Conceptof the Goddess, edd. Sandra Billington & M. Green (London 1996), pp. 152–65, at 158 and 160–1.

90K.L. Maund, Ireland, Wales, and England in the Eleventh Century (Woodbridge 1991), pp. 172–3; J. Jesch, ‘Norse historical traditions and the Historia Gruffud vab Kenan: Magnús berfœttr andHaraldr hárfagri’, in Gruffudd ap Cynan. A Collaborative Biography, ed. K.L. Maund (Woodbridge1996), pp. 117–47, at p. 146. I had not at the time of writing seen the recently rediscovered andpresumptively twelfth-century Latin original: but see now Vita Griffini filii Conani. The MedievalLatin Life of Gruffudd ap Cynan, ed. & transl. Paul Russell (Cardiff 2005).

91Historia Gruffud, §5 – A Mediaeval Prince of Wales: The Life of Gruffudd ap Cynan, ed. & transl.D. Simon Evans (Felinfach 1990), pp. 25 and 56.

92D.E. Thornton, ‘The genealogy of Gruffudd ap Cynan’, in Gruffudd ap Cynan, ed. Maund,pp. 79–108, at p. 103; Jesch, ‘Norse historical traditions’, pp. 146–7.

93Thornton, ‘The genealogy’, pp. 90–3.


Sigtryggr, and Ívarr) founded the three viking-towns of Dublin, Waterford, andLimerick.94 This ‘three-brothers’ motif may be seen elsewhere in literatureconcerning Scandinavian kings.95 Thus we can see that by the twelfth centurythere was a lively interest in legends concerning Scandinavian rulers in Britain andIreland.96 This may have influenced stories of Eiríkr blóðøx, which were evolvingat that time.Vita Griffini and Historia Gruffud also provide evidence to supplement that of

‘The Fragmentary Annals of Ireland’, indicating that kings of Uí Ímair perceivedthemselves as descendants of the Norwegian royal line of the Ynglings. This takesthe form of a genealogy of Gruffudd, king of Gwynedd.97 Gruffudd was relatedto the royalty of Dublin. One of his ancestors was Óláfr cuarán, who ruled Yorkfrom 949 to 952. According to the Historia, Óláfr was a great-grandson of Haraldrhárfagri. Eiríkr blóðøx is therefore not the only mediaeval link made betweenrulers of Northumbria and the Norwegian royal family.

The other point of interest regarding the biographies of King Gruffudd isJudith Jesch’s suggestion that the inclusion of Haraldr hárfagri was influenced bythe fame of his alleged descendant Haraldr hardráði.98 This leads me to suggestthat later events, notably the reign of Haraldr hardráði over Norway, which wasfollowed by his very brief success in Northumbria, could provide a model for thestory of Eiríkr blóðøx. Following Sawyer’s argument about the influence ofMagnús berfœttr on the legend of Haraldr hárfagri in the Scottish isles, we couldsuppose that Scandinavian legends about Northumbria were inspired by laterhistorical examples of interaction.99

Two conflicting accounts which link Haraldr hárfagri with Normandy arepreserved in mediaeval literature. In Vita Griffini and Historia Gruffud, Rollo(Rodulphus), the ancestor of the Norman ducal line (and consequently of theAnglo-Norman kings), is presented as a brother of Haraldr hárfagri.100 However,

94Topographia Hiberniae, III.43: ed. J.J. O’Meara, ‘Giraldus Cambrensis in TopographiaHibernie: text of the First Recension’, Proceedings of The Royal Irish Academy 52 C (1948–50) 113–78, at p. 175; Gerald of Wales, The History and Topography of Ireland, transl. John O’Meara (2ndedn, Portlaoise/Harmondsworth 1982), p. 122.

95Gesta, II.25 (22): Magistri Adami Bremensis Gesta, ed. Schmeidler, p. 84; Adam, transl. Tschan,p. 71.

96Alfred P. Smyth, Scandinavian Kings in the British Isles, 850–880 (Oxford 1977), pp. 101–17.97Historia Gruffud, §4 (A Mediaeval Prince, ed. & transl. Evans, pp. 24 and 54–5).98Jesch, ‘Norse historical traditions’, pp. 139–44.99Sawyer, ‘Harald Finehair’, p. 107. In the time of Knútr, Jarl Eiríkr and Jarl Hákon of Hlaðir

– who controlled Norway – also had authority in Northumbria. These too might have been latermodels around which the ‘Bloodaxe’-legend developed. See S. Keynes, ‘Cnut’s earls’, in The Reignof Cnut, King of England, Denmark and Norway, ed. Alexander R. Rumble (London 1994), pp. 43–88, at 57 and 61.

100Historia Gruffud, §5 (A Mediaeval Prince, ed. & transl. Evans, pp. 25–6 and 56).


in Heimskringla, Rollo (or Göngu-Hrólfr) is portrayed as a son of Røgnvaldr, jarlof Møre, the same who is said to have received Orkney from Haraldr.101 Anotherversion of events is presented in Laxdaela saga. Here we read that Göngu-Hrólfrwas a son of Oxnaþórir, ‘a hersir of good family’ in Norway.102 However, asNormandy is not mentioned in Laxdaela saga, it may be that anotherGöngu-Hrólfr is intended. The account given in Heimskringla of Rollo’s paternalorigins was accepted by David Douglas in 1942, and he has been followed bysubsequent scholars in the field of Norman history.103 Douglas accepted thesaga-evidence because he thought the other main source of early Norman history,De moribus et actis primorum Normanniae ducum, unreliable. This was writtenby Dudo of Saint-Quentin in 996×1020.104 It is significantly earlier than eitherthe biographies of King Gruffudd or Heimskringla. Dudo asserted that Rollo hadnoble Danish origins.105 This view is not substantiated by other mediaeval texts.106

It may reasonably be doubted, for De moribus is a fantastically imaginativecomposition. Eric Christiansen has recently, and I think correctly, pointed outthat Scandinavian sagas and the genealogical claims of the biographies of KingGruffudd are also superb texts but dubious sources for Rollo’s pedigree.107 Thesetexts should not be regarded as more reliable evidence than Dudo. They reïteratethe same pattern noted so far, of Harladr hárfagri or his associates being given alegendary role in the history of Scandinavian colonies.

The Northumbrian link to Haraldr hárfagri should perhaps be considered inthis context. The literary links made between Norway’s most famous king andthe Scandinavian colonies in Britain, Ireland, and Normandy could all behistorically inaccurate. The exaggerated nature of this link may be seen in Egilssaga, §4, where we read that Jämtland and Hälsingland, the Hebrides, Ireland,Normandy, the Northern Isles, and Caithness were all settled by Norwegians in

101Haralds saga hárfagra, §24 (Snorri, ed. Finnur Jónsson, pp. 78–9; Heimskringla, transl.Hollander, pp. 78–9).

102Laxdæla saga, §32: Laxdæla saga, Halldórs þættir Snorrasonar, Stúfs þáttr, ed. Einar Ól. Sveinsson(Reykjavík 1934), pp. 85–7, at p. 85; ‘The Saga of the People of Laxardal’, transl. K. Kunz, in TheSagas, gen. ed. Örnólfur Thorsson, V.1–120, at pp. 42–3.

103David C. Douglas, Time and the Hour – Some Collected Papers (London 1977), pp. 121–40,at 122–5; David Bates, Normandy before 1066 (Harlow 1982), p. 7; E. Searle, ‘Fact and pattern inheroic history: Dudo of Saint-Quentin’, Viator 15 (1984) 119–37, at p. 125.

104Dudo of St Quentin, History of the Normans, transl. Eric Christiansen (Woodbridge 1998),p. xiii.

105De moribus et actibus primorum Normanniae ducum, ed. Jules Lair (Caen 1865), pp. 141–2(II.2); Dudo, transl. Christiansen, p. 26.

106Douglas, Time, pp. 122,125. Richer of Reims (Historia, I.28) stated that Rollo was a son of‘Catillus’ (presumably Old-Scandinavian Ketill), but nothing is known of his ancestry: Richer –Histoire de France (888–995), ed. & transl. Robert Latouche (2 vols, Paris 1930/7), I.64/5.

107Dudo, transl. Christiansen, p. 187 (n. 114).


flight from Haraldr hárfagri.108 These stories served to reïnforce a sense ofScandinavian identity and common origins, which distinguished the inhabitantsof the colonies from their neighbours. This sense of Scandinavian identitypersisting in the colonial settlements abroad is evidenced in the material cultureand literature of these areas in the twelfth and thirteenth centuries.109 Thedevelopment of the legends of Haraldr hárfagri was also much to the taste ofNorwegian and Icelandic historians and saga-writers. Their perceptions of therole of Haraldr hárfagri in their own cultural and political identity may also haveinfluenced the development of the legends. For this reason Scandinavianhistorians and saga-writers may have been keen to develop stories about the sonof Haraldr hárfagri in England.

One further matter to pursue in relation to Eiríkr blóðøx is his allegedassociation with Æthelstan, first king of England. In Egils saga we read thatÆthelstan granted Northumbria to Eiríkr blóðøx on condition that he protect theland from the Scots and Irish.110 In Historia Norwegie and Heimskringla this grantis recorded as having the proviso that he would accept baptism.111 The idea thatEiríkr blóðøx ruled Northumbria with Æthelstan’s consent is at variance with thecontemporaneous tenth-century English chronicle-record. A similar legendappears in the story of Rollo as narrated by Dudo of Saint-Quentin.112 Accordingto this account, Rollo was exiled from Denmark; he fled to England. Followinga peace-agreement, Æthelstan offered him the opportunity to settle in England ifhe would defend the land from aggressors and agree to baptism. This agreement,as David Douglas showed, is chronologically impossible.113 Rollo had alreadysettled in Normandy before Æthlestan came to power. We have two similar storieswhich do not match the historical record.

Neither story challenges the supremacy of Æthelstan in England, but this ‘mostchristian king’ is shown as holding the Scandinavian exile in high esteem.114 Ineach case, positive reference is made by the English king to the deeds and reputeof the exile’s family, and this is said to be a cause of his generosity. These accountstherefore shed a flattering light on the leading families of Norway and Normandyby reference to Æthelstan. Elsewhere, Æthlestan is said to have cultivated links

108Egils saga, §4 (ed. Nordal, pp. 9–12; transl. Scudder, pp. 35–6).109For twelfth-century Dublin, see chapter 8, above.110Egils saga, (ed. Nordal, p. 176 [§59] ; transl. Scudder, p. 116 [§60]); cf. Hákonar saga góða,

§3 (Snorri, ed. Finnur Jónsson, pp. 71–2; Heimskringla, transl. Hollander, p. 98).111Historia Norwegie, §XII (edd. & transl. Ekrem et al., pp. 80–3); Heimskringla, transl.

Hollander, p. 98.112Dudo, De moribus, II.7 (ed. Lair, p. 147; transl. Christiansen, p. 33).113Douglas, Time, p. 135, n. 16; Dudo, transl. Christiansen, p. 189 (n. 137).114Dudo, De moribus, II.7 (ed. Lair, p. 147; transl. Christiansen, p. 33); Hákonar saga góða, §3

(Snorri, ed. Finnur Jónsson, p. 71; Heimskringla, transl. Hollander, p. 98).


with Norway during his reign. According to William of Malmesbury, Haraldrhárfagri sent a fine ship to the king, which had a golden beak and purple sails.115

It is uncertain what is to be made of this, as William was the narrator of many tallstories. There is also a story in Scandinavian literature that Eiríkr’s brother andenemy Hákonn was fostered at Æthelstan’s court. The earliest evidence for thisseems to be a poem attributed to the skald Sighvatr Þórðarson, which may havebeen composed in the 1030s.116 As the two kings appear to have beencontemporaries, a link is possible. It may be in consequence of such contact thatÆthelstan features in the Scandinavian pseudohistory of Eiríkr blóðøx and Rollo.Æthelstan also loomed large in historical record and legend as the king who hadcreated the kingdom of England. It may have seemed appropriate to saga-writersto link this prestigious English king with Norwegian royalty.117

It may be argued that in the twelfth century, when the legends of Eiríkr blóðøxwere first written up, the political cults of Haraldr hárfagri and Æthelstan haddeveloped alongside growing awareness of national identity in Norway andEngland. I suggest that this had a direct bearing on the literary image of Eiríkr ofYork, particularly as Northumbria displayed a fusion of identities which wereEnglish and Scandinavian in origin.118 In sum, I hope to have shown that theidentification of Eiríkr of York as Eiríkr blóðøx may be doubted. The evidencethat they were identical is late and contradicts what is known from tenth-centurysources. I therefore wish to explore whether an alternative identity for Eiríkr ofYork can be suggested.

The evidence from Insular sources may be used to suggest that Eiríkr was amember of the Hiberno-Scandinavian dynasty known in Irish as Uí Ímair.119 Thesesources bear testimony to Uí Ímair providing the main opposition to English rulein Northumbria. At least six kings of Uí Ímair ruled at York in the course of thetenth century.120 Their names are italicised in the genealogical chart below. There

115Gesta regum Anglorum, II.135 (William of Malmesbury, edd. & transl. Mynors et al., I.216/17).116Corpus, edd. & transl. Gudbrand Vigfusson & York Powell, II.145–8, at p. 146, line 26; Page,

‘The audience’, p. 114; Abrams, ‘The Anglo-Saxons’, p. 217, n. 22.117For discussion, see David N. Dumville, Wessex and England from Alfred to Edgar. Six Essays

on Political, Cultural, and Ecclesiastical Revival (Woodbridge 1992), pp. 141–71 (especially p. 168[–9], n. 183); cf. Michael Lapidge, ‘Some Latin poems as evidence for the reign of Athelstan’,Anglo-Saxon England 9 (1981) 61–98, revised in his Anglo-Latin Literature, 900–1066 (London1993), pp. 81–113. Alfred’s fame as a founder of the kingdom seems to have overtaken that ofÆthelstan in the later Middle Ages: cf. S. Keynes, ‘The cult of King Alfred the Great’, Anglo-SaxonEngland 28 (1999) 225–356, at pp. 227–39.

118J. Hines, ‘Scandinavian English: a creole in context’, in Language Contact in the British Isles,edd. P. Sture Ureland & G. Broderick (Tübingen 1991), pp. 403–27, at 417–18.

119On Uí Ímair, see now Clare Downham, Viking Kings of Britain and Ireland. The Dynasty ofÍvarr to A.D. 1014 (Edinburgh 2007; 2nd edn, 2008).

120Ibid., pp. 63–127.


would probably have had to be strong reasons for preferring an outsider on theNorthumbrian scene (a king exiled from Norway with minimal resources) overthe rights of an established royal dynasty which also dominated the major sea-portsof Ireland. If another identity were to be proposed for Eiríkr of York, a candidatefrom Uí Ímair would therefore seem a likely alternative.

Genealogical chart of tenth-century Scandinavian kings of York

Ívarrob. 873

sons or daughters

Sigtryggyr Røgnvaldr Guðrøðrob. 927 ob. 920 ob. 934

Hálfdan Haraldr Guðrøðr Óláfr cuarán Óláfr Røgnvaldr Blákariob. 926 ob. 940 ob. 951 ob. 981 ob. 941 ob. 945 ob. 948

(? Sigtryggr) Cammán fl. 960s

As I have already shown, the skaldic poetry relating to Eiríkr of York (whichcan be separated from that which relates to Eiríkr blóðøx) provides nocontradiction of the idea that Eiríkr hailed from Ireland. The interchange of thekingship of York between Eiríkr and Óláfr from 947 to 954 could therefore beseen as a struggle between two claimants of Uí Ímair rather than between aNorwegian and a Hiberno-Scandinavian faction. A similar situation prevailed inDublin in the 940s, when royal power alternated between Óláfr (945–7, 948–9)


and his rival Blákari Guðrøðsson (940–5, 947–8).121 It is also interesting todiscover from the Irish chronicles that Eiríkr took power over York at times whenÓláfr was distracted by affairs in Dublin. In 947/8, Óláfr was battling againstBlákari Guðrøðsson, and in 952 Óláfr may have been struggling to secure Dublin,which had lost its king Guðrøðr Sigtryggsson late in 951 (the heir apparent, Ívarr,had also died in 950).122 Óláfr ruled at Dublin from 952 until his abdication in980. Version E of ‘The Anglo-Saxon Chronicle’ indicates that Eiríkr instigated acoup against Óláfr in 952: Norðhymbre fordrifan Anlaf cyning 7 underfengon YricHaroldes sunu (‘the Northumbrians drove out Óláfr and submitted to EiríkrHaraldsson’). The timing of Eiríkr’s accession to the throne of York may indicatethat he was well informed of events in Ireland.123

A case may also be put, on the basis of numismatic evidence, for continuity ofrule by Uí Ímair in Northumbria.124 Two issues of King Eiríkr have beenidentified. The first imitated the coins of his predecessor Óláfr Sigtryggsson.125

121The Annals of Clonmacnoise, being Annals of Ireland from the Earliest Period to A.D. 1408, translatedinto English A.D. 1627 by Conell Mageoghagan, ed. Denis Murphy (Dublin 1896), pp. 151–2, 154 (s.aa.933[=940].7, 937[=945].9); Annala Rioghachta Eireann, Annals of the Kingdom of Ireland, by The FourMasters, from the Earliest Times to the Year 1616, ed. & transl. John O’Donovan (2nd edn, 7 vols, Dublin1856), II.640/1, 654/5, 658/9 (s.aa. 938[=940].10, 943[=945].6, 945[=947].6, 946[=948].6); The Annalsof Inisfallen (MS. Rawlinson B 503), ed. & transl. Seán Mac Airt (Dublin 1951), pp. 152/3 (s.a.[947]=[948].1); The Annals of Ulster (to A.D. 1131), I, edd. & transl. Seán Mac Airt & G. Mac Niocaill(Dublin 1983), pp. 392/3 (s.aa. 944[=945].6, 946[=947].1, 947[=948].1); Chronicum Scotorum. AChronicle of Irish Affairs, from the Earliest Times to A.D. 1135; with a Supplement, containing the Eventsfrom 1141 to 1150, ed. & transl. William M. Hennessy (London 1866), pp. 208/9 (s.a. [947]=[948].1).

122‘The Annals of Clonmacnoise’, s.aa. 945[=950].6, 946[=951].2 (ed. Murphy, pp. 155–6); ‘TheAnnals of The Four Masters’, s.aa. 945[=947].6, 946[=948].6, 948[=950].8 (Annala, ed. & transl.O’Donovan, II.658/9, 662/3); ‘The Annals of Inisfallen’, s.a. [947]=[948].1 (ed. & transl. Mac Airt,pp. 152/3); ‘The Annals of Ulster’, s.aa. 944[=945].6, 946[=947].1, 947[=948].1 (edd. & transl. MacAirt & Mac Niocaill, I.392/3); Chronicum Scotorum, s.aa. [947]=[948].1, [949]=[950].4, [950]=[951].4(ed. & transl. Hennessy, pp. 208–11).

123‘The Annals of The Four Masters’, s.a. 950[=952].11 (Annala, ed. & transl. O’Donovan, II.668/9);‘The Annals of Ulster’, s.a. 951[=952].2 (edd. & transl. Mac Airt & Mac Niocaill, I.396/7). Thesechronicles refer to a victory in England by vikings over the men of England, Strathclyde, and Alba. Thishas been associated with Eiríkr’s rise to power: Smyth, Scandinavian York, II.172. The event is notrecorded in English chronicles. The evident Irish interest in the matter may indicate more than a foreignruler being installed in Óláfr’s place at York; it could be taken to suggest that the viking-contingent whichwon was linked with Ireland.

124M. Blackburn, ‘Expansion and control: aspects of Anglo-Scandinavian minting south of theHumber’, in Vikings and the Danelaw, edd. Graham-Campbell et al., pp. 125–42, at p. 125. His articleprovides a convenient summary of the value of coin-evidence and the methods used to provide relativedating for coin-issues.

125C.E. Blunt et al., Coinage in Tenth-century England from Edward the Elder to Edgar’s Reform (Oxford1989), pp. 223–5; Philip Grierson & M. Blackburn, Medieval European Coinage, with a Catalogue of theCoins in The Fitzwilliam Museum, Cambridge, I,The Early Middle Ages (5th to 10th Centuries) (Cambridge1986), pp. 323–5; Smyth, Scandinavian York, II.159.


This could merely be an indication of administrative continuity. However, Eiríkr’ssecond coin-issue imitated that of Óláfr’s father Sigtryggr.126 These coins bore asword-emblem on the reverse. This has been regarded as a symbol which ‘evokedthe Hiberno-[Scandinavian] conquest of 919’,127 and it has been identified morespecifically as portraying ‘the sword of Carlus’. This sword is recorded as part ofthe royal insignia of Uí Ímair in the late tenth century. The association seemsplausible.128 Even if the sword were taken to mean something else, it seems strangethat an outsider-king would use images which could provoke memories of a rivaldynasty’s claim to his position. Indeed, Mark Blackburn emphasised the level ofpolitical consciousness and creativity which determined coin-emblems inViking-Age Northumbria and the degree of royal control exercised over theirselection.129 It may be that this harking back to an earlier coin-design was intendedto legitimate Eiríkr’s right to rule Northumbria, both as a member of Uí Ímairand possibly as a descendant of Sigtryggr.

Beyond these indications there are personal circumstances in Eiríkr’s life whichcould point to his origins among Uí Ímair. The Life of St Cathroe reports thatEiríkr of York married a kinswoman of the saint, who was therefore related toroyal families of Alba and/or Strathclyde.130 This assertion contrasts with skaldicpoems which describe Gunnhildr as the wife of Eiríkr blóðøx.131 The author ofthe Life of St Cathroe should have been well informed about the events, as theLife is attributed to the 980s, a short time after Cathroe’s death, and it seems tohave been written at Metz where the saint had resided.132 The marriage-alliancewhich is described between Eiríkr of York and the kinswoman of Cathroe, fitswith the Irish-Sea/Northumbrian interests of Uí Ímair.

126Blunt, Coinage, pp. 228–9; Grierson & Blackburn, Medieval European Coinage, I.323–5;Smyth, Scandinavian York, II.160, where it is suggested that this was due to Eiríkr’s ignorance of,or lack of say in, coin-design. In the light of Mark Blackburn’s comments (‘Expansion’, pp. 135–6) about earlier issues of Anglo-Scandinavian coins, royal involvement in coin-design is likely.Smyth’s portrayal of Eiríkr as a mere puppet of Wulfstan seems exaggerated.

127Michael Dolley, Viking Coins of the Danelaw and of Dublin (London 1965), p. 26; Smyth,Scandinavian York, II.160.

128Grierson & Blackburn, Medieval European Coinage, I.323, 325; Smyth, Scandinavian York,I.107. This is, of course, speculative.

129M. Blackburn, ‘The coinage of Scandinavian York’, in The Archaeology of York, ed. R.A. Hall,VIII, Anglo-Scandinavian York, part 4, Aspects of Anglo-Scandinavian York (York 2004), pp. 325–49.I am grateful to Dr Blackburn (whose recent death is a great loss) for allowing me sight of this articlebefore its publication.

130Vita Sancti Kaddroe, §17 (Acta, ed. Colganus, p. 497; Early Sources, transl. Anderson, I.441).See n. 35, above.

131Egils saga, stanzas 11 and 31 (ed. Nordal, pp. 113, 170 [§§45, 57]; transl. Scudder, pp. 84,113 [§§45, 58]).

132Dumville, ‘St Cathróe’, p. 173, and Celtic Essays, 2001–2007, I.166.


The record of Eiríkr’s death on Stainmore (in that part of modern Cumbriacalled Westmorland until 1974) also fits with the spheres of influence of Uí Ímair.Stainmore lay on a main route following the old Roman roads from York acrossto the Irish-Sea coast.133 The location could suggest that Eiríkr was attempting toreach home-territory. The significance of Stainmore’s location en route from Yorkto Dublin (or into Strathclyde) is not represented in Scandinavian sources.134

The twelfth- and thirteenth-century authors Symeon of Durham, Roger ofHowden, and Roger of Wendover (in a chain of interdependence) asserted thatEiríkr was slain by one Maccus.135 This name does not appear in the Scandinaviansources. According to David Thornton, Maccus is a name of mixed Gaelic andScandinavian usage, developed from macc (Old and Middle Gaelic for ‘son’).136

The appearance of this name and the alleged location of Eiríkr’s death couldsuggest a link with the viking-colonies in Gaeldom.137

This begs a question: if Eiríkr was a member of Uí Ímair, who exactly was he?There are several persons called Haraldr (Aralt) recorded in Irish chronicles. Three

133Collingwood, ‘King Eirík’, p. 322; Smyth, Scandinavian York, II.174. In English texts of thetwelfth and thirteenth centuries Eiríkr’s death is recorded in the same year as his expulsion fromNorthumbria: Historia regum Anglorum, Part II, s.a. 1072 (Symeonis Monachi Opera, ed. Arnold,II.197, and The Church Historians, transl. Stevenson, III.2, pp. 556–7); Flores historiarum, s.a. 950(Rogeri de Wendover Chronica, ed. Coxe, I.401; Roger of Wendover’s Flowers, transl. Giles, I.256);Chronica Magistri Rogeri, s.a. 952 (ed. Stubbs, I.57; The Annals, transl. Riley, I.67). Roger ofWendover is the sole source for Stainmore as the location of Eiríkr’s death.

134Ágrip, §7 (ed. & transl. Driscoll, pp. 16/17); Historia Norwegie, §XII (edd. & transl. Ekremet al., pp. 80–3); Hákonar saga góða, §4 (Snorri, ed. Finnur Jónsson, pp. 72–3; Heimskringla, transl.Hollander, p. 99). Notwithstanding the assertion in Ágrip and Historia Norwegie that Eiríkr died inSpain, Heimskringla shows the legendary Eiríkr raiding far and wide before falling in his last battleat an unidentified location in England.

135Historia regum Anglorum, Part II, s.a. 1072 (Symeonis Monachi Opera, ed. Arnold, II.197; TheChurch Historians, transl. Stevenson, III.2, pp. 556–7; Chronica Magistri Rogeri, s.a. 952 (ed. Stubbs,I.57; The Annals, transl. Riley, I.67); Flores historiarum, s.a. 950 (Rogeri de Wendover Chronica, ed.Coxe, I.401; Roger of Wendover’s Flowers, transl. Giles, I.256).

136D.E. Thornton, ‘Hey Macc! The name Maccus, tenth to fifteenth centuries’, Nomina 20(1997) 67–94. The Maccus who is associated with Eiríkr is called a son of Óláfr in two closelyrelated texts, Historia regum Anglorum, Part II, and the Chronicle of Roger of Howden: Historiaregum Anglorum, Part II, s.a. 1072 (Symeonis Monachi Opera, ed. Arnold, II.197; The ChurchHistorians, transl. Stevenson, III.2, pp. 556–7); Roger of Howden, Chronica, s.a. 952 (ed. Stubbs,I.56; transl. Riley, I.67). W.G. Collingwood and Alfred Smyth regarded him as a son of ÓláfrSigtryggsson: Collingwood, ‘King Eirík’, pp. 321–3; Smyth, Scandinavian York, II.175.

137For debate about the reliability of Roger of Wendover’s Chronicle, see English HistoricalDocuments, c. 500–1042, transl. Dorothy Whitelock (2nd edn, London 1979), pp. 281–4 (no. 4);David N. Dumville, ‘Textual archaeology and Northumbrian history subsequent to Bede’, in Coinagein Ninth-century Northumbria, ed. D.M. Metcalf (Oxford 1987), pp. 43–55, at 44–5, reprinted inhis book Britons and Anglo-Saxons in the Early Middle Ages (Aldershot 1993), essay X.


of these belong to the tenth century.138 None can be proved to have had a soncalled Eiríkr. However, it has been noted that Irish chronicles provide littlereference to the careers of members of Uí Ímair at York and that, though extensive,the chronicles are not comprehensive in their coverage of political events.139

Irish chronicles do not yield reliable evidence for someone called Eiríkr in thetenth century.140 It is possible that the ‘Laraic’ who raided St Mullins (Co. Carlow)from the sea in 953 has a corrupted form of this name.141 The only referencegiven outside the chronicles is to an Eiríkr, king of the Isles (presumably theHebrides), in the twelfth-century saga Caithréim Chellacháin Chaisil.142 This isan untrustworthy source based on events in the reign of Cellachán, overking ofMunster, who reigned from 940 to 952. The author(s) quarried genealogical

138‘The Annals of Clonmacnoise’, s.aa. 921[=926].3, 933[=940].7, 992[=999].4 (ed. Murphy,pp. 148, 151, 164); ‘The Annals of The Four Masters’, s.aa. 924[=926].4, 938[=940].10,998[=999].10 (Annala, ed. & transl. O’Donovan, II.614/15, 638–41, 740/1); ‘The Annals ofInisfallen’, s.a. [940].1 (ed. & transl. Mac Airt, pp. 152/3); The Annals of Tigernach, ed. & transl.Whitley Stokes (2nd edn, 2 vols, Felinfach 1993), II.351 (s.a. [997]=[999].1); ‘The Annals of Ulster’,s.a. 998[=999].8 (edd. & transl. Mac Airt & Mac Niocaill, I.428/9); Chronicum Scotorum, s.aa.[925]=[926].4, [939]=[940].2, [997]=[999].4 (ed. & transl. Hennessy, pp. 196/7, 202/3, 236/7).

139Smyth, Scandinavian York, II.162: ‘the kings of Dublin are described in Irish sources withlittle reference to their careers at York’.

140In their annal for 937, ‘The Annals of Clonmacnoise’ record an ‘Arick McBrith’ who fell atthe battle of Brunanburh. Collingwood identified this name as Eiríkr Barðarson, whereas AlistairCampbell deemed it to be Hárekr Barðarson: Collingwood, ‘King Eirík’, p. 317; The Battle ofBrunanburh, ed. Alistair Campbell (London 1938), p. 57; Smyth, Scandinavian York, II.39. Oneof the copies of ‘The Annals of Clonmacnoise’ has ‘abbot of Arick McBrith’, which is likely to bean error. The sources of this chronicle’s account of the battle of Brunanburh are not certain, andsome of the names of the fallen may therefore be unreliable: The Battle, ed. Campbell, pp. 56, 159.Allen Mawer and David Douglas reported that Irish pirates named Eiríkr and Barðr sacked Toursin 903: Allen Mawer, The Vikings (Cambridge 1913), p. 51; Douglas, Time, p. 126. Mediaevalsources do not record that these vikings came from Ireland: Auguste Eckel, Charles le Simple (Paris1899), pp. 67–9. Mawer stated that his hypothesis was based on the frequency of the name Barðr(as Old- and Middle-Gaelic Barid, Barith) in Irish records and on the fact that this raid followedsoon after the expulsion of vikings from Dublin. This is an interesting idea but remains speculative.

141‘The Annals of The Four Masters’, s.a. 951[=953].10 (Annala, ed. & transl. O’Donovan,II.698/9). This does not seem to be a Gaelic name: ibid., VII.193 (index). For discussion, see above,chapter 7 (pp. 140–1). In conversation, Martin Syrett has commented that this name may beScandinavian, ending with the element ríkr, but he was not sure what name that might be. Thereis a slight possibility that an original reading la Araic was corrupted to ó Laraic. It would be necessaryto examine the autograph manuscripts of ‘The Annals of The Four Masters’ before commentingfurther on this point.

142Caithréim Chellacháin Chaisil, §44: Caithreim Cellachain Caisil, The Victorious Career ofCellachan of Cashel, or The Wars between the Irishmen and the Norsemen in the Middle of the 10thCentury, ed. & transl. Alexander Bugge (Oslo 1905), pp. 25, 83.


literature and Cocad Gaedel re Gallaib as a source of names.143 Limited use mayalso have been made of chronicle-evidence.144 From Donnchadh Ó Corráin’sthorough analysis of this saga, it emerges that almost half of the people named init were contemporary with the events being described.145 Ó Corráin hasdemonstrated that the author(s) of that text deliberately sought out and usedtenth-century names in the story (apart from names included for political reasonsor through artistic licence). This may have been an attempt to add an air ofauthenticity. Many other names in the text seem to refer to eponymous ancestorsof dynasties who were significant in the twelfth century when the text was written.Their inclusion appears to represent the political agenda of the patron of thetext.146 Other names in the saga may be pure invention. While the content ofCaithréim Chellacháin Chaisil is unreliable, Alexander Bugge and Alfred Smythunreservedly identified the Eiríkr of the Isles who is mentioned in it with Eiríkrof York, and consequently with Eiríkr blóðøx.147 Ó Corráin has more cautiouslysuggested that ‘they may be identical .... It may equally be a coincidence.’148

It would be hasty to jump to conclusions regarding an alternative identity forEiríkr of York. The evidence is fragmentary. The only possibility which I cantentatively suggest from the available source-material is that he was a son ofHaraldr Sigtryggsson. Haraldr ruled Limerick and died in 940.149 He has beenidentified as the father of meicc Arailt, ‘the sons of Haraldr’, who were active on

143D. Ó Corráin, ‘Caithréim Chellacháin Chaisil: history or propaganda?’, Ériu 25 (1974) 1–69, at pp. 60–2.

144Ibid., pp. 20, 30.145Ibid., pp. 9, 10, 11, 12, 15, 19, 23, 25, 26, 27, 28, 31, 32, 36, 38, 42, 45, 47.146The anachronistic names in the text have been discussed by Ó Corráin, ibid., pp. 10, 11, 12,

19, 20, 21, 23, 24, 27, 31, 32, 33, 34, 35, 36, 39, 42, 47, 52, 53. Some of these anachronisms areone or two generations out of place. This may be due to the author’s attempt to gathermid-tenth-century names from unreliable genealogical sources: ibid., p. 17. It is interesting thatthe poem which refers to Eiríkr in Caithréim Chellacháin Chaisil states that Sigtryggr was king ofDublin. The idea that there was a king of Dublin named Sigtryggr in the mid-tenth century isfound elsewhere: ‘The Annals of The Four Masters’, s.a. 939[=941].5 (Annala, ed. & transl.O’Donovan, II.642/3); D. Ó Corráin, ‘Muirchertach Mac Lochlainn and the Circuit of Ireland ’, inSeanchas. Studies in Early and Medieval Irish Archaeology, History and Literature in Honour of FrancisJ. Byrne, ed. Alfred P. Smyth (Dublin 2000), pp. 238–50.

147Caithreim Cellachain Caisil, ed. & transl. Bugge, p. 148; Smyth, Scandinavian York, II.177.148Ó Corráin, ‘Caithréim Chellacháin’, p. 57.149‘The Annals of Clonmacnoise’, s.a. 933[=940].7 (ed. Murphy, p. 151); ‘The Annals of The

Four Masters’, s.a. 938[=940].10 (Annala, ed. & transl. O’Donovan, II.638–41); ‘The Annals ofInisfallen’, s.a. [940].1 (ed. & transl. Mac Airt, pp. 152/3); Chronicum Scotorum, s.a. [939]=[940].2(ed. & transl. Hennessy, pp. 202/3).


the Irish Sea in the late tenth century.150 Two sons of Haraldr ruled in Man andthe Isles, and Uí Ímair maintained power there (perhaps intermittently) foranother century.151 Smyth and Ó Corráin have argued that the involvements ofUí Ímair in the Hebrides began in the early tenth century.152 If Eiríkr was a sonof Haraldr Sigtryggsson, he would have had a claim to the Northumbrian thronethrough his uncle Óláfr, and from his grandfather Sigtryggr (it may be worthkeeping in mind that Eiríkr’s second issue of coins imitated those of Sigtryggr).153

Eiríkr might also have been a king of the Isles, if Caithréim Chellacháin Chaisilcan be allowed to carry any weight.

In sum, the identification of Eiríkr of York with Eiríkr blóðøx can be regardedas late, unreliable, and in contradiction of what earlier sources tell us about Eiríkr’slife. This identification should be questioned. On the other hand, an origin forEiríkr among Uí Ímair is plausible and should be considered. It seems that Irishevidence was somewhat overlooked in the nineteenth-century debate about thepaternal origins of Eiríkr of York.154 My solution, to explain the contradictions

150This identification is based on the record in ‘The Annals of The Four Masters’, s.a.972[=974].9 (Annala, ed. & transl. O’Donovan, II.698/9). I am indebted to Kristin Bornholdt forpointing out that a hoard from Mungret, Co. Limerick, datable to 953, is the only Irish hoard fromthe mid-tenth century which contains several cut halfpennies, and that this indicates a link with theIsle of Man, where hoards with cut halves are relatively common. See R. Hall, ‘A check list of VikingAge coin finds from Ireland’, Ulster Journal of Archaeology, 3rd series, 36/37 (1973/4) 71–86, atp. 75.

151F.M. Stenton, Anglo-Saxon England (3rd edn, Oxford 1971), pp. 369–71; ‘The Annals ofClonmacnoise’, s.a. 982[=989].2 (ed. Murphy, p. 160); ‘The Annals of Inisfallen’, s.aa.[1004]=[1005].5, [1064].7 (ed. & transl. Mac Airt, pp. 176/7, 222/3); ‘The Annals of Tigernach’,s.a. [988]=[989].3 (ed. & transl. Stokes, II.346); ‘The Annals of Ulster’, s.aa. 988[=989].4,1004[=1005].1, 1064.9 (edd. & transl. Mac Airt & Mac Niocaill, I.420/1, 434/5, 502/3);Chronicum Scotorum, s.aa. [987]=[989].2, [1003]=[1005].1 (ed. & transl. Hennessy, pp. 230–3,242/3).

152Smyth, Scandinavian York, I.61; Ó Corráin, ‘The vikings in Scotland’, pp. 336–7.153Chronicum Scotorum, s.a. [939]=[940].2 (ed. & transl. Hennessy, pp. 202/3), ‘Aralt mac .h.

Imair, .i. mac Sitric’; Ó Corráin, ‘Cathréim Chellacháin’, p. 54; cf. G. Broderick, ‘Irish and Welshstrands in the genealogy of Godred Crovan’, Journal of The Manx Museum 8 (1980) 32–8.

154This may in part be due to Victorian scholars’ obsession with saga and to the comparativelyslight attention given to Gaelic sources for viking-activity until recently: Andrew Wawn, The Vikingsand the Victorians. Inventing the Old North in Nineteenth-century Britain (Cambridge 2000);A. Wawn, ‘Hereward, the Danelaw and the Victorians’, in Vikings and the Danelaw, edd.Graham-Campbell et al., pp. 357–68; P. Holm, ‘Between apathy and antipathy: the vikings in Irishand Scandinavian history’, Peritia 8 (1994) 151–69; F.J. Byrne, ‘Ireland before the Norman invasion’,in Irish Historiography, 1936–70, ed. T.W. Moody (Dublin 1971), pp. 1–15.


both within the Scandinavian evidence and between Scandinavian and Insularsources, is to suggest that there were two persons called Eiríkr. One ruled York,and one ruled Norway. These later became confused in Scandinavian literature.The proposed solution may not be entirely satisfactory. It may be less satisfyingto admit uncertainty as to who Eiríkr of York was, but this may do greater justiceto the available evidence.155

155My thanks go to Lesley Abrams, Kristin Bornholdt, David Dumville, and Ralph O’Connorfor reading drafts of this chapter and providing useful comments and corrections. Since this paperwas first written (Summer 2000) much relevant work has been published, some of which has beenincorporated into the text and footnotes. Particular attention should be drawn to A. Woolf, ‘AmlaíbCuarán and the Gael, 941–981’, Medieval Dublin 3 (2001) 34–43, which is a helpful guide to theperiod covered by this chapter.


10Cold feet: King Magnús and the Irish Sea

THE expeditions made by Magnús, king of Norway, to the Irish Sea in1098×1103 represent an interesting episode in Insular history. His campaignsseem to have been intended to secure Norwegian control over the Hebrides whilechallenging the influence which the overkings of Munster wielded there.1

In 1102, Magnús arranged a year’s peace with Muirchertach Ua Briain,overking of Munster and soi-disant king of Ireland. In the thirteenth-centuryChronica regum Mannie et Insularum, s.a. 1098, there is a fanciful description ofMagnús’s actions following this treaty.2

Murecardo regi Ybernie misit calciamenta sue precipiens ei ut ea super humeros suos indie natalis Domini per medium domus sue portaret in conspectu nuntiorum eius, quatinusinde intelligeret se subiectum esse magno regi. Quod audientes Ybernienses egre ferebantet indignati sunt nimis. Sed rex saniori consilio usus, “Non solum”, inquit, “calciamentieius portare, uerum etiam manducare mallem, quam Magnus rex unam prouinciam inYbernia destrueret”.

‘To Muirchertach king of Ireland he sent his shoes instructing him to carry them abovehis shoulders through the middle of his house on Christmas Day in full view of his envoys,that he might understand from that that he had subjected himself to King Magnús.3Whenthey heard this, the Irishmen were greatly offended and deeply incensed. But the king,

1 A. Candon, ‘Muirchertach Ua Briain, politics and naval activity in the Irish Sea, 1075 to 1119’,in Keimelia: Studies in Medieval Archaeology and History in Memory of Tom Delaney, edd. GearóidMac Niocaill & P.F. Wallace (Galway 1988), pp. 397–415; E. Curtis, ‘Muirchertach O’Brien, highking of Ireland, and his Norman son-in-law, Arnulf de Montgomery, circa 1100, Journal of TheRoyal Society of Antiquaries of Ireland 51 (1921) 116–24.

2 Chronica Regum Mannie & Insularum, Chronicles of the Kings of Man and the Isles, ed. & transl.George Broderick (2nd edn, Douglas 1996), folio 35r. William of Malmesbury also exaggerated theinfluence which Henry I wielded over Muirchertach: William of Malmesbury, Gesta Regum Anglorum,The History of the English Kings, edd. & transl. R.A.B. Mynors et al. (2 vols, Oxford 1998/9), I.738/9(V.409); C. Downham, ‘England and the Irish-Sea zone in the eleventh century’, Anglo-NormanStudies 26 (2003) 55–73.

3 Christmas was a recognised occasion for demonstrating political relationships: cf. ChronicumScotorum. A Chronicle of Irish Affairs, from the Earliest Times to A.D. 1135; with a Supplement,containing the Events from 1141 to 1150, ed. & transl. William M. Hennessy (London 1866), pp.232/3 ([987]=[989]); Magnúss saga berfœtts, §1, in Snorri Sturluson, Heimskringla, ed. BjarniAðalbjarnarson (3 vols, Reykjavík 1941~51), III.210, and Snorri Sturluson, Heimskringla: Sagas of theNorse Kings, transl. Samuel Laing (3 vols, rev. edn by P. Foote & J. Simpson, London 1961–4), I.250.

The Journal of Celtic Studies 5 (2005) ISSN 1781-1406

209

adopting a more sensible approach, said, “Not only should I prefer to carry his shoes butshould eat them, rather than that King Magnús destroy one province in Ireland”.’

This ‘peculiar anecdote’ was designed to magnify Magnús’s achievements.4

However, it may also have been invented to explain Magnús’s epithet berfœttr(‘Barefoot’ or ‘Barelegs’). That is, the Norwegian king was rendered ‘barefoot’when he sent his shoes to Muirchertach. Although the connexion with Magnús’sby-name is not made explicit in the story, comparison may be made with otherexplanations of the epithet.5 Saxo Grammaticus published one.6

Qui cum sedulo Sueones adortus in Hallandenses etiam arma proferret, inopinata eorumirruptione perculsus, ut erat calceamentis uacuus, probrosum ad naues recursum habuiteiusque fugæ deformitatem etiam cognomina usurpauit.

‘While Magnús was vigorously invading the Swedes, he had also attacked the Hallandersbut was surprised while bare-footed by an unexpected charge which they made, and heretreated ignominiously to his ships, a sorry flight from which he took his by-name.’

The author of Heimskringla gave another story.7

Svá segja menn, at þá er Magnús konungr kom ór vestvíkingu, at hann hafði mjök þá siðuok klæðabúnað, sem títt var í Vestrlöndum, ok margir hans menn. Gengu þeir berleggjaðirum stræti ok höfðu kyrtla stutta ok svá yfirhafnir. Þá kölluðu menn hann Magnús berfœtteða berbein.

‘People say that when King Magnús came home from his viking-cruise to the Westerncountries he and many of his people brought with them a great deal of the habits andfashion of clothing of the western parts. They went about on the streets with bare legs andhad short kirtles and over-cloaks; and therefore his men called him Magnús barefoot orbareleg.’

4 S. Duffy, ‘Irishmen and Islesmen in the kingdoms of Dublin and Man, 1052–1171’, Ériu 43(1992) 93–133, at p. 122.

5 R. Power, ‘Magnus Barelegs’ expeditions to the West’, Scottish Historical Review 65 (1986)107–32, at pp. 122, n. 1, and 123, n. 4.

6 Saxonis Gesta Danorum, edd. J. Olrik et al. (2 vols, København 1931/57), I.342 (XIII.i.2);Saxo Grammaticus, Danorum Regum Heroumque Historia, Books X–XVI, transl. Eric Christiansen (3vols, Oxford 1980–1), I.108. The translation has been adjusted.

7 Magnúss saga berfœtts, §16: Snorri Sturluson, ed. Bjarni Aðalbjarnarson, III.229; SnorriSturluson, transl. Laing, I.269. For the authorship of Heimskringla, see M. Cormack, ‘Egils saga,Heimskringla, and the daughter of Eiríkr blóðøx’, Alvíssmál 10 (2001) 61–8.


Such onomastic play was common to both mediaeval Scandinavian and mediaevalGaelic culture. Each tale also has its political spin. The account of SaxoGrammaticus emphasised Magnús’s cowardice in war against the Swedes, whereasHeimskringla and Chronica regum Mannie et Insularum give attention to hissuccesses in the Irish Sea. In each case the by-name was used to substantiate arather exaggerated story.

One of these tales may be nearer the truth than the others; however, the realorigin of Magnús’s epithet eludes us. Nevertheless, by interpreting the Manxaccount as an onomastic legend, one can explain some features of this story. It alsodraws attention to the significance of Muirchertach and Magnús in Manx history.Magnús planned the marriage of his son Sigurðr to Muirchertach’s young daughterin 1098. He thus hoped to unite two royal lines and secure Norwegian influencein the Hebrides. These hopes were not fulfilled. Magnús was killed in Ulster in1103; his son Sigurðr hastily returned to Norway, leaving his child-bride behind.

KING MAGNÚS AND THE IRISH SEA 211

No Horns on their Helmet? Essays on the Insular Viking Age

Documents

Transcript of No Horns on their Helmet? Essays on the Insular Viking Age