Uses of Wodan The Development of his Cult and of Medieval Literary Responses to It Philip Andrew Shaw

Submitted in accordancewith the requirements for the degree of Doctor of Philosophy The University of Leeds Centre for Medieval Studies December zooz

The candidate confirms that the work submitted is his own and that appropriate credit has been given where referencehas been made to the work of others This copy has been supplied on the understanding that it is copyright material and that from be the thesis may published without proper acknowledgement no quotation

Abstract
Scholars working on Germanic pre-christian religion have generally considered Wodan to have been a deity of considerable importance to most if not all Germanic failure however, based This is, to approach the available tribes. on a understanding This for Wodan thesis reevidence within appropriate contemporary contexts. Germanic building a model of the general nature of contextualises the evidence, therefore, heathenisms in the Migration Age, within which the cult of Wodan can be located. Set its due to social, political and religious against this model, and with consideration given beginning be for Wodan of a christian reseen as the contexts, the earliest evidence can imagination of this deity. A plausible model of Wodan's cult is established, which sees first half being limited, this cult as and originating probably within the geographically be Obinn Common Era; the cult of millenium of the would appear, moreover, to Furthermore, from Wodan. development in that of a complex set of substantially separate have Wodan shaped subsequent eighth-century scholarly re-uses of are shown to day. deity, both in the medieval period and up to the present understandings of the Having considered how the traditions of eighth-century scholarship have misled

further development of these traditions modern scholarship, the thesis then examinesthe in Anglo-Saxon England. In this context, Wodan assumes still more various guises,and is conflated with Obinn, thus helping to cement modern scholarship'sbelief in the figures. This by is process strengthened, moreover, the strong original unity of these two influence which Anglo-Saxon England exerted on Scandinaviaboth around the time of Scandinavia Scandinavian the conversion of and at the period when much of the extant down. mythography was written This Scandinavianmythography is examined briefly in the final chapter, which important points out some areasof misreading of pre-christian mythology in thirteenthScandinavian for century mythography, aswell as arguing substantial extra-Scandinavian influences on such mythography. This leads, finally, to a consideration of how Obinn little for him. in appears what certainly pre-christian evidenceexists

Acknowledgements
I should like to expressmy gratitude to the many individuals, groups and My helped have supervisors, organisations who me along the way with my thesis. ProfessorsJoyce Hill and Ian Wood, and, latterly, Dr Mary Swan have helped me Swan has been Dr a enormously, guiding my researchand putting up with me patiently. Leeds. during in constant supportive presence my time books in I am also grateful to the libraries which have allowed me access the to Libraries School English Boyle Brotherton, Edward at course of my research:the and of Archaeology University Main Institute British Library; Leeds; the the the and of of Libraries at University College London, University of London; Goldsmith's College Library, University of London; the John Rylands Library at the University of Manchester. Studies for have Centre Medieval The students and staff associated with the
I provided me with a wonderful community within which to work, and, although cannot list them individually here, they have all helped me to develop and formulate my ideas, Nor have been is their time this the only and always and expertise. generous with from I found I have like to thank all the scholarly community which support; would also Texts Identities have taught me much about the early and members of the group, who Middle Ages and non-Insular cheeses.

My parents have bankrolled a substantial proportion of this undertaking, with the Arts and Humanities ResearchBoard providing the other part. I am grateful to both, but immeasurably more so to my parents, since I could not have done this, or indeed My for love, I them. thanks to them their anything, without very great trust. support and hope, but do not expect, that I have not omitted anyonewho has helped me in the last half `and left I've / best, three and a the that years: all rest out, you were the without a doubt' (Richard Rodney Bennett). Being stubborn, I have not alwaysfollowed the sage have (except I been bear advice given; none of the above perhapsmy parents) any for bear my stubbornness,and consequently,none of them responsibility any for in any errors which remain the thesis. responsibility

Table of Contents
Note on Terminology i .......................................................................... Preface .................................................................................................. iii

Previous Scholarship i. I ............................................................................ Introduction i. i i ...................................................................................................... The Beginnings Modern Pre-Christian Religions Scholarship Germanic i. 2 of on
................................................................................................................ I

Gods Norsemen: Recovering Edda Saga in Nineteenth Century4 the I. 3 the of and Attempts Altgermanische Systematising Religionsgeschichte: Histories I. 4 at an of Heathenism 6 ............................................................................................ Volk Folklore: Reading Modern Reflexes Heathenisms 8 1.5 and of ........................ Heathenism, Nationalism Racism: Second World The War 1.6 and Io .................. Religion Society: Linguistic-Religious Fallacy Dumezil I.7 within and the 12, ........ The Age Improbable: Directions Scholarship from in 1.8 the 196osto the of the Present 13 ................................................................................................... Understanding Literary Evidence: Re-Using Re-Imagining Wodan 1.9 and 16 ...... Conclusion I. IO .................................................................................................... 17

Models Methodologies 2. and ................................................................

Iconography z. z. i ........................................................................................

Introduction z. i i9 ................................................................................................... Approaches Methodologies 2.z and 23 .....................................................................

Toponymy 2.z. z ......................................................................................... 27

19

2,3

Conclusion 2.4 .....................................................................................................

Philology 2.2.3 ........................................................................................... 35 Modelling Heathen Cults Societies 2.3 and 42......................................................... Patterns Germanic Pre-Christian Religion 2-3-1 of 43 .................................... Syncretism Exoticism in Germania: Patterns Religious, Cultural 2.3.2 and of Literary Exchange and .................................................................So
7o

Classicising Heathen: The Invention Wodan Eighth-Century in the 3. of Intellectual Milieux 72 .................................................................... Introduction
3.1

Thunaer Uuoden Saxnote: Imagining Saxons Frankish in Circles8z 3.3 the ende ende Mercurium Woden: The Corpus Glossary Frankish-Kentish Classicism 3.4 and 92 Uoden, de 3"5 cuiusstirpe multarum prouinciarum regiumgenusoriginem duxit: Bede Use Woden Anglo-Saxon in Genealogies of and the 97 ............................ Vodano Columbanus, Suebi, Cult Wodan the the 3.6 nomine: and of ios ...............

72, .................................................................................................... Refert hoc loco Paulus Diaconus Heathen 3.2. antiquitas ridiculamfabulam: and the Antique 72, ..................................................................................................

Conclusion 3.7

Cult.. Alamannic Brooch Nordendorf The Wodan: Logapore io6 and 3.6.1 Ethnogenesis Langobard The Wodano: Quem fanatice as nominant 3.6.2 Evidence for Cult III ........................................................................ Cult Columbani Sancti Vita The Mercurius: Identifying the of and 3.6.3 Wodan 118 ........................................................................................
................................................................................................... 127

Late Complex in Woden-Obinn The Figures: Authority Ancestors 4. and 129 Anglo-Saxon England ..............................................................

Introduction i29 4.1 ................................................................................................. O lion Uses inn.... Anglo-Saxon Late Denisc: 132of 4.2 gehateno6rum naman on His Circle Wulfstan 138 4.Z.! and ................................................................... Culture.... Literary iElfrician Post Deities Danish in iElfric: After 147 4.2.2 Figure Magico-Medical Woden Da Woden VIlli iso as wuldortanas: 4.3 3enam .... Charm Merseburg Second The 15o 4.3.1 .......................................................... Charm The Nine Herbs 156 4.3.2 ................................................................... False Creator Woden Woden 16o as 4.4 worhte weos: ............................................... Conclusion 169 4.5 ...................................................................................................

171 Drinking in the Past: Snorri, Saxo and the Invention of OSinn .......

Introduction 171 S. I ................................................................................................... O Verse Role Skaldic Central in inn's Fagnafundr Friggjar 177 5.z ni5ja: ................ Heathenism Creation Influences Hpkon External for hei in of and the 5.3 god: me6 in Skaldic Verse 19z ................................................................................... O Asgard Classical Hero Troy: Re-inventing inn zoo as a 5.4 and ........................ Gesta Danorum By-Names Obinn's in the z16 5.4.1 ....................................... Danorum Gesta Exiles Obinn's in 9 the 5.4.1 .............................. Gesta Danorum Physician Obinn in the 211I 5.4.3 as ....................................... From Saxo Snorri Skalds to zzz 5.5 and ................................................................... Obinn Cult U1fuR Traces of zz4 Auk upin Auk HutiuR: of the 5.6 .......................... Conclusion 2-30 5.7 ...................................................................................................

6. Conclusion: Reading and Writing Wodan 233 ....................................... Bibliography 237 ..........................................................................................

List of Illustrations
Figure i- the Nordendorf brooch ion ........................................................................... Figure z- the Ribe skull fragment 225 ..........................................................................

List of Tables
Diagram of Primitive Germanic to Proto-Scandinavian Sound Change 37 .................... Diagram of French-English phoneme equivalents 40 .................................................... Table of Roman Coin Finds in Scandinavia 56 ............................................................. Table of Germanic Individuals in Roman Service in the Fourth Century CE........... S9 ............................................................................... Table of Manuscripts of /Elfric's De FalsisDeis 147 ....................................................... Table of Early By-Names of Obinn 179 ........................................................................ Germanic Charon's Obol Finds 66

A Note on Terminology
A number of terms appearin this thesis with a more specific meaning than that The discuss It be briefly helpful, them. term to therefore, normally attached to them. will `heathen' is synonymouswith `pagan'in quotidian usage,but is here used to refer to nonchristian religious expressions and manifestations originating specifically and uniquely Germanic `Pagan', is in tribes among or peoples. contrast, used to refer to non-christian be The punor Germani. religions not originating solely among cult of might therefore described as heathen, but the cult of Mars Germanic by evenwhen practised would

individuals, and even when Mars is conflated with a deity of Germanic origin' be described as pagan. It will also be noted that the term `myth', although popular in previous

This here. is imprecise, term scholarship, appearsrarely and therefore usually mythology be draw distinction or mythography will used, since these terms allow us to a clear between oral and written myths. In a few cases, however, in which both mythology and `myth' is to, mythography are referred used. The forms of namesused should also be discussedhere. The names of Germanic deities appearin different forms in different Germanic languages(on which seesection below). The forms in be Old z.2.3, texts tend to most widely used modern normalised Icelandic forms (Obinn, prr), normalised Old English forms (Woden, punor), modern German forms (Wotan, Donar), or some common modern English forms, often basedon Old Icelandic forms (Odin, Thor). The last two groups mentioned are problematic becausethey do not alwaysprovide forms for some of the lessfrequently-mentioned deities (e.g. Ullr, Forsete), and their uniformity createsthe misleading impression of in deity. Normalised forms uniformity medieval use and understanding of the of medieval least in forms for the names are more useful that they at versions of provide standardised

Winter,

1 Both these conditions are evidencedby the inscription from Housesteads, Hadrian's Wall, to on `Mars Thincsus' (seeKarl Helm, Altgermanische Religionsgeschichte, GermanischeBibliothek, i. S. i,; Bibliothek, Religionswissenschaftliche [vol (Heidelberg: in Carl also S, 2 vols two z, parts]
1913-1953), 1,366-67)-

ii

It be impractical deities, but the names of all to quite the second criticism remains. would Wodan in discussing in that text; not the presentation of adopt the usageof each text but in between texts confusing, only would this render comparisons any casemany individual medieval texts spell a given name in severaldifferent ways. One can broadly distinguish between the presentation of heathen deities in Scandinavianmythography, their presentation in Anglo-Saxon England, and their presentation outside England and Scandinavia(which last is generally early, and is therefore often thought to provide some for Wodan). The here be the therefore, to refer to evidence cult of usageadopted will,
i

Obinn (normalised Old Icelandic) when dealing with Scandinaviansources,Woden (normalised Old English) when dealing with English sources,and Wodan (the form Wodan, Nordendorf in inscription the which appears the earliest text to mention on the brooch; this is discussedin section 3.6.1below) when dealing with continental sources. This indicates the variety of spellings and pronunciations of the name without creating The be far undue obscurity. samerules will adopted,as as possible,with the names of Germanic (e. deities Scandinavia, England prr for punor for ponar for other the g. and When Wodan he in talking about continent). than generally,rather specifically as appears Scandinavia,England, or on the continent, the form Wodan will be used.

111

Preface
The study of pre- and non-christian religious expressionamong the Germanic if forth has brought tribes some of the most various and exciting, not alwaysthe most has deity One in attracted a great particular plausible, scholarship of the modern period. deal of attention, namely Wodan. The extraordinary variety of evidenceevidenceor supposed

for the cult of Wodan and for mythological narratives relating to Wodan has

done much to create this situation. The huge geographicaland chronological range of The developments. in has methodologies and such evidence also played a part these been have however, in well-defined not always approachesapplied studying this evidence, indeed, capableof producing plausible results. or, Many studies of heathen religion and mythology have failed to set out their lack from have Moreover, they a of attention to the often suffered methodologies. from failures found in to their evidence,and geographical and chronological patterns identify all the useful evidenceor to exclude evidencewhich cannot safelybe used. This has resulted in the development of a modern, scholarly picture of Wodan which bad, into indifferent a and representsa monolithic construction of evidencegood, Germanic figure have believed world pervadedthe pre-christian somehow to composite from the Iron Age through to the conversion of Scandinavia. The present study seeksto dispersethis monolithic figure. It sets out and for it discusses the the methodologies which employs, and these methodologies allow development of a very different picture of Wodan. In fact, the starting point for this study is the understanding that we cannot have a picture of Wodan; Wodan is not a single developing for but figures in various contexts, phenomenon, a complex of various reasons. We must allow for the possibility of chronological and regional variations. We must also does his Wodan in for heathen that the possibility not alwaysappear capacity as a allow deity. With this in mind, I will now outline the scopeand aims of this study. It seeksto Wodan for development figure, by the as a cult of asking provide a plausible new model When Wodan deity, following questions: and where was actually worshipped as a the and

iv

by whom? How did his cult arise, and how did it changeand develop throughout its lifespan? What were the causesof this development?One must also recognise, however, Wodan figure in has that appearing christian contexts; an extremely vigorous afterlife as a in fact, this study arguesthat he was probably more important, in much of the Germanic The following before. area,after christianisation than questions are therefore also posed: What forms does Wodan take at different times and in different places?How and why is he re-used and re-imagined in christian contexts?Can one trace patterns in these developments, and connections between manifestations of Wodan? To begin answering these questions,we would do well to consider briefly some Wodan; is, the previous scholarship on that post-medieval readings and representationsof Wodan.

Scholarship Previous i.

Introduction i. i
Obinn, Wodan, also known asWoden, Godan, Up in and by a host of other is Germanic Not his heathen deity does names, the most versatile only name ever created. from but vary place to place, time to time, author to author, and even within a single text, his In he also usesvary enormously. some contexts appears as a war-god, or as a god of his he king in is human trickery; name others portrayed as a or magician; sometimes be This nothing more than a vaguely-definedword with magical connotations. seemsto is as true of modern representationsof Wodan as it is of medieval representations.This thesis is circular; in examining the medieval representationsof Wodan, one must begin discussion Such is, these with a of previous scholarshipon representations. scholarship however, merely a continuation of the processes Wodan of reading and representing be in the examined this which created medieval traditions and representationswhich will Wodan. The indeed, is itself thesis; this thesis sheer volume yet another representationof Wodan, heathen of post-medieval scholarship relating to religious expressionand and to its medieval literary representations,precludesa detailed examination of any substantial This limited here. in surveywill, spaceavailable proportion of the scholarship the broad therefore, examine the trends of such scholarshipwith referenceto certain specific (although be examples many others could mentioned).

The Beginnings Modern Scholarship Germanic Prei. 2 of on Christian Religions

Acknowledging that, as argued above,any attempt to pinpoint the origins of heathen deities, is their their cults and representations, merely to modern scholarship on Olaus dub Magnus's Historia Gentibus let de Septentrionalibus us point to an arbitrary text,

first First Rome deities. in heathen the published at 1555, piece of modern scholarship on heathen deities; is historiographical first this work admittedly not the work to treat of Langobardorum, distinction Origo Gentis both Chronicon belongs Fredegar's that to or the discusses first CE. Nor historiographical is it the the work which of the seventh century heathen for deities origins of and specifically attempts to synthesisethe author's sources heathenism into a more or less coherent picture; Saxo Grammaticus and Snorri Sturluson Snorri Magnus. Saxo from direction before Both in made efforts this quote centuries and both their sources,moreover, and the validity of their make some attempt to assess Magnus fact in follows his sources. sourcesSaxo looming large among these very

division his between the closely, and work and that of the thirteenth-century is division mythographers really an arbitrary chronological which may reflect other developments Europe between in the thirteenth and sixteenth political and social heathen but fit does development centuries, of traditions concerning which not easily the first Magnus, original work of the early religion and mythology. nevertheless,provides the (the Saxo's first heathen modern period on printed edition of religion and mythology GestaDanorum also belongs to this period, appearingin 1514), his work continued to and be influential into the nineteenth century.' What, then, does Magnus have to sayabout Wodan? He repeatsinformation
from Adam of Bremen's description of the temple at Uppsala but ascribes it to loannes Magnus, Archbishop of Uppsala; draws in Saxo's euhemeristic portrait of Obinn; derives `Odens dag' from Obinn; and suggests that Jordanes's Gothic Mars was Wodan and was ' Other human from Saxo Obinn offered sacrifice. material also appears, portraying as a (pp. Magnus did know soothsayer IIS, I22-23). all the sources which are now available not (if but he knew to scholars, several of the richer not necessarily more reliable or useful)

' On the date of the first printed edition of the GestaDanorum, and Walter Scott's use of Magnus Olaus, seeAndrew Wawn, The Vikingsand the Victorians:Inventing the Old North in On Nineteenth-Century Britain (Cambridge: Brewer, woo), pp. 17-18 the relationship of and 71. Danorum Saxo's Ivan Gesta Boserup, `The princeps the to the editio of original composition, see Angers Fragment and the Archetype of GestaDanorum', in Saxo Grammaticus:A Medieval Norse and Latin Culture, ed. by Karsten Friis-Jensen (Copenhagen:Museum Author between Tusculanum Press, i98i), pp. 9-26.
2 Olaus Magnus: Historia de Gentibus Septentrionalibus: Romae iffy, introduction by John Granlund (Copenhagen: Rosenkilde and Bagger, 1972), p. 100.

information He Saxo's Danorum. Gesta ones, such as makes some attempt to order the With his his from him, draw availableto conclusions, sources. and to conclusions however, scholarship on heathen religion, and particularly on Wodan, was already faltering. The interpretation of Jordanes'sMars asWodan created a new Wodan who had Wodan day the not existed previously, and who remains to this of most popular deity, This Wodan is the the pan-Germanic conceptions and much scholarship. bloodthirsty war-god, demanding human sacrifice. By establishing Wodan as a deity known to the Goths, Magnus set up an orthodoxy in which Wodan must have existed (many Gothic Germanic if tribes prior to the scholars migration, and among most not all for firm but is there no still assumethat this was the case, evidence this understanding of The Wodan; below). the geographicalspreadof the cult of suggestion seesection z.2.3, human Wodan, depictions became in of sacrifice also of although a commonplace modern Wodan human late involved the evidencethat the cult of sacrificerests solely on relatively Ecclesiae Scandinavian Bremen's Gesta Hammaburgensis Adam medieval sources,such as of ' As Pontificum and Old Icelandic sagas. we shall see,such sourcesprovide very poor (see for heathenisms below). evidence extra-Scandinavianearly medieval section Z.i, Magnus's method of reading a Roman pagandeity-name as equivalent to a heathen deity, highly is moreover, set a undesirable methodological precedentwhich often unthinkingly 4 followed to this day. It is true that medieval texts had equatedheathen and pagandeities CE, but in the these since seventh century equations arise the context of specific attempts between heathen draw deities, deities to parallels paganand or even to argue that these

3 GestaHammaburgensis Pontificum, in Quellendes Ecclesiae Jahrhunderts Geschichte 9. und ii. zur Kirche und desReiches, G. Waitz, H. der hamburgischen by B. Schmeidler, Bresslau G. ed. and Pertz, trans. by Werner Trillmich and K. Nobbe, Ausgewhlte Quellen zur deutschen Geschichte des Mittelalters, ii (Berlin: Rtten & Loening, i96i), p. 472 (book 4, chapter 27). Seealso the Starkabr's King V'karr Obinn Gautreks in in to well-known narrative of sacrificeof saga, Fornaldar SgurNor6urlanda, ed. by Gubni Jonsson,4 vols (Reykjavik: Edda, 1950),Iv, 30-31.
4 See, for instance, H. R. Ellis Davidson, Pagan Scandinavia, Ancient Peoples Places, and 58 (London: Thames and Hudson, 1967), p. 7z; David Wilson, Anglo-Saxon Paganism (London: Routledge, 1992), p. 23; Hilda Ellis Davidson, The Lost Beliefs of Northern Europe (London: Routledge, 1993), p. 79; Rudolf Simek, A Dictionary ofNorthern Mythology, trans. by Angela Hall (Cambridge: Brewer, 1993), pp" 211-I2; Lotte Motz, The King, the Champion and the Sorcerer: A Study in Germanic Myth, Studia Medievalia Septentrionalia, I (Vienna: Fassbaender, 1996), p. ioi; Richard North, Heathen Gods in Old English Literature, Cambridge Studies in Anglo-Saxon England, zz (Cambridge: Cambridge University Press, 1997), p" 79"

deity-name identical. Such however, do were prove that the use of a pagan attempts not, felt had for if heathen deity; medieval writers alone could refer to a quite the opposite, that they could use a pagandeity-name its own, and without explanation on to

draw between, have heathen deity, parallels or they would scarcely refer to a attempted to for deities. heathen identity the argue of, paganand

Saga Gods Norsemen: Recovering Edda in the 1.3 and of the

Nineteenth Century

So much for the father of modern scholarshipon heathen religion and mythology.
Magnus's implicit methods and assumptions are still to be found in recent scholarship, but the next major change in scholarship on heathen religion and mythology came with the modern dissemination of the Old Icelandic sources which were, apparently, not used by Magnus. The nineteenth century saw a rash of publications of syntheses of Old Icelandic mythography. A typical example is R. B. Anderson's impressively-titled Norse Mythology: or, The Religion of our Forefathers, containing all the Myths of the Eddas, less impressive `that interpreted, it is the claim no systematizedand whose preface makes English first Norse language i5 in presentation of the the completeand systematic mythology the In Anderson's work are reflected the nationalistic and romantic ideals of his day. Describing Longfellow as `the Nestor among American writers on Scandinavian themes' (p. ii), he goes on to invent heathens as noble savages,`[nearly] in sympathy with Christianity (p. z, full `The in touch with nature and 7), old mythological of poetry:

Norsemen in abound stories of the poetry of the truest and most touching character. These stories tell us in sublime and wonderful speech of the workings of external nature' (p" 33)"

5 R. B. Anderson, NorseMythology: Or, The Religion of Our Forefathers, Containing All the Myths (Chicago: Interpreted, London: Systematized Griggs; Eddas, Trbner, and znd edn, 1876),p. of the follow in 8 (Anderson's italics). Further referencesto this work in the paragraphs which are given in parentheses the text.

The nationalistic side of Anderson's views is perhapsmore disturbing to a modern He harangue begins first his reader. against the use of non-Germanic chapter with a loanwords in English: `we havewatered our mother tongue long enough with bastard Latin; let us now brace and steel it with the life-water of our own sweet and soft and rich (p. ENGLISH' and shining and clear-ringing and manly and world-ranging, ever-dearest This `world-ranging in is z3). strangely echoed,moreover, the opening paragraphof Anderson's section on Wodan (Anderson refers to him as `Odin'), in which he claims that deity's Wodan `is derived from is `to the the allthe verb vada walk', and that name pervadingspirit of the world' (p. zis; Anderson's italics). It is curious that Anderson's linguistic nationalism allows him here not only to use, but even to stress,the Latin loanword `pervade',but, leaving this oddity aside,Wodan pervading the world seemsto English Anderson takes the euhemerism of recall ranging the world; the more so, since Saxo and Snorri at facevalue, but credits his `historical Odin' (pp. z3z-136) with teaching `the art of poetry to the Norsemen' (p. z35).Anderson's Wodan is a euhemerised historical hero, not only of inward-looking linguistic nationalism, but also of nationalism (p. Latin-speaking Rome kind, directed of a more militant 2-32-)against These are strands which prove common enough in the nineteenth-century handbooks of Old Icelandic myth. Nationalism and Romanticism combine in much of Germanic idealised this material to produce an vision of pre-christian religion which takes Icelandic by Old evidence,and which no account of the chronological problems posed the is, in many ways,not dissimilar to Tacitus's equally unreliable presentation of the Germani as noble savages.

The nineteenth century also saw,by contrast, the beginnings of scholarly efforts disentangle developments from to the the of pre- and non-christian religious expression H. Chadwick's M. Cult complex chronological and geographicalspreadof evidence. of Othin representsan important example of this trend, and one which deserves a special large-scale directed in study specifically at the cult of place this survey as a relatively

Wodan and its development.6 To this day, it remains an admirably complete and wellfundamental Unfortunately, its argued piece of scholarship. premisesand assumptions are quite untenable, and these insecurefoundations prejudice the value of the majority of the book's conclusions. Chadwick makes extensiveuse of sagamaterial to build up a picture of Wodan, 7 and, using this picture as a guide, then seeksto establish where, when and how ' Chadwick's Wodan was culted in Germania from the first century CE onwards. him but his fail arguments are generally well-constructed, completely; assumptions Saxo in Grammaticus's Gesta Danorum, and a few ambiguous skaldic bymaterial sagas, for Chadwick names suffice to view hanging as a preferred method of sacrificing to Wodan. 9 The argument is plausible, but only if one accepts(and few scholars now do) that all of these sourcesconstitute reliable evidencefor pre-christian religious practice. On this basis Chadwick goes on to argue that all instancesin ancient writers of Germanic 1 hanging As Wodan. is tribes people are evidenceof those tribes culting usual, moreover, Chadwick reads classicalreferencesto Germani worshipping Mercury (and, ll These Ares) Wodan. flaws contradictorily, also to the asreferences render worship of Chadwick's conclusions almost worthless, although his work does assemblemuch of the important for Wodan, for later textual evidence the cult of that most and some of Wodan. representationsand usesof

Attempts Altgermanische Religionsgeschichte: 1.4 at an Systematising Histories of Heathenism

Chadwick's work is notable also as an attempt at a more systematic study of heathen religion, which attempts to contextualise mythology within cult practice. His

' Chadwick,pp. 3-2-8-

6 H. M. Chadwick, The Cult Othin: An Essay Religion in Ancient North (London: the the of of 1899). Clay,

'Chadwick, pp. 2-9-48'Chadwick, pp. 16-2o.

oChadwick,PP-35-39Chadwick, pp. 30 and 53-54.

difficulties, but his reliance on medieval mythography presentsconsiderablechronological does from the common nineteenth-century work represent a significant move away tendency to view pre-christian religion largely in terms of anachronistically well-ordered leads into This tendency, moreover, a new and still useful approach mythological cycles. heathen heathen to mythology and cult practices, religions, which attempts to synthesise both heathen viewing as manifestations of religions. This approachis broader in its use of evidence,drawing in both medieval late historiography hagiography mythography and the classicaland early of the and heathen is The broaden bases the evidential cult of studies of medieval periods. move to focus in increased also reflected an on archaeological,toponymic and epigraphic evidence. To this movement towards a more broadly-basedapproachto studying Religionsgeschichte, by heathenisms belong two important works entitled Altgermanische one Jan de Vries, the other by Karl Helm. 12Both works were published in two volumes over final instalment, the the course of the earlier part of the twentieth century, with the Helm's Both in scholarsrecognisedthe second part of secondvolume, appearing 1953. for Germanic North disjunction between for the the evidence cults and that problematic Helm beliefs Germanic tribes, although religious and practicesof the extra-Scandinavian North Germanic his never published the projected third part of secondvolume, on They between took early archaeological religions. also account of the chronological gap hand, historiography and the medieval evidenceand classical and ethnography, on the one hand. sources,on the other This understanding of the complexities of the subject is important, and continues These be history to so. grand projects of religious are still valuable as referenceworks, drawing together prodigious volumes of evidencefor pre-christian religion among the Germani. The interpretations of this evidencethat are put forward in these works are, however, not alwaysso useful. Although their accountsare often plausible, they are also by drawing injudicious between disparate the of often marred connections evidence.

12 Religionsgeschichte; Jan de Vries, Altgermanische Helm, Altgermanische Religionsgeschichte, Grundriss der germanischenPhilologie, iz, z vols (Berlin: de Gruyter, 1935-1937); Helm's seealso Wodan: Ausbreitung Wanderung Kultes, Giessener Beitrge study, und seines more specific zur deutschen Philologie, 8S(Giessen:W. Schmitz, 1946).

Thus, although some of the major problems of non-contiguous patterns of evidenceare fall in recognised these works, they still prey, when examining phenomena on a smaller hardly be between scale,to making unjustified associations piecesof evidencewhich need heathenisms Germanic This from just in related. as perhapsresults part a sensethat, forms be large-scale, in of religious could examined monolithic publications, so these homogenous themselves expressionwere more and monolithic than we might now 13As the Germanic but histories suppose. sharedtitle says,these were religions, of not of Germanic religion. Despite their shortcomings, however, these ambitious attempts at a total history heathenism but information, of are not without their usefulness,not only as a mine of because lay builds foundations they also which much subsequentscholarship substantial The developments focussed on and modifies. which more studies of particular regions and have followed these works owe much to them for beginning to disintegrate the myth of a homogenous heathenism, and at the sametime the amassingof evidence,undertaken in heathenism has history for to an attempt create a as a whole, capableof accounting build. if provided a valuable, sometimes simplistic, model on which to

Volk Folklore: Reading Modern Reflexes Heathenisms of 1.5 and

The attempt to broaden the evidential basesof studies of pre-christian religions did not seek simply to bring into consideration under-explored classicaland medieval Another began be to evidence. possible source of evidencewhich considered was modern folklore, which, it was thought, could provide evidencefor heathen cult beliefs and be if it folklore from practices, could establishedthat particular elements of stemmed preThis is for times. christian a perfectly reasonableposition to take, save the fact that it can determined degree be folklore has with any never of certainty that modern such ancient

13Such a position was not without its early critics, however,such asWilhelm Boudriot, who both difficulties is the of unconnected evidence, and of evidencewhich essentially points out dependent on literary sources,in the prefaceto his Die altgermanische Religion in der amtlichen

belief did begin Occasionally be origins. not one can reasonablycertain that a practice or does but later, until the medieval period, or even this not prove the antiquity of those begun have in beliefs be practices and or after the medieval period. shown to which cannot Only unequivocal contemporary evidencecan prove the antiquity of a cult element, and, hardly is later where such evidenceexists, recourseto necessary. evidence The folkloric method is foreshadowedin Benjamin Thorpe's impressive focuses folk Germanic beliefs, but this text on compendium of nineteenth-century description more than analysisof such beliefs.14For the development of a genuine folkloric method we must look a little later. An important application of the folkloric der Geheimbnde Hfler's Kultische Wodan Otto in method to the cult of occurs ls It does Germanic Germanen. implausible tribes, military that, among some not seem " Hfler (and, indeed, Wallacespecialist groups with an exclusivemembership existed. beliefs Hadrill) may be correct that such groups had some religious observances specific or however, This, is is but it is hard the them, to to establish certainly that this case. Hfler late folkloric does, maleprecisely what medieval adducing as evidencemodern and Wild Robin Hood Hunt, flavour, from through the only groups with a martial ranging 17 Morris dancers. and company, to The results of such a method are vivid, but hardly uniformly convincing. Nevertheless, less extreme arguments from folklore are often accepted.Wodan's supposed leadership of the Wild Hunt, for instance, continues to appearin some accounts of his

Jahrhundert (Bonn: Rhrscheid, Ludwig bis Abendlandes kirchlichenLiteratur des zl 1964), vom y pp. -8.
Northern Mythology: Comprising the Principal Popular Traditions and Superstitions of Scandinavia, North Germany and the Netherlands, 3 vols (London: Lumley, 185IThorpe, 14Benjamin

1852), II and in.

is Otto Hfler, Kultische Geheimbndeder Germanen (Frankfurt M.: Diesterweg, Moritz a. 1934)" 16See, for instance, J. M. Wallace-Hadrill, The Long-Haired Kings and Other Studies in Frankish History (London: Methuen, 1962), pp. 154-55-

" Hfler, pp. 36-37,48-49. De Vries, alsowriting in the 1930s, in dealing urged extreme caution `when knowledge folklore that an accurate evidence,concluding we possess of the origin of a with development downwards, but line from tradition trace the to we may of popular seek modern known folklore is to the to us, seemsto me a way a sourcewhich only superficially modern in His Relation to fruitless task' (Jan de Vries, Contributions to the Study of Othin Especially Agricultural Practicesin Modern Popular Lore, Folklore Fellows Communications, 94 (Helsinki: 63). SocietasScientiarum Fennica, 1931), p.

10

18The folkloric be cult. method should perhaps seenwithin the wider context of 1930s German nationalism, with its development of the neo-paganreconstructive religiosity of Jugendbewegung, lifeblood farmer German its the of as the peasantor and emphasison the Second dead-end in the nation; an emphasis,moreover, which cameto a chilling the World War. At the sametime, a folkloric approachto pre-christian religion has had less for formed basis Britain, it in the unpleasant popular reflexes,as post-war another where 19 Gerald Gardner. founded by Wicca, known reconstructive religious movement as

Heathenism, World The Nationalism Racism: Second 6 i. and War

In 1936,looking out from a relatively untroubled Switzerland at the increasingly Germany, `Wotan'; life Carl Jung stormy political of an wrote an essayentitled, simply, be ten essaywhich, yearson, appearedto as much prophecy as reflection on contemporary `Wotan', Ergreifer Jung he is `is troubles. really the only explanation of men and argues, an [for National Socialism], unlesswe wish to deify Hitler i20.Not that Jung literally saw a National Socialism, deity he his rather one-eyed with spearand ravensas the root causeof Wodan understood as a personification of the psychologicaland social patterns and imperatives which, in his view, drove the development of National Socialism: Wotan is a fundamental characteristic of the German soul, an irrational, factor, high like psychic pressureof civilization which acts a cyclone on the ] [... Wotan Germanic factor, blows it and away representsa primeval and [...] is the most accurateexpressionof a basic human quality which is 21 German. particularly characteristic of the

18 For examples,Motz, pp. 78-79. 19 Gerald B. Gardner, Gardner sets out his ideasin two monographspublished in the 195os: Today (London: Rider, 1954)and G. B. Gardner, The Meaning of Witchcraft Witchcraft (London: Aquarian Press,1959)" C. G. Jung, Wotan', trans. by BarbaraHannah, in C. G. Jung, Essays Contemporary Events, on (London: Barbara Hannah Mary Briner Welsh, Kegan Paul, Elizabeth by and trans. 1947),pp. I(p" 7)16 21 Jung, p. 8.

11

This, in its own right, constitutes a scholarly approachto the study of pre-christian figurative be religions, arguing that they can all representations understood essentiallyas human be Such of psychology through myth. seenas a reasonableenough a method could human form for that matter, any other way to read religious practice, or, of endeavourat One human by human is all, since all psychology. should simply shaped activity necessarily Switzerland however, in the the early that the psychology of the citizens of note caution, twentieth century may well have been quite dissimilar from that of early medieval Germani; attempting to read from human activity to human psychology is difficult but human is best from human not unreasonable,whereasreading activity at psychologyto useless. What is remarkable about Jung's essay, in fact, is not his central psychological `psychic National Socialism, Wodan but his the reading of recognition that, although factor' acted essentiallywithin the sphereof the irrational, even the intellectual life of Germany was profoundly influenced by National Socialism. In few areasof researchis this Germanic in heathenisms, it is than more apparent studies of precisely this areaof and Jung himself his but in study which picks out, not only obliquely the title of essay, also 22 his Schicksalsglaube. in Martin Ninck's Wodan specifically comments on und germanischer Ninck is, Jung avers,one of those men ergriffenby Wodan the Ergreifer, and this is 23 his in abundantly clear work. Hitler's fondness for the operatic works of Wagner has often been remarked; 24 Wagner's These indeed, has popularity artistic reworkings of suffered as a result. Germanic mythography were not, however, the only works on heathenism which found favour in Nazi circles. Wilhelm Grnbech's Danish work Vor Folketeti Oldtiden, Copenhagen German into between in late in the i9o9 and i9i2., was translated published by Otto Hfler in the translated edition, a preface 193os,and received, which appealsto German Volk National its Bauern informed Socialism the the rhetoric of to and which degree: such a great

22Martin Ninck, Wodan und germanischer (Jena:Eugen Diederichs, 1935)Schicksalsglaube ' Jung, p. 9.

12

Die Volkskunde findet bei Grnbech die lang entbehrte groangelegte Darstellung des altgermanischenBauertums, daswir kennen mssen, historisch Volkstums Entfaltungen die wenn wir unseres spteren begreifen wollen. 25 As we have alreadyseen,studies of heathenism were often used as platforms for forms here, of early still more sinister nineteenth-century nationalism, and, aswe can see Nor, found in twentieth-century nationalism also their scholarly reflexes such studies. (seesection i. 8, below, on Danish scholarly indeed, has this tendency entirely disappeared has been it with nationalistic movements after nationalism), although not since associated National Socialism. the stamp of

LinguisticReligion Dumezil Society: 1.7 within and the

Religious Fallacy

The French scholar GeorgesDumezil arguedthat heathen mythology reflects an deities involving a central trinity of ancient pan-Indo-European theological structure ('souverainete"), functions, for three crucial socio-political namely rulership responsible 26One the difficulties ('force'), (`fecondite'). fertility of with this military power and Scandinavian drawn from Dumezil's largely for his is is approach that evidence views Gesta Danorum Sturluson's Edda Saxo Snorri the and of mythographic materials, such as Grammaticus; as is discussedbelow (section z.i), basing models of heathen cult on such late, literary evidenceis not particularly safe.A still greater difficulty, moreover, is that Dume'zil's model simplistically imagines religious and mythological ideasas developing in languages do, branching in the that sameway a chronological tree, usually only much disturbed by loanwords There is cross-dialect transfers such as and pidginisation. slightly

24See, for instance, the examination of Thomas Mann's changing attitudes towards Wagner in George Bridges, `The Almost Irresistible Appeal of Fascism, Or: Is It Okay to Like Richard Wagner?, The Germanic Review, 64 (1989), 42-48. 25Wilhelm Grnbech, Kultur und Religion der Germanen, trans. by Ellen Hofineyer, ed. by Otto Hfler (Hamburg: hanseatische Verlagsanstalt, 1937),p. 9.

26GeorgesDumezil, Les Dieux desGermain: Essaisur la Formation de la Religion Scandinave, Universitaires de France, 1959),P. I. Mythes et Religions (Paris: Presses

13

develop in do ideas this way. simply no reason to supposethat religious and mythological The fact that proto-Germanic * Tiwaz is cognatewith Greek 7. first element evs and the Latin from Jupiter developing does be *Tiwaz of a pan-Indomay seenas suggestthat European god who also gaverise to Jupiter and Zeus, but no other heathen god can be 27 develop On is in to this way. the contrary, there considerableevidenceof shown transfers between heathen and non-heathen religions, and such transfers can occur much dramatic linguistic transfers; the most more readily and on much greater scalesthan can Germanic is, example of course, the conversionsof the various groupings to christianity, in which a more or less complete replacement of heathen cults by christian cults was The in achieved a remarkably short spaceof time, and over wide geographicalareas. Celtic Germanic dialect languages by replacement of a English in the British Isles,

by contrast, has been underway for more than a millennium and a half, and is not yet in complete some regions.

The Age from Improbable: Directions Scholarship in 1.8 of the Present the i96os to the
In general, scholarshave become increasinglyreluctant to attempt studies of heathen religious expression.Since the 196osthis subject has been relatively little studied, largely in reaction to some of the more extreme and improbable previous scholarship, and Germanic for perhapsalso as a responseto the misreadingsand misappropriations of myth (see Second World before during War the nationalistic and racist ends and section i. 6, above). There has, nevertheless,been a steady,if slight, production of work in this area. There are those who still champion fanciful interpretations. The very title of Otto Hantl's Der UrglaubeAlteuropas:Die Edda als Schlssel Steinzeit is zur enough to arouse here is Jagd, the them; and contents amply confirm one's suspicions, a real wilde chasing

27 Wrterbuch, Jan de Vries, Altnordisches (Leiden: Brill, Tyr. etymologisches i,nd edn 1977),sub

14

likenessesof deities out of rock formations and using the Poetic Edda as the key evidence for a reconstruction of a proto-Indo-European mythology and religion. 28 A quite different tack is exemplified by Richard North's HeathenDeities in Old English Literature, in which North conjures heathen deities from the vernacular literature Anglo-Saxon England of christian with remarkable ease,and, of course, the aid of the Scandinavianmythographic key which is so often thought to unlock every secret of heathen religion. The work of Karl Hauck and others on the bracteatesshould also be mentioned here. Again, Scandinavianmythography is used as a key to far earlier, and far less clear, In Wodan, bracteates depicting interpreted evidence. this casethe and, again, there are as 29 little is or no evidenceto support the thesis. Hauck's work has found enthusiastic support in Denmark, whence many of the bracteate-finds originate. The importance of certain areasof Denmark as centres for the Wodan Hauck's bracteates, with production of when taken together with connection of in has is bracteates, the reinforced a new scholarly nationalism which particularly evident Scandinavia,and especiallyin Denmark. Erik Moltke has written a monograph on the Denmark development was the cradle of the origins and of the runic script, arguing that 3o futhark. elder A similar emphasiscan be seenin Lotte Hedeager'srecent monograph, Skygger of Oldnordiske literary draws myter, which on archaeologicaland en anden virkelighed: (primarily Wodan Icelandic Old in originated mythography) to claim that evidence Scandinaviaand spreadfrom there throughout Germania as the chief deity of the 31 Age's Margaret Clunies Ross dismisses Migration this warrior aristocracy. rightly Hedeager's highly her book, but implausible in argument as review of she nevertheless

28Otto Hantl, Der Urglaube Alteuropas: Die Edda Schlssel Steinzeit, Verffentlichungen als zur Wissenschaft Forschung, (Tbingen: Hochschule, Grabert, und 9 1983). aus 29See sections z. z. i and 2.3.2. See also Kathryn Starkey, `Imagining an Early Odin: Gold Bracteates as Visual Evidence?, Scandinavian Studies, 71 (1999), 373-92, who argues convincingly, between later bracteate basis iconography, textual that the comparisons of sources and on Wodan/Obinn need not be depicted on the bracteates; even by the exercise of Hauck's own insecure. his identifications are method, then,

30Erik Moltke, Runesand Their Origin: Denmark and Elsewhere, by Peter G. Foote trans. (Copenhagen: National Museum, 1985).

15

Esir Age `a Migration known acceptsthat the,, were widely pantheon' through Germania.32As we shall see,the situation Wodan least rather was at as regards more complex than this suggests. A casebearing some similarities to that of Hedeager literature and folklore, and less on archaeology based more on although

is made by Lotte Motz in The King the

Champion and the Sorcerer: A Study in GermanicMyth. This book presentsa Wodan's `the development figure, sophisticated view of chief arguing that as a cult divinity of the Germanic nations did not find his origin in a mythical but in a cultic he fashioned image in that context, the was of an ecstaticvisionary, possibly a wandering Mother Cybele, by Phrygian Mountain priest, as exemplified the metrogyrtes and of the he into in that was reshaped a mounted warrior-magician the context of a martial Unfortunately, from being based this culturei33. model suffers mainly upon the evidenceof Scandinavianmythography and contemporary folklore, and by ignoring crucial earlier Langobard evidencesuch as the various versions of the ethnogenesis. A rather different, but no more plausible,understanding of Woden appearsin an Kenneth by Harrison, Woden king human article which arguesthat was a who actually lived in the Migration Age.34The difficulties with this interpretation are discussedin below. Thor Heyerdahl Per have Lilliestrm and also espoused such neosection 3.5, Wodan Germania but Azov fact but in in lived in the not arguing that euhemerism,
35

area.

31Lotte Hedeager,Skygger (Haslev: Virkelighed: Oldnordiske Myter Samleren, Anden of en 1997)" (p. 32Clunies Ross, Alvissmal,8 (1998),I13-I5 114). 33Motz, The King, the Championand the Sorcerer, p. 70. 34`Woden', in Famulus Christi: Essays Thirteenth Centenary in Commemoration Birth the the of of (London: Gerald Bonner SPCK, by Bede, Venerable 1976),pp. 35I-56. ed. of the 35Thor Heyerdahl and Per Lilliestrm, takten pct Odin: P Sporetav vr Fortid (Oslo: Stenersen,
200I).

16

Understanding Literary ReEvidence: Re-Using 1.9 and Imagining Wodan

One responseto the problems inherent in dealing with the cult of Wodan has been more constructive than the silencewhich is the norm. Severalscholarshave Wodan for recognised the possibility and value of studying referencesto as evidence David deity. Work by christian re-uses and re-imaginings of the scholarssuch as Dumville and Craig Davis on the Anglo-Saxon royal genealogies has done much to help hence, us understand the nature and usesof these particular texts, and why they use Woden. 36This work clearly shows the political stimuli for the development of these from Woden believed be idea to the simplistic that genealogies,and moves us away an was by heathen Anglo-Saxons, from implausible the ancestor view that the cult and still more Woden heathen Anglo-Saxons form involved kingship. the of among of sacral a

In responseto an article by J. S. Ryan, which attempted Gods,but with more restraint -

like North's Heathen

to discern tracesand reminiscencesof the cult of

Woden in Old English literature, Audrey Meaney arguedfor understanding literary, iconographical its toponymic and evidencewithin contemporary cultural context, rather 37 for life. This it than viewing as evidence much earlier cultural recognition of the need to (see is important seeevidencewithin plausible contexts an element of the present work Meaney discussion below). in importance the also notes the section z.i, of this of making `a clear enough distinction between Woden and Othin', a point which is taken up and

36David N. Dumville, `The Anglian Collection Royal Genealogies Regnal Lists', Angloof and Saxon England, S (1976), z3-5o; Craig R. Davis, `Cultural Assimilation in the Anglo-Saxon Royal 0991), 2-3-36.See also Kenneth Sisam, `Anglo-Saxon Genealogies', Anglo-Saxon England, 2-1 Royal Genealogies', in British Academy Papers on Anglo-Saxon England, ed. by E. G. Stanley (Oxford: Oxford University Press, 199o), pp. 145-zo4, and Thomas D. Hill, Woden as "Ninth Father": Numerical Patterning in Some Old English Royal Genealogies', in Germania: Comparative Studies in the Old Germanic Languagesand Literatures, ed. by Daniel G. Calder and T. Craig Christy (Cambridge: Brewer, 1988),pp. 161-74. 37J. S. Ryan, `Othin in England: Evidence from the Poetry for a Cult of Woden in Anglo-Saxon England', Folklore, 74 (1963), 460-80; A. L. Meaney, Woden in England: A Reconsideration of (1966), Folklore, Evidence', 77 Ios-IS. the

17

38 in has however, (see Ryan's below). the extended present work approach, section 2.2.3, been Gods in is, Heathen North's not entirely abandonedmore recently; many ways, a larger-scale application of Ryan's method, while Raymond Tripp has also argued for the implicit representation of Obinn's ability to conversewith the dead in the Old English elegies. es.

Thomas DuBois's recent book Nordic Religionsin the Viking Age is a good, if late DuBois slightly mis-titled, piece of scholarship. concedesthe problems of using book literary for his medieval evidence pre-christian religion, and therefore structures but features important around not only an account of some also of pre-christian religion, Scandinavian literary to, and representationsof, some casestudies of responses heathenisms in the sagas. He also points out the problems of specificity and localisation of cult difficulties the alongside of using sagas as evidencein his recognition of

from literature, for Valhpll instance in the caseof conflicting evidence as a centralised destinations destination, in family-specific localised, encountered afterlife as againstthe 4' The however, from work still suffers, some sources. a slight tendency to assumethe fourteenth-century reliability of thirteenth- and witnessesto pre-christian practices,and DuBois's for Finno-Ugric impacts influences this particularly on arguments shamanic on Scandinavianreligious practices.

Conclusion moo
The development of scholarship on Germanic heathenismsand their literary is, in reflexes and representations general, a processof recognition of new sourcesand new Germanic Attempts for have identify to pre-christian religion new sources methods. not

38Meaney, Woden in England', p. ios.

39Raymond P. Tripp, `Odin's Powers and the Old English Elegies', in The Old English Elegies: New Essaysin Criticism and Research,ed. by Martin Green (Cranbury, NJ: Associated University Presses, 1983),pp. 57-68.

Thomas A. DuBois, Nordic Religionsin the Viking Age (Philadelphia: University of PennsylvaniaPress,1999),PP-74-81.

18

been bracteates, folklore always entirely successful;the possibilities of modern and of the for instance, have been rather overstatedby some their more enthusiastic proponents. of Equally, some new sourceshave not immediately been complemented by effective for for Eddas, interpretation; their methodologies the evidenceof sagas, and of the instance, has been used as evidencefor far longer than it has been viewed within its in thirteenth-century, literary context. A number of approaches which seemedvaluable but is it being earlier scholarship are not now regardedas asvaluableas they once were, hard to deny that present scholarshipwould be much the poorer and less developedhad had day. With they not once their the observation,then, that our own methods,
but do in ten years' time as they approaches and assumptions may not seem as useful now, be heathenisms, invention should understood as part of an ongoing process of rather of let discussed in than as a culmination of the previous scholarship, this chapter, us turn to `Models and Methodologies'.

19

Models Methodologies z. and

Introduction z. i
The title `Uses of Wodan: The Development of His Cult and of Medieval Literary Responses to it' points towards several of the major areasof methodological difficulty in the In deities, Germanic deities. study of pre-christian general, and especially of pre-christian has discuss in in this to not attempted scholarship area any particularly methodology be by the methodologies employed systematic way; a particular scholar can often only deduced from his or her work, rather than being set out explicitly as a part of that work. Previous scholarship has often focussed, moreover, either on the cult of Wodan or on his literary figure. Although it is representations as a possible to examine either of these fundamental isolation, in bases is one of the of this work the assumption that phenomena Wodan inter-related is the together uses of as a complex of cultural phenomena examining all figure Wodan. There important to our overall understanding of the are certainly crucially discontinuities many become in as will clear the course of this thesis in the not only

It for but is also amongst the phenomena themselves. such cultural phenomena, evidence how figure figures develop from important to try to appreciate the who or nevertheless Wodan function as parts of a whole phenomenon, since this greatly improves our chances of development be identifying the to sequencesand patterns of and evidence which are correctly found in our material.

A revision of some of the basicassumptionson which most studies of pre-christian Germani for have been based is in also order, all too often they are simply religion among the far for Scholars have been instance, Germanic too to that the ready assume, not plausible. days loan-translated Germanic into languages the the the of week were theophoric namesof

20

Yet date. is is at an early there no reasonto suppose that this the case,and a number of good (see for learned borrowing fact in in reasons supposingthat the took place a christian context below). section 3.4, Scholarshave also tended to focus on evidence form familiar in is and therefore which
This has be historiography, to easy understand. and causedundue weight to attached to Germanic Tacitus's many scholars are still prepared to accept society and religion portrait of in the Germania as accurate, despite strong arguments for the unreliability of Tacitus. l Carole M. Cusack, in her recent study of conversion, notes these difficulties with Tacitus's but Tacitus's Germania for evidence, nevertheless argues that cults of provides evidence Wodan, Thonar, Tiw and Njgr&r, and for a form of `king-cult'. 2 This is hardly satisfactory; if Tacitus's evidence is problematic, then at least some attempt to read past the distortions in familiar The fact is is if is be historiography this evidence required the evidence to that used. in form and easyto understand does not necessarilymean that it is easyto interpret, and is to that simply acknowledging that this evidence problematic and then scholars need accept hardly it is does the using anyway a viable methodology; noting not remove them. problems Let us use historiography as what it is; evidence for how historians viewed, depicted and reinvented Germanic pre-christian religion. For actual evidence of cult, and for a model of Germanic heathen cult against which to evaluate historiographical evidence, we must turn to directly The in evidence which relates more and a more straightforward manner to such cult. Romano-Germanic votive inscriptions to Matrons, found principally along the Roman side Rhine, Roman Empire, in just the although also elsewhere provide such evidence, and of the later in building Germanic this chapter as a means of a working model of preare considered christian religions and religious practices.

An over-relianceon historiographical texts for evidenceof pre-christian religions has for figures from to tendency view all evidence arising such religions as gone together with a

' Rosemary Woolf, `The Ideal of Men Dying with their Lord in the Germania and in the Battle of Maldon', Anglo-Saxon England, S (1976), 63-81(pp. 64-65).
2 Carole M. Cusack, The Rise of Christianity in Northern Europe, 300-1000 (London: (p. 36). pp. 19 and 35-37 Cassell, 1998),

21

in direct This has been for beliefs providing most obvious evidence religious and practices. In dealing Old Norse deities for heathen literary recent work with and cult practices. evidence however, heathen has been the weakness of this approach years, recognised and studies of have dwindled Richard has North, however, religions recently argued considerably as a result. fact is in in danger of becoming a restrictive `orthodoxy which is preventing useful that this because has is difficult because the subject to approach, and work simply often previous work ' In fanciful. basis discussing proved relation to the problem of of pre-christian religion on the thirteenth- and fourteenth-century Old Icelandic texts, he has argued that `Icelandic evidence late fourteenth-century Flateyjarbok it be is if the even as as argued that the can admissible, from Scandinavia )4 is likely be indigenous borrowed tradition expressed more to than outside This methodological basis has allowed North to create a sophisticated model of heathen cult in early Anglo-Saxon England based largely upon thirteenth- and fourteenth-century Icelandic texts, which is ingenious but implausible, based, as it is, upon the untenable Scandinavian English heathenisms assumptions that and were very similar, and that indigenous Scandinavian traditions have necessarilysurvived severalcenturies without alteration.

The current work will therefore subscribeto the premisethat all evidenceshould be Although historical in it we may context which appears. understood within the cultural and least Eddaic in their that the older elements, preservation some of at poemspreserve suppose fit begin how by indicates they that we should examining a thirteenth-century manuscript ' identify In it be some cases may possibleto within their wider thirteenth-century context.

North, p. x. North, pp. 93-94. Against this view, Bjarne Fidjestol argues that it is useful and possible (although difficult) to Eddaic he datings Ashley Amos's in date to the of verse; praises restraint not attempting attempt Old English but `given Old history that that the written verse, suggests of possible oral origins of Norse poetry is much shorter than that of Old English poetry, applying this sagaciousrestraint to the dating of Eddic poetry would amount to backing away from the task' (Bjarne Fidjestol, The Dating of Eddic Poetry: A Historical Survey and Methodological Investigation, ed. by Odd Einar Haugen, Bibliotheca Arnamagnaeana,41 (Copenhagen: Reitzel, 1999), p. zog; see also Ashley Crandell Amos,

22

date and early material preserved in late manuscripts with some degree of confidence (particularly in the caseof skaldic verse), and in these casesit may also prove fruitful to how Townend fits its that examine material within the contexts of original composition. for literary is, in `skaldic the points out need such contextualisation, given that praise-poetry terms, a genre that has not survived its contexts at all well, as so much of its meaning appears to be contingent on the environment of original production and reception', and he also argues in that, the case of skaldic verse, we can place some reliance on dating of composition, since our assumptions as to the accurate transmission and editorial reassembling of such skaldic 6 This is true, in fact, `reasonable verse are and generally accepted'. as of mythological skaldic but both forms these verse as of praise-poetry, one should note that of skaldic verse can be fourteenth-century in thirteenthusefully examined within the which they and contexts fact Snorri the that were preserved; may misinterpret skaldic verse of the ninth and tenth does it his for do interpretation; to to centuries not render pointless examine some attempts just that, see sections 5.2,and 5.3, below. For most other forms of evidence, moreover, it is date is to not possible original composition reliably; even when this possible it still remains desirable to question the circumstances of preservation of the material as much as the While does determine this circumstances of composition. reduce what we can about pre- and findings it firmer basis, and also that are produced on a non-christian cult, places the increases our understanding of the various ways in which christian authors and audiences used figures. and understood pre- and non-christian religion and religious

Bearing in mind thesebasicpremisesand assumptions, this chapterwill examinea iconographical, toponymic and philological. number of problematic typesof evidence,namely These types of evidenceare potentially useful but difficult to interpret satisfactorily,and their literary Wodan to the cult and reflexesof specific application an examinationof will therefore

Linguistic Means of Determining the Dates of Old English Literary Texts, Medieval Academy Books, 9o (Cambridge, MA: The Medieval Academy of America, 198o)). 6 Matthew Townend, `Contextualizing the Kntsdrdpur: Skaldic Praise-Poetry at the Court Cnut', of (PP. Anglo-Saxon England, 30 (2. ooi), 145-79 147-48).

23

be considered. The second major section of the chapter is devoted to outlining a working heathen for model of cults and societies within which to evaluate the more specific evidence O literary developments Wodan, Woden inn discussed in the cults and subsequent of and chapters.

Approaches Methodologies 2.2 and

Iconography 2.2.i

Two major sourcesof iconographyrelating to Wodan havebeen identified, namely

Wodan/monster bracteates In little both is the the and sceattas. cases,there reason to Wodan fact, is, in depicted in indeed, in that the caseof suppose on the artefacts question; for bracteates iconography in the there are good reasons that their supposing was understood Roman (see This below). terms of socio-religious models section considers the section 2.3.2, bracteates and the sceattasas a casestudy, and outlines the methodological implications of by iconographic this reassessmentof the evidence provided such artefacts.

A number of scholarshaveclaimed that at leastsomeof the extant bracteates depict The most important of theseis undoubtedly Karl heathen mythological figures or scenes. Hauck, whose extremely lengthy seriesof articles and monographs`Zur Ikonologie der Goldbrakteaten' attempts to explain almost everyknown bracteatedesign in such terms.7 It is

' Hauck's writings on the subject are too extensive to list, and the present author has examined a for An details bibliography. Hauck's list the of which see almost complete of representative sample, Mrit Gaimster's Vendel Period in bibliography Bracteates Gotland: is On the of available writings on Germanic Art, Lundensia, Acta Archaeologica (Lund: in Almqvist Significance the series octavo 2'7 of & Wiksell, 1998),pp. 2.88-9o.Hauck himself provides a readily-accessiblesummary of his for his identifications, his in `Brakteaten' identifications, in these and arguments entry mythological Johannes Altertumskunde, by Hoops, der (Berlin: Reallexikon ed. vols i- , z, germanischen the nd edn

24
important Wodan, his dealing particularly that one should review with the cult of work when he identifies depictions Wodan The bracteates. since of catalogueof gold on the majority of bracteates Die Goldbrakteaten Katalogder Vlkerwanderungszeit: Ikonographischer which

Hauck has preparedin collaboration with several Axboe, is Morten including other scholars, in terming this figure the `Er-gott', but Hauck's sole-authoredwork rather more conservative identifies the figure asWodan. usually
What, then, are the basesof Hauck's identifications? The identification of Wodan, Baldr and Loki on the so-called Drei-Gtter bracteatesrests entirely on a comparison with Snorri's account of Baldr's death.' If the bracteatesand Snorri's account were more nearly bracteate if design based depicting contemporaneous, and the on solidi were not so clearly Victory be identification implausible. winged crowning the emperor, this would not entirely The existence of one such bracteate on which the central figure (Baldr) appearsto have a his in her does little detail spear or arrow embedded to strengthen the case,since the or chest is unique to this bracteate, but central to the story as told in the Edda. If the story were depicted on the bracteates, one might, therefore, expect this detail to appear on most, if not bracteates. these all, of The identification of the profile head or human figure of the C-bracteates as Woden is based on the SecondMerseburg Charm. 9 Hauck argues that the animal beneath the head on bracteates be identified lamed horse head itself the these this the can with of charm, while Wodan healing Even discounts if between the animal. the chronological gap represents one

Walter de Gruyter, 1973-); see this entry for a discussion of the identifications mentioned in this For drawings bracteates, together with photographs and of almost all of the extant section. Karl Hauck, Die Vlkerwanderungszeit: Goldbrakteaten der and others, commentaries, see Katalog, Mnstersche Mittelalter-Schriften, 2.4,ed. by H. Belting and others, 7 vols Ikonographischer (Munich: Wilhelm Fink, 1985-i989).Consider also similar claims by Lotte Hedeager, for instance, in her `Myth and Art: A Passport to Political Authority in Scandinavia during the Migration Period', in History, `h from Sachsensymposium, York, Papers Anglo-Saxon Studiesin Archaeology the 47 and ro: September1996, ed. by Tania Dickinson and David Griffiths (Oxford: Oxford University Committee for Archaeology, 1999), pp. i 5i-S6.
8 See Gaimster, pp. 30-32.

9 See Gaimster, pp. 44-48, and the SecondMerseburg Charm, in Althochdeutsches Lesebuch, by ed. Wilhelm Braune, 13th edn, rev. by Karl Helm (Tbingen: Niemeyer, 1958),p. 86.

25

bracteates fairly looks The hardly the and the charm, this animal often seems plausible. horse-like, and sometimesthe headbecomesan entire or partial figure riding the horse. There are, however, C-bracteateson which the animal doesnot look at all horse-like, as it is
10 horse is Even be definitely if it always often shown with toes or claws. established that a can for intended distinctive horse is hardly the us to animal, a man on or with a sufficiently identify him as Wodan. It would, moreover, seem to be the casethat the motif with just a head precedes the version with a rider (which is relatively rare). 11That the head could, in Second Merseburg be some cases, re-interpreted as a rider suggeststhat the story of the Charm did not underlie the motif, since such an iconographic context would make the development of the head into a complete rider a misinterpretation; a head alone could signify Wodan healing a horse, but a man riding a horse cannot also be healing that horse. Since one (the (the depict from head), develops the them to the other we should expect motif rider) be (whether but Merseburg healing) they cannot narrative, same element riding or of the doing this. This sequenceof iconographic development clearly does not support Hauck's heathen Roman in terms of models were re-imagined mythological scenes; contention that Roman loss development is the much more reminiscent of a gradual of understanding of the in interestingly, of a gradual ornamentation and elaboration of those models models, or, more Germanic decorative tastes. accordance with

The fundamental difficulty with the Hauckian position is that its identification of depictions of heathen mythological figures restsprimarily on texts, using a method which 12 This method is crucially flawed. A much safer `Kontext-Ikonographie'. Hauck has dubbed

lo See, for instance, Die Goldbrakteaten,n, 99-ioo; z6I-6z (nos 78,79 and zoo. i). 11See Morten Axboe and Anne Kromann, `DN ODINNP FAUC?: Germanic "Imperial Portraits" Image Bracteates', Message, by Tobias FischerGold Ancient Portraiture: Scandinavian in and ed. on Hansen and others, Acta Hyperborea, 4 (Copenhagen: Museum Tusculanum Press, 1992-), pp. 2-71Roman horseman Kromann fig. (p. type that showing a 14, p. 284); a reverse points out 2-81 and 305 folds Emperor's but been have the original model, of the prefers the suggestion that the may for fact This by into the the that paludamentum were elaborated would also account no means animal. horselike. all of these animals appear 12For a typical example of this method, attempting to explain Migration Age bracteate iconography by drawing on much later texts including the SecondMerseburg Charm (see section 4.3.1) and the

26

first havesome supernatural identify depictionswhich must necessarily methodology would formulate Only by iconology element or elements. scrupulouslyexamining the alone can one be iconology depicts; a clear picture of what that the context can then examinedto attempt broaden by An to the picture thus obtained. extensiveexamination the present author of Hauck's catalogueof Migration Age bracteates hasnot revealeda single bracteatesatisfying depicting a supernaturalfigure or scene. the criterion of necessarily
A very considerable proportion of the known bracteates, however, do depict figures or This is Roman found not entirely sceneswhich are strikingly similar to those medallions. on for Hauck for Roman himself medallions provided the original models surprising, admits that 13 Hauck's Roman iconography bracteates. is despite the that, this of origin, the contention Roman iconography in bracteates the represents a radical re-use and re-interpretation of the heathen figures. As below will show, mythological scenesand section 2.3.1 order to represent highly have development is occurred; the of the major strands of unlikely to such a process bracteate iconography can far more readily be understood within a model of adoption and bracteates Roman within socio-religious significance of the comprehension of the original Germania Libera.

In 187o,Dirks, writing in the Revue de la Numismatique Belge, noted that in 1863`en decrivant la trouvaille de Terwispel, [... ] nous avons demande si la tete en face des pieces de 14. The Wodan Odin' fameux he'ros-dieu `pieces' le indiquer ou cette trouvaille ne pouvait pas in question were sceattas,and, in his article in AN, Dirks attempted a classification of this `Wodan-monstre' he four into group, named the principal groups, one of which coinage following his identification of the coins in this group as possessingthe face of Wodan on one

Karl Bildinhalten Hauck, `Fnens den besonderer Anteil ProseEdda (see sections S. 2 and 5.3), see an der vlkerwanderungszeitlichen Brakteaten (Zur Ikonologie der Goldbrakteaten, XLIX)', Frhmittelalterliche Studien, 2,6 (199z), io6-48 (pp. III-27). 13Hauck, `Fnens besonderer Anteil', PP-111-1414 J Dirks, `Les Anglo-Saxons et Leurs Petits Deniers Dits Sceattas:Essai Historique et Numismatique', Revue de la Numismatique Belge,5th series, i (1870), 81-iz8; 269-320; 387-409; 511-41 (p. 397).

27

" beyond he In the mere side, and a monster on the other. this article, moreover, went well he'ros d6ifi6 la le his Wodan Odin &ait speculation of souche, ou earlier article, arguing that There la is buste, 6tait duquel la nothing sous protection placeei16. monnaie, en acceptantson be Dirks's Wodan, his identification to substantiate or suggestionthat a currency could of Wodan; by however, speciallyprotected sceattas subsequentlyreferred to as of this type were, Wodan/monster sceattas, In label. however, the origins of this time, simply as a convenient label were entirely forgotten, and scholarsbeganto treat the identification of this bearded faceas that of Wodan asproven to be accurate.No one thought to check whether anyone had ever managedto prove that this wasindeed Wodan. The power of tradition alone had into turned a speculation a closedcase. Even more clearly than in the caseof the bracteates, iconographic interpretation this
is without basis. A bearded face could be intended to be almost anyone, and there is little ponar, Wodan is, is likely figure depicted be than to the say, reason to suppose that any more In fact, human figure is indeed, given of authority. a or a god whose name unknown to us, or, human the use of portraits of rulers on the mediterranean coinageswhich provided the for Germanic depicted likely human is it that a on the models coinages, seems ruler early sceattas.

Toponymy z. 2,. 2.

Place-namesare perhapsthe most problematic sort of evidence,or potential evidence, fact from for heathen cults. Part of the difficulty with place-names the that scholars arises highly is homogenous body tend to treat place-names asa which either significant of evidence for instance, There that theophoric placeare those who accept, or scarcelysignificant at all. deity is in demonstrate in the presence an areaof a cult of the whose name referenced names

15For this classification, see Dirks, p. z, 71. 16Dirks, p. 398"

28

the place-name.This has,historically, been a common tendencyin scholarshiprelating to Scandinavia, by because early medieval perhaps evidenceother than that provided archaeology " for Such has in is Scandinavia. in toponymy and scarce this period continued an approach " in do Then use some quartersuntil more recently. not accept again,there are those who A dialogue between these polar opposites such an understanding of theophoric place-names. is difficult to find, as is a recognition of the complexity and individuality of place-names. Although paucity of other evidencein Scandinavia has producedgreater interest in in the Germanic area,in England placetheophoric place-names than is common elsewhere have been indicate name scholars content to acceptthat theophoric place-names probably

17See, for instance, Magnus Olsen, Farms Norway: Fanes Place-Names The and ofAncient of a Country Discussed in their Bearings on Social and Religious History, trans. by Th. Gleditsch, Instituttet for Sammenlignende Kulturforskning, Serie A: Forelesninger, 9 (Oslo: Aschehoug; Leipzig: Harrassowitz; Paris: Champion; London: Williams & Norgate; Cambridge, MA: Harvard University Press, 1918), which counts over six hundred `heathen places of worship testified to by place-names' (pp. 2,67-68). Olsen provides a district-by-district discussion of these place-names in his Minner om Guderne og Deres Dyrkelse i Norske Stedsnavn (Kristiania: Steensball, 1913 [offprint from P. A. Norrone Gude- og Heltesagn]), and suggests that place-names containing the elements aker, for CE by indicate templesites open-air ritual practices which were replaced around ioo yang and vin based worship, reflected in place-names containing the element hof (pp. 35-37). Besides such grand Munch, surveys there also exists a strand of scholarship which uses the toponymy of relatively small areas as a key to reconstructing the religious organisation of the district in pre-christian times; a typical example is the discussion of the place-names of Vrend in Elias Wessen, `Schwedische Ortsnamen und Philologica Nordisk Spro, (1929Acta Scandinavica: Tidsskrift for Mythologie', 'orskning, 4 altnordische (97-100). 30), 97-115

18See, for instance, the extensive use of toponymic evidencein Odd Nordland, Valhall and Helgafell: Syncretistic Traits of the Old Norse Religion', in Syncretism:Basedon PapersRead at the Symposium Abo Held Cultural Meeting Religions, Syncretism Contact, September, 8th-Toth the at on 1966, of of on Hartman, Instituti (Stockholm: by Sven S. Scripta Donneriani Aboensis, Almqvist Wiksell, & 3 ed. Kr. Hald, Cult Odin Danish Place-Names', `The in Early in English 66-99; also of 1969), pp. and Norse Studies:Presentedto Hugh Smith in Honour of his Sixtieth Birthday, ed. by Arthur Browne and Peter Foote (London: Methuen, 1963),pp. 99-IO9. John Kousgrd Sorensen points out difficulties loss ignorance in the of extent of of pre-christian place-names the course of conversion, such as our but accepts implicitly that theophoric place-names are pre-christian in origin and cultic in nature ('The Change of Religion and the Names', in Old Norse and Finnish Religions and Cultic PlaceNames: Basedon PapersRead at the Symposiumon EncountersBetweenReligions in Old Nordic Times Held Finland, Place-Names Cultic August Tore by Ahlbck, Abo, the at on rgth-isst of 1987,ed. and on Scripta Instituti Donneriani Aboensis, 13(Stockholm: Almqvist & Wiksell, 1990), PP-394-403 (pp397-402)).

29

few basis the cult, on theophoric place-names reassuringlynon-committal that there are so 19 little that they prove very anyway. We A way through this difficulty may be to understandplace-names astexts.
interests that texts recognise early medieval are complex witnesses reflecting the concerns, biases The is but and not only of their authors, same also of their copyists and audiences. true of place-names, although we do not understand the processesof their preservation and transmission as well as we do those of literary texts. In particular, place-names rely to a much Nevertheless, if do transmission than greater extent on oral we attempt to many texts. be discern approach place-names as texts, we may able to more regional and chronological have in be a more patterning than we might expected possible, and we may able to use them for heathen treating them all as equally useful or uselessas evidence useful way than simply cult. At the same time, we must recognise that place-names, unlike literary texts, are inextricably linked with particular landscapes,and we should therefore be following Margaret 20If the Gelling's lead in thinking more about how place-namesinteract with landscapes. landscape its is place-name a text, constitutes a crucial part of the extra-textual context Although doubt the audiencesand users of a place-name no within which that text operates. differed across time and space,the landscapeattached to that place-name tends to remain landscape different by different in if ways people at was read more constant, even that different times and places.With this in mind, let us consider how, if at all, English placefor heathen cult. names provide evidence

19North, pp. 79-80; E. O. G. Turville-Petre, Myth and Religion of the North: The Religion ofAncient Scandinavia (London: Weidenfeld and Nicolson, 1964), pp. 71-7z; Kenneth Cameron, English PlacePast: PlaceNames, 3rd edn (London: Batsford, 1977),pp. 119-2o; Margaret Gelling, Signposts to the Names and the History of England, znd edn (Chichester: Phillimore, 1988),pp. 158-61(although Gelling does note of such place-names that `their significance is far from exhausted' (p. 161)). 20For an example of her attempts to read place-nameswithin their landscapes,see Margaret Gelling, `Further Thoughts on Pagan Place-Names', in Otium et Negotium: Studiesin Onomatologyand Presentedto Olof von Feilitzen, ed. by Folke Sandgren, Acta Bibliothecae Regiae Library Science Stockholmiensis, i6 (Stockholm: Norstedt, 1973),pp. io9-2,8.

30

In the late Anglo-Saxon period people lived with the perceived traces of nonChristians on the landscape. Numerous Anglo-Saxon charters have lists of bounds which include `bone haepenanbyrgels' ('the heathen burial'). " As Reynolds has pointed out, such burials may not have been pre-christian burials, but burials of criminals and other people who 22 `hxpen' buried if is need not were outside consecrated ground; clearly, this correct, the term `pre-christian', but mean simply can refer to a variety of non-christian states and positions. Being able to use an Anglo-Saxon charter successfullyrequires a way of reading the landscape, and, crucially, it requires that this way of reading the landscapebe a shared `heathen' burial is, The then, an understanding, common to all who use and write charters. immediately recognisable landscapeelement; it is a standard reference point that many individuals can recognise. Such labelling of the site as heathen, moreover, is necessarily a Anglo-Saxon from landscape; it is perspective that such a christian christian reading of the burials are constructed as explicitly heathen.

heathen What, then, of place-names cult? which are thought to provide evidenceof
Can we see in such place-names evidence for the way that heathens read landscapesin cult for Christians landscapes for the traces of non-Christians the way that terms, or read (whether heathens or those excluded from the majority christian society, such as criminals In Anglo-Saxon be individuals constructed as non-christian)? who can otherwise and England we have evidence for theophoric place-names, and for place-names which refer to a hearg, for Hill, instance, Harrow which refers to some contains the element on the cult site; 23 have little but for temple, we reason to the enactment of ritual, perhaps a sort of space England. The heathen Anglo-Saxon buildings were the normal ritual spacesof suppose that distribution of sites with names involving the elements heargand weob(which also refers to a

21Fifty-one are listed in Andrew Reynolds, `Burials, Boundaries and Charters in Anglo-Saxon England: A Reassessment',in Burial in Early Medieval England and Wales, ed. by Sam Lucy and Andrew Reynolds, The Society for Medieval Archaeology Monograph Series, 17 (London: The Society for Medieval Archaeology, zooz), pp. 171-94(PP. '9o-93). Reynolds, pp. 175-79. Cameron, English Place-Names, p. izo.

31

24 Wilson is, has The shrine of some sort) as pointed out, complementary. weohsites are typically to be found near important routeways, mostly Roman roads but also pre-Roman hearg high A distribution the sites normally occur on routes, while ground. complementary hearg date this tends to that such as suggest to around the same time, and were and weohsites known to the same groups these sites between otherwise we might expect more geographical overlap

it indicate fulfilled functions, different that they and may also religious

how differed is One they although unclear. admits, of course, the possibility that the distribution is historical complementary of such sites a accident relating to the survival of indeed, different different to place-names, or, times; on the whole, religious circumstances at however, the claim for contemporaneity of these sites seemspreferable. In this distribution, how heathen Anglo-Saxons landscapes in then, we can glimpse read some religious terms. Wilson has suggested, on the basis of the landscapecontexts of these two sorts of site, that hearg sites were set up, and exclusively used, by specific tribal groupings, while weohsites were intended for the use of anyone passing by.25This suggestion is supported, Wilson argues, by first is Cusanweoh, `Cusa's element a personal name, such as some weohnames whose shrine' Patchway, by bearg `Paeccel's first is and shrine', and some names whose element a tribal ('temple Besings') hergae' ('temples `Gumeninga `Besingahearh' of the name, such as and of 26 This however, Wilson's Gumenings'). indicates, that the evidence perhaps reading of the landscape be While between their these names and should modified slightly. relationships Wilson's hearg does the these these position, personal nature of names support weohnames We should consider, rather, that the not clearly suggest public availability of the shrines. be for by reasons other than their accessibility to routeways could siting of weohnames If have been devotees. weohswere personal shrines, then they may well very close to passing have living been building. their the the proprietor, and may even quarters of part of same

24Wilson, Anglo-Saxon Paganism, pp. 6-11. 25Wilson, Anglo-Saxon Paganism, p. io. 26Wilson, Anglo-Saxon Paganism, p. 7. Seealso Cameron, English Place-Names, p. Ito, which notes Peper Harrow as meaning `Pippera'sheathen temple'; this suggeststhat 'hearg'-sites need not always be tribal, but could be personal sites.

32

Such a proprietor would havepractical, quotidian reasons for wanting to be closeto routes of It is communication, such as a need to transport or receivegoods. possible,then, that weohs family. heargs, being by the proprietor and their were even more exclusivethan used only Their relationship with the surrounding landscape, however,indicatesa reading of landscapes by heathens,but not necessarily have had little to a religious reading; their choice of site may do with religious concern. Given this evidencefor the importance of consideringheathen cult siteswithin the landscape, the theophoric place-names context of their surrounding of one might try to assess Anglo-Saxon England in this light. When we examinesome of the landscapes associated
indicate these that these names with names we are presented with a picture which seemsto have heathen landscape, but may originally represented not christian readings readings of the Unlike landscapes like heathen burial, being the which, of were recognised as non-christian. have heathen burial, however, landscapes the the non-christian status of such seemsto depended more on a recognition that they were ancient but man-made (and therefore preindividuals who were christian), rather than on an understanding of the sites as related to perceived as non-christian.

There tends to be an assumptionthat combinations of the name of a deity with a

(other feature for for than a temple or shrine) represent evidence word a natural or man-made This form by heathens. dedication landscape landscapes the reading of may take the of a of feature to a deity, and perhaps the use of that feature as a focus for religious rituals. Or it can be thought that man-made features were perceived as having been made by gods. Thus Wilson suggests that `a mound, if particularly large and pre-Saxon, might have been regarded Thus impressive, its the name of a construction ascribed to that godi27. and as mysterious and feature such as Wansdyke linear England, in the earthwork south-west of an extensive is to be thought of

Old English (Woden's dic' from derives `Wodnes ditch) whose name Woden landscape in heathen the to reading of as relating some way. as a

Z' Wilson, Anglo-Saxon Paganism, p. 13.

33

It seemsat least asprobable,however,that such toponyms constitute evidencefor Just by Anglo-Saxons, be heathen. landscapes readings, christian of which they perceivedto late heathen burial landscape feature in instantly as the as a categoryor was recognisableeven Anglo-Saxon England, so too namessuch as Wodnes dic' were clearly readily analysable as Woden. At Woden in deity is to the time, the relating same especially commonly used names for man-made landscape features,such asbarrowsand linear earthworks. This calls to mind ('the Roman Anglo-Saxon England `enta designation in the poetic of remains as geweorc' handiwork of giants').28Strangely,this is often viewedasevidencefor ideasof supernatural builders, and the associationof Woden with earthworksis alsoviewed in this light. 29There
is no reason to suppose, however, that the Anglo-Saxons were unaware of the human origins Britain, Roman ignorant and of such monuments; they were not of the earlier occupation of Romano-British had lived in their ancestors towns when they were still and around 30`Enta figuratively `giants'. to earlier peoples as geweorc' can perfectly well refer occupied. Such peoples would be understood to have been non-christian, and Woden could well be figurative in indicate to the non-christian character of these used a or symbolic manner landscapes.It is tempting to suppose, then, that at least some of the theophoric names Woden but heathen indicate landscapes inventions, are not christian readings of referencing heathen being in origin. as

from outside England. In This is confirmed by comparisonwith evidence Obinn in Scandinavia, in contrast to England, appears place-names with a suffix which clearly

28As, for instance, in Maxims II, in The Anglo-Saxon Minor Poems,ed. by Elliott van Kirk Dobbie, The Anglo-Saxon Poetic Records, 6 (London: Routledge, 1942), pp. 55-57(p" SS)'and in The Ruin, in The Exeter Anthology of Old English Poetry: An Edition of Exeter Dean and Chapter MS 35-or,ed. by Bernard J. Muir, z, vols (Exeter: University of Exeter Press, 1994), I, 36o-6i (p. 360). 3oAs Butler points out, `in the early fifth century the Roman armies were withdrawn, but evidence from cemeteries at Aldborough, Catterick and around York shows that Saxon foederati were settled A. S. Butler, Age Leeds Region, (L. `Dark Archaeology', in its by M. W. and ed. around these towns' Beresford and G. R. J. Jones (Leeds: British Association for the Advancement of Science, 1967), PP(p. 97))" 97-100

29Gelling, Signposts Woden Past, Meaney, England', in to the p. 149; pp. 107-o8.

34
31 (cognate Old English Scandinavia refers to a cult site, such as-vi with also evidences weoh). in England only Tiw, a much greatervariety of deities in its theophoric place-names; 32 Thunor and Woden are certainly present. One might wish to modify slightly the view that Woden- place-names in England
landscapes being importance in may reflect christian readings of view of the as non-christian, Woden in Anglo-Saxon (see for below). This of royal genealogies sections 3.5 and 4.2, role Woden presents him as a pre-christian king, but strips him of divinity; he is an important but human king, That he be and prestigious ancestor, only a not a god. should associated barrows former is with and earthworks, which are potent symbols of power and prestige, hardly surprising. This association, then, does not simply identify a landscapeas heathen in heathen deity it, but heathen landscape then to origin, and attach a recognises a as and heathen Just Woden it but is to attempts claim as part of a prestigious past. as reclaimed and for heathen landscapes be re-used christian ends, so recognisably can re-read as part of a past is despite being which prestigious non-christian. A consideration of place-names and their landscapesindicates, then, that we have to be careful about the idea of `heathen place-names'; identifying theophoric names as heathen indicators is At its the same time, one of cult not without pitfalls. straightforward function be if that texts, they can therefore should recognise place-names as read and re-read by different audiences in different ways. While there are good reasons to think that some historical Anglo-Saxons in is theophoric terms, this read place-names christian and political have deny in that the to some such names may existed pre-christian times, and possibility not have been read in different ways in those times. As with other texts with an apparently lengthy, but irrecoverable, transmission history, however, we may be well advised to consider known in in these place-names relation to their than and recoverable audiences, rather but unknowable earlier audiences. relation to posited

31Hald lists several of these (pp. 99-ioi). 32Gelling, Signposts Past, Olsen, Farms in Fanes, identifies to the 158; p. and a much greater variety of deities.

35

If we want to go any way towards reconstructing pre-christian Germanic religions, we

build In initial be heathens. need to models according to evidence that can safely attributed to looking at theophoric place-names, we are, in reality, not much better off than in looking at Old Norse mythography. Yet there are more useful forms of evidence for the general scale Germanic heathenisms, discussed inscriptions in the to and nature of some such as matrons below. section z. 3.i, There is, then, a need to recognise that place-names do not operate in a simple way; they do not simply reflect a single act of naming, but a successionof uses and re-uses of a Some be for heathen but name. place-names can probably we should used as evidence cult, have different been in recognise that many place-names that used this way could reflect very landscape. Crucially, readings of the we should not attempt our own readings of place-names landscapes, by the out of context; careful consideration of and social and political contexts, be to able to make more productive use of this evidence, which attach place-names, we may but not necessarily use as evidence for heathen cult.

Philology 2-2-3

The etymology of the name Wodan has provokeda lot of rather inconclusive 33It is Old (`mad') English Old be related to wd usually thought to and scholarly attention. Norse Or ('mad'), or to the distinct Old Norse word o'er('poetry'), and it is generallyassumed Obinn West Germanic is Scandinavian the that the a straightforward cognateof name name Wodan. This is perhapsnot the case,however. The Old English form Woden cannot be related to Old Norse 'r (`poetry), which Woden If English Old in was etymologicallyrelated to w, we would expect as w66. appears Old English have developed from Primitive but Germanic Woden, *Woben. wb must not *wpazas follows:

Religionsgeschichte, 33For a summary, see Helm, Altgermanische 1, z6o-64.

36

[No Verner's Law] wpaz -*

[loss of

--z] [note (1 (6) interchangeable that wp and are usually

wpz -+

in Old English orthography] Old English wed,in contrast, could havedeveloped in two waysfrom Primitive Germanic: [b becomesd] w&az -+ [loss of wdz --+ --z] wd

[loss of -z]
wdaz -4 wd

The Old Norse form o'er,however,demonstrates that the former is the correct development,since Primitive Germanic *wdazwould produce Old Norse *odr (and in any Primitive Germanic have /d/ *wdaz initially in case, could not existed,since existedonly and 34 Germanic). /nd/ Primitive The form *wbaz,however,would regularly in the sequence developas follows: [loss of w-]
wbaz -+ 66Z [-z becomes -r] --+ Or

Wodan, then, developsfrom Primitive Germanic *wbaz,meaning something like

11 `mad, inspired, raging'. That Obinn develops from this root, however, seemsunlikely. Early Scandinavian runic inscriptions represent *wbaz as <wod(z)>, 35making a clear distinction between the phonemes /b/ and /J/, the former being represented by <d> ([d] and [b] are Obinn latter 4>. It /6/ by is striking that never appears allophones of at this period) and the Scandinavian Lobinn inscription, does in although the name runic with <d> any appear once, Obinn 36 It is also striking that in a late medieval inscription, as <loden>. is associatedwith (see from Scandinavian for him in below). These poetry very early our evidence section S. i,,

A. Campbell, Old EnglishGrammar (Oxford: ClarendonPress,1959, (no. repr. 1997),p. 164 398.3). 35Elmer H. Antonsen, A Concise Grammar of the Older Runic Inscriptions, SeriesSprachstrukturen, Reihe A, Historische Sprachstrukturen,3 (Tbingen: Niemeyer, 1975), (nos 6, from PP.31and 44-46 Sweden,c. zooAD, and 27, from Norway, c. 4ooAD). 36Samnordisk N 69. Runtextdatabas,

37

Obinn facts, from develops two taken together, seemto suggestthat the name not actually *wbaz, `mad, but from *wpaz,`poetry'.This would therefore mean that the continental Wodan and English Woden were, initially at least,a deity distinct from the Scandinavian Obinn) Obinn. Although (Wodan/Woden deities have the two and very similar names,and be in been deities. have come to conflated christian thought, they could originally separate

Nielsen arguesthat the Ribe inscription (on which seesection 5.6, below) illustrates the sound changementioned above,and outlined in the table below, by which the allophones [d] and [6] of the Primitive Germanic phoneme*/d/ split, aspart of the changesleading to 37The former becamethe Proto-Scandinavian Proto-Scandinavian. */d/, while the phoneme latter becamean allophone of Proto-Scandinavian*/p/, asfollows: Primitive Germanic Phonemes Runes Allophones */p/ [p] 111 Allophones ProtoRunes Scandinavian Phonemes [p] */_/ [i] [d] t */d/ [] */d/ d [d]

Nielsen arguesthat the P of upin is derivedfrom Primitive Germanic *Id/ 01, which appears Primitive pAim. derived from Germanic Old English), in in A are while those uipR and as <d> " Old English). There can be little doubt that Nielsen is */p/ (which gives <p> or <6> in had Ribe taking taken place,or was place,when the correct that this sound change in this inscription inscription waswritten, since Primitive Germanic */d/ ([d]) is represented by t in tuirk- (= Old Icelandic dvergr,'dwarf). The suggestionthat Proto-Scandinavian*/p/ from Primitive Germanic */d/ is represented here in addition to Proto-Scandinavian*/p/ from Primitive Germanic */p/ is, however,lesscertain. Since the runic script of the Ribe

37Hans Frede Nielsen, The Early Runic Language of Scandinavia: Studiesin Germanic Dialect Geography(Heidelberg: Carl Winter, Zooo), pp. 145-46. 38Nielsen, The Early Runic Language, p. 146.

LEEDS UNIVERSITY LIBRARY

38

inscription makes no distinction between Proto-Scandinavian */p/ from Primitive Germanic */d/ and from Primitive Germanic */p/, Nielsen's argument that upin contains ProtoScandinavian */k/ from Primitive Germanic */dl rests solely on the assumption that upin is Uroden, form from Nordendorf (on Old English brooch the cognate with or the wodan As have forms below). derive /d/ these their which see section 3.6.1, we seen above, must from Primitive Germanic */d/. If we had a Scandinavian inscription from before the sound Nielsen form is if to the change outlined above, we should expect see correct that the udin, Wodan. We however, do have is inscription, the same as so we cannot name upin not such an be certain that upin is the same as Wodan. In fact, as outlined above, upin could just as easily have developed from Primitive Germanic *wpaz, which gives Old Icelandic o'er and Old English w6, both meaning `poetry'.

That just such a developmentfrom *wkazis somewhatlikelier than development

from *wbaz can, furthermore, be evidenced from the Ribe inscription. This inscription futhark Germanic form between the early somewhere z4-character represents a transitional of futhark and the later Scandinavian i6-character futhark; 39most of the characters in the inscription belong to the i6-character futhark, but the inscription still preservesthe older (in from Gotland) be Another inscription this case which can probably runes a, g and R.

dated to the eighth century - as can the Ribe inscription (seesection 5.6, below)40 - reads 41 Another Gotlandic `iu pin udR rak'. In this inscription, the older a and d runes appear. 42Clearly, Denmark in inscription of this period preserves the older g rune. this period and on Gotland sawa transitional phasein which rune-carverswere awareof elementsof the older futhark. The fact futhark that the carverof aswell asthe newer i6-character 24-character Ribe inscription inscription G 40 used the old d rune suggests that the carverof the could felt [d] if he distinguish from have the to the need or she sound used this rune probably also

39On the developmentof the older and younger futharks seeMoltke, pp. 24-30.
4oNielsen, The Early Runic Language,p. 146, dates the Ribe inscription to around 7Z5 CE on dendrochronological grounds. 41Samnordisk Runtextdatabas, G 40. 42Samnordisk Runtextdatabas, G 157.

39

[t], or the sound [b] from [p]. We can be certain that the carver did not feel the need to distinguish [d] from [t], becausehe or she uses t to represent [d] in tuirk- ('dwarf-'). The have [cs], d fact he indicate the sound that used to carver could very easily, then, or and the does be indication is in then, that this she not could, seen as an sound not present the words being written.

It is quite possible, then, that Wodan and O sinn are not cognate deities, or were not They in however, be the understood as such were, eighth century. clearly understood to later in literatures, for Wulfstan instance, iEf in iElfric, the works of cognate as, and elweard, Snorri Sturluson (see later in Allowing for below). those of the still, sections 4.2 and 5.!, or, fact that there may well not have been a nearly pan-Germanic deity *Wobanaz in the Germanic Wodan for dating primitive period, we must still consider the possible evidence and Obinn by for former the the throughout the so-called provided existence of cognates

West Germanic languages,and for the latter throughout the so-called North Germanic languages.

The medievalWest Germanic languages, few (notably Gothic), with a exceptions all Wodan. Old English has form forms `Woden', Old Saxon `Uuoden', the the name of exhibit Old High German `Uuotan' and Langobardic `Guodan'.Equally, cognatesof Obinn appearin Danish Icelandic (e. It North Germanic Old `Upin', Old `Obinn'). dialects g. could all the dialects differing forms be if these these the that could only possess of name perhaps argued they receivedthe namethrough processes of regular phonological changes from

West North Germanic Germanic, from dialects the the and proto-languagesof or primitive If have Wodan then this the to that were case, one would, of course, accept and respectively. Obinn had Germanic before began into dialects to split the existed either primitive which least West North Germanic formed the medievalGermanic languages, before the or at and languagesbeganto developsignificant dialectal differences.This would, then, arguefor the Obinn. Wodan and considerableantiquity of the cults of

40

This need hardly be the case,however.Speakers dialects intelligible of mutually are

dialect into the translating the the usually capable of some or all of phonemes of other in dialect. Thus Standard their own etymologically equivalent phonemes a speaker of modern English recognises their /au/ (as in `down') as the Cockney speaker's/a: /. One can match up be the numerous phonemic systems with one's own, moreover; same speakerwill not only his her Cockney West Country but to connect able or phonemes with phonemes, also with 43Still (/au/) (/u: Scots /). do interestingly, this, phonemes phonemes and more one can even to a more limited extent, with other languagesas well as other dialects. Thus English French learning French schoolchildren recognise that they can sometimes create a correct by English English For instance, translating the phonemes of an word merely word. and French share a word `attention', whose phonemes transfer between the languagesas follows:

Letter(s)
<a>

English phoneme(s)
/a/

French phoneme /a/

<tt>
<en> <ti> <on>

/t/
1e/ + /n/ 1s/ /a/ + /n/ /5/

- --->

/sj/ /5/

They can perform this translation of phonemesdespitethe fact that the English and French do different English far dialects; do those of as phonemic systems not correspondasreadily as doing English nothing more complex than sayingan as the children are concerned,they are French accent. word with a

43Matthew Townend makes precisely this point in discussing Viking settlers' treatment of pre(Matthew Danelaw Townend, in `Viking England Old English Age the place-names as a existing Bilingual Society, in Cultures in Contact: Scandinavian Settlementin England in the Ninth and Tenth Centuries, ed. by Dawn M. Hadley and Julian D. Richards, Studies in the Early Middle Ages, 2, (Turnhout: Brepols, zooo), pp. 89-io5 (p. 99)). I am very grateful indeed to Professor Katie Wales for her patient help with the phonetic transcriptions here and in the following paragraphs, and for the below. `plastic', example of the word

41

Clearly, the correspondences betweenthe phonemic systemsof two English dialects fact due dialects have If is invented to the that the are essentially a common ancestor. a word however, Glaswegians London English, hear in imitate today the who and use word will not London English; in the exact sound of the word to take an examplefrom around the
beginning of the twentieth century, the word `plastic' would be spoken, at that time, by a (/plastik/), London English, speaker of upper class as whose phonemes their putative Glaswegian interlocutor would automatically translate to produce /plastik/. In fact, then, different in but dialects forms in newly-invented words can appear related which suggest that have leading differentiation dialects, those words the to the the undergone sound changes of has fact happened. in when this not Given the relatively high levels of linguistic influence and interchange among early Germanic dialects, as well as the quite ready mutual intelligibility least of at some of these

languages in the early medieval period, we need hardly doubt that the names Wodan and Obinn have developed late, have quite and simply could undergone translation of phonemes 44 Germanic It forms in languages. follows from to produce the they assume the various also determine dialect first developed in these names this that we cannot which indeed develop late on deities. if they did

for linguistic dating localising criteria or use or cults of these

from the outset that Wodan It would seem,then, that we would be unwise to assume Germania; linguistically in it his is deity perfectly possiblethat was a of considerableantiquity At Germania the the sametime, we within early medievalperiod. name spreadthroughout Obinn direct Wodan be are straightforward cognates. and should also wary of assumingthat Certainly they were equatedby christian authors, but this need not indicate a connection Again, lingustic for it in times. that seems supposed evidence actually present pre-christian

4' On the mutual intelligibility of early medieval Germanic dialects, see Townend, p. The Saxon go; B' and the Old Saxon Vatican `Genesis', An Edition of the West Saxon `Genesis Genesis: N. by A. ed. Doane (Madison: University of Wisconsin Press, 1991),P. 49; William G. Moulton, `Mutual Intelligibility among Speakersof Early Germanic Dialects', in Germania, ed. by Calder and Christy, pp. 9-Z8.

42

the antiquity and ubiquity of Wodan and his cult in Germaniais not asstrong ashas often been supposed.

Modelling Cults Heathen Societies 2.3 and

Thomas DuBois, asmentioned above(seesection 1.9),presentsa model of heathen based This is textual cult essentiallyon a comparative-anthropological reading of sources. in The DuBois by than unsatisfactory more one way. method employed posits necessary frameworks in in the parallelisms all religious cults, regardless of cultural and chronological
45This is which these cults exist. a common enough objection to comparative-anthropological phenomena. historical One his to cultural approaches should also recognise, moreover, that Old Norse literary if to texts attempt read as they are the notebooks of anthropologists, for describing by individuals the actions performed minutely and accurately and groups ritual 46Old Norse literary texts is highly purposes, problematic. are not even eye-witness accounts detailed let heathen alone and accurate accounts written with an awareness of ritual activities, of and an attempt to eschew biases cultural and colouring.

It would seemprudent, then, not to attempt to build working models of heathen framework by literary basis those outside the social texts written of the cults cults on the of based This discussion, themselves. will, therefore, proposea model on an examination of inscriptions. texts which are themselvesthe result of cult activity, namelyvotive

4sSee,for instance,his efforts to define religion (DuBois, pp. 30-31).

46Consider, for example, his reading of an account of a divinatory ritual in Eir{ks sagarauk (DuBois, PP, 17-3-2-5)-

43

Religion Pre-Christian Patterns Germanic 2.3.1 of

A crucial basisfor any considerationof the cult of Wodan must, then, be a plausible have developed in likely the to cult was model of the overall socio-religious contexts which Germanic little In idea have been pre-christian religion, we can practised. the caseof and is times, sinceany earlier evidence about cult prior to the written recordsof classical difficult interpret iconographic to therefore and very without or archaeological, necessarily fifth From first discussed through to the the centuries above. clearersupporting material, as CE, however,we have far more evidencethan is often assumed. The Roman Empire contains a very large number of votive stonesto deities with
Germanic names. Most of these are found along the Roman-German border, as well as in Upper Italy and Eastern Gaul, but they also occur in smaller numbers elsewhere in Britain, Gaul, and even Spain. The vast majority of these stones are to deities identified as `matrones' 47On the Lower Rhine, the `matrones' `matres' `matrae'. cult of such or or sometimes as CE, inscriptions beginning have flourished to while of the third century around the seems to 48 CE. Some of these figures `matrones' are known dating from the first to the fifth centuries have clearly Germanic names, some have clearly Celtic names, and some have names whose having The dedicators is clearly of the stones are also mixed, some uncertain. origin Germanic or clearly Celtic names, whilst others have Roman names, or names of uncertain Germanic female, dedicators While most are male; some were clearly are some of the origin. have been Roman Roman in names may also service, while some with ostensibly warriors 4' Germanic individuals. Roman socio-religious organisation was male-oriented, and we have a Germanic The fact Roman Germanic that men eyes. tendency to see society through fighting abroad typically seem to have set up votive stones to female deities or ancestor-

47Siegfried Gutenbrunner, The germanischen Gtternamen der antiken Inschriften, Rheinische Beitrge (Halle: Niemeyer, Volkskunde, Philologie Hlfsbcher 24 1936), pp. zoound zur germanischen und Simek, o5-o6. pp. 2, 37; 48Simek, p. zs and p. 2-04.

49Gutenbrunner, pp. 2-00-37.

44

figures, however,would seemto suggestthat Germanic socio-religious organisationat this female-centred time was more than one might otherwisethink. One might also note that these `matrones'tend to be quite tightly localised.The (for features localities instance, in the namesgiven thesevotive texts often refer to natural or `Ambioreneses'were probably, on the basisof their name, deities local to the Rhine). 50The texts are at the sametime, moreover,inextricably linked to the artefactson which they are found, and the geographicaldistribution of thesevotive stonesoften indicatesa very specific largely locality for a given group of matrons, such as the Veteranehae/Veterahenae, who are sl Embken. in restricted to a very small area and around
Simek argues that the Matron-cult is Gallic in origin, spreading from the Gauls to Germanic individuals on the Lower Rhine. 52This does not seem likely, for a number of Rhine Lower Matron-cult Firstly, Simek himself and notes, the centres on the reasons. as Eastern Gaul; if the cult was a Gaulish cult in origin, we might expect a more even distribution of the cult across Gaul. The actual distribution suggeststhat, within the Roman Empire, the cult spread outwards from the Lower Rhine. Such a spread suggestsnot that Germanic groups and individuals were receiving a Gaulish religious practice, but that Germanic groups and individuals, at a major point of contact with the Roman Empire, passed Gaulish frontier inhabitants the the of region. on one of their own religious practices to

Secondly,although a number of Germanic Matron-names seemto relate to the locality in which most inscriptions to them are found (presumablya centre of their cult), a 53 This find-site. from distant their extant cult centre or number seemto relate to places development beginning Matron-cult Germanic on the wasnot a again suggeststhat the

soSimek, p. 14. sl Gutenbrunner, pp. 135-36. s2Simek, p. z05. 53Matrons whose names appearto be related to the locality in which the majority of inscriptions to (see (see Ambioreneses Simek, Simek, Albiahenae found include 6), the the p. p. 14), the them are Iulineihiae (see Simek, p. 177),and the Renahenae (see Simek, p. z63). Matrons referenced on inscriptions found at a considerable distance from the main centres of their cult or votive stone (see Simek, Anesiaminehae (see Simek, Almaviahenae include deposition the the p. io) and p. i6).

45

Lower Rhine, but a more widespreadGermanic cult whosecontact with, and influence on, Roman Gauls Empire Lower Rhine. the the the and occurredon Thirdly, a number of Matron-names clearlyrefer to tribes from GermaniaLibera. The Frisavae,although evidencedin an inscription from Wissen, nearXanten in Germany, deities Frisian Roman Frisiavi tribes the are clearly of one of of the periphery, the or 54likewise, the Suebae, Frisaevones; from Cologne, inscriptions twice evidenced on and once 55 from Deutz, inscription Suebi. This for the on an tribe suppliedsoldiers clearlyrelate to the Roman army, including a not inconsiderablenumber of high-ranking officers (seesection below), but largely Roman Empire in late the Z.3.z, remained outside the period. classical This indicates once more that the Germanic Matron-cult wasknown outside the areas from Germania Libera. inscriptions in survive,and particularly which votive
A similar picture emerges if one examines the Germanic gods and goddesses from Suebi Inscriptions Britain Frisians, in Latin inscriptions. show and an evidenced votive 56Again, these individual with a Celtic name fulfilling vows to Germanic goddesses. cults Empire Germani Roman Germanic in be within the seem to origin, with the presenceof individuals. allowing them to spread to non-Germanic One should also note the gynocentricity of this evidence. Three Germanic gods occur in Latin votive inscriptions, compared to over thirty goddesses.The Germanic Matrons (assuming that they are not goddesses,but perhaps a slightly less important order of 57 The Germanic hundred. figure) gods, moreover, number well over a supernatural authority Mars Mercurius inscriptions Even if include four inscriptions. in just to and with we occur Germanic epithets (Mars Thingsus, Mercurius Channin(i)us, Mercurius Cimbrianus; the Simek Celtic by Roman, `Mercurius+epithet' are either clearly or combinations noted other Germanic is increased be the to number of gods only read convincingly), or too corrupt to

When inscriptions to thesegods to eleven. one considersthat many of six, and the number of

s4See Simek, p. 94" ss See Simek, p. 30Z. s6See Simek, pp. 5-6, p. i9, p. ioo. Simek, p. Z07.

46

Matrons in inscription, in the goddesses than and appear more that one and some appear very large numbers of inscriptions, it becomesclearthat Germanic religious life in the first to fifth CE female figures centuries wasvery much centred on of religious authority. This is confirmed by classical for Tacitus Historiae his in writers, recordsthe
important by Veleda Batavian CE, in 69 extremely the role played a seeresscalled uprising of 58 That Germani. that and suggests seeresses were even considered goddessesamong some Veleda was as important as Tacitus claims is confirmed by the fact that Papinius Statius also 59 her CE. Tacitus Germania, in in his mentions as a prisoner 77 makes a similar point Veleda, Aurinia: to referring again and also to a woman named divo Vespasiano Veledam diu loco vidimus sub apud plerosque numinis habitam; sed et olim Auriniam et complures alias venerati sunt, non facerent deas60 adulatione nec tamquam

Although Tacitus here points out that Aurinia and her fellows are worshippedbut not as deities, this nevertheless in important that women were extremely the socio-religious suggests life of at least some Germanic tribes. This is confirmed, moreover,by CassiusDio, who
Ganna, lived Semnones who sometime around the end of the called refers to a seeressof the 61 Domitian Rome. Ganna Again, first century CE, and who was received by was on a visit to her its in individual important society and religious practices. clearly a very

It is striking, then, that Germanic individuals in the Roman Empire (often soldiers)
individuals, The lives Matrons of these socio-religious typically made vows to and goddesses. Germania Libera, have been individuals Germanic to seem within groups and and of many highly gynocentric. Nor should we be surprised at warriors engaging in gynocentric cult, for a have Romano-Germanic inscriptions in names which appear to number of goddessesrecorded

58Cornelii Taciti Historiarum Libri, ed. by C. D. Fisher (Oxford: Clarendon Press, i9ii; repr. 1956), book 4, chapters 6i and 65; book S, chapters zz and z4. 59P. Papini Stati Silvae, ed. by E. Courtney (Oxford: Oxford University Press, 1990), p. z3 (book i, line go). chapter 4, 60Cornelii Taciti Opera Minora, ed. by Henricus Furneaux and J. G. C. Anderson (Oxford: Clarendon Press, 1900; repr. 1958),chapter 8, section 3. 61Cassii Dion is CocceianiHistoriarum Romanarum Quae Supersunt,ed. by Ursulus Philippus Boissevain, 4 vols (Berlin: Weidmann, 1955),III, i8o (book 67, chapter 5).

47
62 Tacitus in his Annales Frisian Baduhenna, to relate war, and records the a goddess called first element of whose name is probably cognatewith Old English beadu(`battle'), strongly 63 figure is in Furthermore, suggestingthat this concerned someway with martial pursuits. although the christianisation of the Roman Empire -- and of Latin Matron-cult,

causes votive

inscriptions to non-christian deities to peter out in the fifth century CE, the Germanic did, it Bede, Germanic did existing, as acrossmuch of the area, not. writing

in CE, heathen festival Anglo-Saxon the early the eighth century records of modranect: Incipiebant autem annum ab octauo kalendarum lanuarium die, ubi nunc domini Et ipsam natalem celebramus. noctem nunc nobis sacrosanctam,tunc Modranect, id gentili uocabulo est matrum noctem, appellabant, ob causam, ut suspicamur, ceremoniarum quas in ea peruigiles agebant64 They used, however, to count the beginning of the year from the eighth day January, kalends Lord. birthday the of of when we now celebrate the of the And that same night which is now holy for us, they used then to call by the Modranect `night because, the pagan name of mothers' we - which means during the they to the course of that suspect, of rites which used perform night. It is very tempting to see a connection between `Modranect' and the matron cult. One can at least say that Bede thought it probable that female figures of veneration, identified as life heathen important in Anglothe socio-religious mothers, were very and calendar of Saxons. A number of scholars have gone further than this, however, and accepted that this Matron-cult from descended Germanic Anglo-Saxon the continental passagerefers to an Matron-cult. 65They are probably right to do so.

62For (Simek, It Vihansa Harimella has been Hariasa, pp. 131,131and 361). also and example, Nehalennia (the Frisian goddess remains of whose cult centre on suggested that the name of the Walcheren and Noord-Beveland) is related to Latin necare, `to kill' (Simek, pp. zz8-z9).

63Cornelii Taciti Annalium ab Excessu Divi Augusti Libri, ed. by C. D. Fisher (Oxford: Oxford University Press, 19o6; repr. 1956),book 4, chapter 73. See also Simek p. z6. 'Bede, De TemporumRatione,ed. W. Jones, in Bedae VenerabilisOpera, 8 vols, Corpus Christianorum, Series Latina, 113B (Turnholt: Brepols, 1977),Vl.z, 330 (chapter is). My translation. 65Audrey L. Meaney, `Bede and Anglo-Saxon Paganism', Parergon, n. s. 3 (1985),1-z9 (pp. 5-6);
North, p. 2-2-7.

byCh.

48
The model within which we have to Wodan, is, then, one in which cults were place highly localised in time and place, and usually gynocentric, although did deities often male exist as well. The tribes of Germania Libera may have had common `great gods', but, if so, there is remarkably little evidence for them prior to the seventh century CE. There is, in contrast, abundant evidence for localised cults of Matrons and goddesses,and some evidence that such cults continued into the sixth and seventh centuries. The importance of women in even the cult of a seemingly male-oriented great god such as Wodan is perhaps evidenced, in the central role of women in the Langobard ethnogenesis (particularly in the moreover, in Origo Gentis Langobardorum Historia Langobardorum; the account given and the see below). section 3.6.2,,

It is possible,moreover,to corroboratethis gynocentricmodel of early Germanic

life basis In the religious on of archaeology. recent years, the archaeology of early medieval has become focus for In Denmark, interest in settlements the particular a research. led has identification high-status to the settlement archaeology of a number of apparently 66 be These for thought to centres, the associatedwith royal power. centres provide evidence development kingship in Denmark. There the origins and system of of early medieval also 67 between be finds bracteate hoards. appearsto such centres and a geographical correlation of It has been argued, on iconographic grounds, that the majority of bracteates depict the deity Wodan (see section z. z. i, above). The considerable difficulties with such an interpretation of their iconography have been discussedabove, in section z. z. i, and the possibility of in bracteate production and use terms of external social, political and religious understanding influences on areasof Germania is considered below in section z. 3.z. Of particular interest here, however, is the correlation between bracteate hoards and royal centres, which has been Wodan importance indicate in to thought to that emergent royalty was of particular

66Charlotte Fabech, `Society Landscape: From Collective Manifestations Ceremonies to and of a New Ruling Class', in Iconologia Sacra: Mythos, Bildkunst und Dichtung in der Religions- und Alteuropas : Festschrift fr Karl Hauck zum 75 Geburtstag,ed. by Hagen Keller and Sozialgeschichte Nikolaus Staubach, Arbeiten zur Frhmittelalterforschung, 23 (Berlin: Walter de Gruyter, 1994), PP(P" 133) 131-43

49
Denmark, perhaps as a legitimating ancestor.6' The presence of Woden in so many AngloSaxon royal genealogies might be seen as corroborating such a view (on such genealogies see below). sections 3.5 and 4. z,

Charlotte Behr has attempted to identify a similar early royal centre in Kent. 69She
follows earlier scholarship regarding the iconographic significance of the bracteates, stating that they are `clearly linked to the cult of Wodeni70. She adds the evidence of the place-name `Woodnesborough' to the casefor Woden as a royal patron- or ancestor-deity, stating that This is `indicates Wodeni71. the name necessarily the veneration of the god statement surely is best; has been in simplistic at 2, place-name evidence extremely as argued above, section z. z,. difficult to date, and we can rarely have any idea whether a theophoric name such as this was instituted by heathens becausethey worshipped in the place, or by christians becauseWoden England, figure Anglo-Saxon in important as the extant royal royal ancestor was an extremely 72Nor find a placeare these the only possible reasonswhy we might genealogies clearly show. Woden for Woodnesborough The Woden. was as a site at which case name referencing falls far short of proof. worshipped clearly The new element which Behr brings to her identification of an early royal centre -

here interest is it is this element which of particular and -

is her emphasison the

73 importance of women in the socio-religiouslife of the centre. The fact that bracteates found in hoards and gravesare almost alwaysfound in femalecontexts is not often mentioned
74 Wodan. Behr, however, by those who -wish to see them as manifestations of the cult of

67Fabech, `Society and Landscape', p. 133. 68Axboe and Kromann, pp. 287-300. 69Charlotte Behr, `The Origins of Kingship in Early Medieval Kent', Early Medieval Europe, 9
(2000), 25-52.

'o Behr, p. So. 71Behr, p. 43. 72See, for instance, Dumville, `The Anglian Collection'; The Chronicle of lEthelrueard, ed. by A. Campbell (London: Thomas Nelson, 1962), p. 7, p. 9, pp. I2-13, p. i8, p. 25 and p. 33. 73Behr, PP. 47-52-74But see Gaimster, pp. 215-16.Bente Magnus also implausibly suggestsa depiction of Wodan on (although bog), found brooch in in female in found this a case a relief graves another object usually Serpent: Magnus, Myth Religion Firebed (Bente `The Sweden Uppland, Ekeby, in from of the and

50

Kentish `the bracteates find in in the the richest graves argues that circumstances of Kent, some with royal connections, suggest that particularly wealthy cemeteries of eastern Anglo-Saxon Kent highest in in women of the rank early society played an ostentatious role in the religious and ideological legitimization of early Kentish kingshipi75. This is a very interesting suggestion, but the supposed connection between the bracteatesand Wodan does little to explain the importance of women in the development of early Kentish kingship. As bracteates, in importance and any religious we shall see section 2.3.2, the socio-political of the The fundamentally Romanitas. have was related to their significance they may possessed, Behr bracteates, between by women, and socio-religious power, then, can connection noted be better understood as indicating once again the existence of gynocentric areasof early Anglo-Saxon society, entirely consonant with the model derived from the Matron-cult, from late classical and early medieval authors. and

Patterns Religious, Cultural Germania: Exoticism in Syncretism of and 2.3.2 Exchange Literary and

Cultural expression -

iconographical it is or textual religious, political, whether

Indeed, interconnected function people. such expression within a social network of can only interactions, its be through the mediated of cultural network own constructing as can seen it larger forming itself, originated. social network within which a subset of the and expression This clearly has some important implications for any study of religious expression and literary representations of religious expression.

To take an examplefrom the earliestperiod coveredin the presentwork, the finds inside Germania of Roman medallions,and of statuettesof deities and mythological figures,

in the Migration Period Mirrored through Some Golden Objects', in The Transformation of the Roman World AD 400-900, ed. by Leslie Webster and Michelle Brown (Berkeley: University of (p. zoi and plate 58)). California Press, 1997),pp. 194-2-027sBehr, p. SZ,.

51

76 for Roman Germanic What testify to the existenceof audiences paganreligious expression. is far from clear, however,is the route or routes by which such Roman paganiconography Germania. We its into identifying in made way the are also presentedwith problems high but that they are all audiencesprecisely;we may suspect status, we could perhapsexplore other possibilities individuals Germania inside in there perhaps were and groups who were

fact Roman pagans, high This although not necessarily of statuswithin their own society. how Germanic Roman the the raises question of audiences understood and used pagan for how have Roman that matter, they may artefacts,and, understood and used paganism. Crucial to any attempt to answerthese questionsis a good model of how Roman paganism its Germania; into is, their that and material manifestationsmade way communities within develop dynamics Germania, in to the the we need a senseof of socialnetworks operating and frontiers Roman its across with the empire. At the later end of our period, the relationshipsbetweenOld Icelandic literature and its intra- and extra-Scandinavian The interpretation. sourcespresentnumerousproblems of Iceland between nature of connections and the outside world are the subject of considerable Icelandic be literature it by that to argument, and often seems assumed was affected other literatures and cultures only as a result of very specialconditions. A considerationof the North Sea throughout the social, cultural and political networks obtaining around the Iceland however, wasvery much more open to perhapssuggestthat medievalperiod would, has been influence than generally assumed. external Much of the work that has been done on Romano-Germanic interaction has focussed Lee frontiers. has interaction recently produceda monograph entitled on at and across Information and Frontiers, which specificallystudiesthe flow of information acrossthe

76A Roman found limes both in the the the modern on sides of pagan statuettes catalogue of Netherlands is A. N. Zadoks-JosephusJitta, W. J. T. Peters and W. A. van Es, Roman Bronze from the Netherlands, Scripta Archaeologica Groningana, i-i,, z vols (Groningen: Wolters, Statuettes Kings North Sea, AD by Evert See Kramer, the the exhibition catalogue of also z5o-8yo, ed. 1967). Ingrid Stoumann and Andrew Greg (Newcastle: Tyne and Wear Museums, zooo), p. iss (nos 136 Germania, in below. On medallion-finds see and 137).

52

Romano-Germanic border." He builds a theoretical model for information exchangebased higher density faster in density; information the the the upon population a given area, will 78 for frontier This travel acrossthat area. model works well regions,where relatively small distancesare involved. In both the examples higher however, levels mentioned above, much In different of spatial separationare the norm. consideringsuch sparse cultural networks, a approachto network structure may prove useful. Whilst it is obvious that information in the medievalperiod could only travel asfast
fastest from it how is information travelled as the means of transport available, not so clear Roman Germania, North In Sea. latter the the case,the presence empire across or around the North distances land Sea, faster long it than would while enabled connections over of the have done, precludes any connections other than long-distance ones. In the caseof information exchange between the Roman Empire and continental Germania, however, we 79 from if in information to travelled small stages, village village, or might wonder whether Roman between deep inside long-distance the there were empire and areas connections Germania? We do know of long distance connections, such as those created by trading links 80 be discussed below, Roman between northern Germania and the those empire, and, as will Such Roman legions. in Germanic connections would military service the created through have had a considerable impact on the understanding of Roman cultural and religious models Germania. within

" A. D. Lee, Information and Frontiers: Roman Foreign Relations in Late Antiquity (Cambridge: Cambridge University Press, 1993).
7sLee, pp. 149-61.

79This seemsto be the model implicit in Bjorn Myhre's discussion of Romano-Germanic longdistance interactions in `Germanic Kingdoms Bordering on Two Empires', in Kings of the North Sea, (pp" Greg, Stoumann Kramer, by 41-42-)pp. 41-54 and ed. " The considerable importance of long-distance trading links and infrastructure to the lower Vistula Vistula Area as a "Region of Power" and its Lower Urbanczyk, `The Przemyslaw in is area considered Early Middle Power Jong Ages, by Mayke de in Continental Contacts', in Topographies the ed. of and Frans Theuws with Carine van Rhijn, The Transformation of the Roman World, 6 (Leiden: Brill,
200I),

PP.

509-32'

53

Watts has recently re-examined what may be termed the `small world phenomenon'. 81 This is the phenomenon by which everyone in the world is supposedly connected through just six other people. People tend to be clustered quite tightly in terms of who they have friends have and acquaintances, a relatively small group of contact with; each person will for be by, instance, defined their social status, their employment, or their religious which may beliefs. Many of the people within that group will know each other, but few will know people In into be theory then, the world could structured clusters of people who outside the group. have little or no contact with those outside their cluster. In practice, however, a relatively far become its small number of contacts reaching outside a cluster will cause members to better connected with the members of other clusters than one might have expected; in fact, the averagenumber of steps connecting any given member of a cluster to a member of found fall in to almost the number a world without any clustering at all. another cluster can Yet the world remains very clustered, despite the shortness of the paths linking people to others outside their cluster.

in This clearly has relevance to the problemswhich recur this thesis, of

Here influences. flow information we will concentrate on early and of communication and Germanic connections with the Roman empire, as a casestudy which is applicable also in later periods and other areasin the Middle Ages. We have every reason to believe that Germania was highly clustered in the early medieval period. It seemshighly probable that it large number of relatively small tribes, often competing, perhaps occasionally at consisted of a local scale with peacewith each other, and sometimes very poorly connected on the density. low due to population neighbouring tribes,

This situation is reflected in the religious practicesof the Germani, discussed above, The Roman locality. have empire, on the other variedwidely accordingto which seemto hand, seemsto havebeen very unclustered,particularly in the religious sphere;public cults of

81Duncan J. Watts, Small Worlds: The Dynamics of Networks betweenOrder and Randomness (Princeton: Princeton University Press, 1999), P" 4"

54

the emperor and the great godswere availableto everyone throughout the empire through public sacrifices. The existenceof evena small number of links betweena Germanic tribe and the
Roman empire could, then, create a situation in which the members of that tribal grouping be better Roman Still the might connected with empire than with neighbouring tribes. more individuals the startlingly, averagenumber of connecting any member of that tribe to a Roman citizen might be very few indeed; perhaps around five. For instance, we might expect Germanic individual importance know head household, to the their a of no particular who of know head know then the their the tribe, then might of who might whoever sub-group of If leader Roman in the tribe. the tribal ambassador, was overall control of received a but directly individual therefore, the ambassadorwould not with whom we started, meet the information from the ambassadorcould reach this individual through only three intermediaries.

The sameconditions, moreover,applyin later periodsaswell. Icelandic links with

highly-clustered England, in Scandinavia, which the create conditions mainland and with but Iceland significant society of enjoyed contacts with the outside world through a small North Sea important in The important individuals. terms of tenth century was number of Viking but in in the terms of settlement, spread of also terms of cultural networks, not only what one might term skaldic mobility; the movement of skalds throughout the northern North Sea area, both with patrons and as part of the processof seeking a patron. The twelfth (although more and thirteenth centuries see not only a continuation of the skaldic network Iceland but Scandinavia in with sources of this period), also the connection of confined to literary authority in England, as well as in mainland Scandinavia. The English bias of such in his English Snorri's indicated in is royal genealogical material adoption of connections 82 Obinn's treatment of sons.

82See Anthony Faulkes, `The Genealogiesand Regnal Lists in a Manuscript in Resen's Library', in Sjtiu ritgerrir helgakr Jakobi Benediktssyni,zo. jli r977; ed. by Einar G. Petursson and Jnas Kristjansson (Reykjavik: Stofnun Arna Magnssonar a Islandi, 1977),pp. 177-9.

55

An understanding of the dynamics of sparse,highly-clustered networks with a few long-distance is, levels, in then, effective connections extremely useful considering the interchange Europe. This has prevalence and speed of cultural within medieval clearly important implications for any examination of pre-christian religions in Germania. The Germanic have depended to a availability of external religious models within societies must large extent on the existence or absenceof socio-political or economic networks with these At be literary influence time, the characteristics. can also same and artistic exchange and from Bearing in this approached perspective. mind, therefore, that this understanding of light is the that this can medieval cultural networks more widely applicable, an outline of bracteates development throw on the cultural and religious circumstances of the and use of be attempted. will now

is, in It has been shown above,in section z.z.i, that the iconographyof the bracteates
itself, not compatible with the identification of these artefacts as usually or always related to bracteates, Wodan depicted is Wodan. Not is on the the cult of only there no evidence that bracteates fifthis were there also considerable evidence that the and sixth-century have been There Roman in almost ten times as many terms of their models. understood finds of Roman coins and medallions (Roman in the sensethat they originate in the Roman Denmark, Norway in Constantinople) Rome in and rather than empire, not specifically Sweden as of bracteates. While around goo bracteatesare known, about 9000 coins and 83A Kromann's found Axboe be in is found. have been and useful summary to medallions 84 following AUC', PF ODINN `DN table: whence the article

83Axboe and Kromann, pp. 2-76-79. 84Axboe and Kromann, p. z85.

56

Norway Jutland unen Zealand Bornholm Gotland

Main land TOTAL Sweden

c. 150 C. 900 7040

Denarii
. 6o-7-3o

C. 10

c. 500

C. 28o

c. 500

C. 1200

c. 4600

ureVSolidi
c. 2.70-39

4
2

3
2

c.8o
1

S
I

96
7

Medallion
4th
C.

Medallion
imitation

IS

Siliquae C. 4th Solidi

3
I

6
c. IO

C. 300
c. I5

9
S C. ISO c. 26o C. 310

318
C. I45 86S

C. 6th-6th

Kromann traces the iconography of certain frequent bracteate motifs back to

in Kromann medallions, and, one case,to solidi. that the coins and medallions `were suggests
less for `most probably more or the reserved upper classesi85, and also states that of them were

The Kromann into jewelieryi86. transformed way that presentsthe usesof the coins and medallions failure indeed, her difference between to the and, signal clear coins and

All however, but is, her iconographic two of medallions -oversimplifiedand misleading. links are betweenbracteates it is be important therefore to and medallions,and clear about how medallions in particular were used,both by Romans and by Germanic individuals, and how they reachedGermania.The table aboveclearlyshowsthat the influx of medallions into Germania was tiny, both absolutelyand in comparisonto the arrival of coins. Even the aurei Funen, large fourth number arriving on centuriesshow a although and solidi of the third and few comparedto the arrivalsof denarli and fifth- and sixth-century solidi. This, however, is According the to overall sizeof production of medallions. not surprising when one considers

85 Axboe and Kromann, p. z86. 86Axboe and Kromann, p. z86.

57

87 Toynbee, thesewere piecesintended for specialoccasions, few and relatively were minted. Toynbee points to Tiberius II's gift of `aurei'(interpreting theseasmedallions) to Chilperic in the sixth century, and suggests that such a gift wasnot without earlier precedent:`from know Rome's dealings during fourth what we the tribesmen the third and of with northern it had is for bestowing barbarian that centuries clear she many occasions on princes gifts of a In Toynbee's complimentary, remuneratory and evenpropitiatory character"'. view, then, Roman medallions reachedGermaniaasdiplomatic gifts. This would seemto be confirmed 89 by the rarity of medallions in Germania. Another possibility should, however,be considered,especially bearing it has as some
how Germanic iconographic the the on aware recipients were of significance of the This hands Germania is in the that the medallions reached medallions. possibility of Germanic warriors who received them in the course of imperial military service. Rausing has finds Roman decorations the examined of certain military more problematic patellae and corniculi particularly phalerae, and the

in Germania, and suggestsa model in which

Germanic warriors may often have served in Roman auxiliary units, thereby gaining Roman family (which in hereditary) traditions and consequently establishing of service was citizenship 9 legions Those legions. in the ordinary were, of course, eligible to receive serving the because Rausing does decorations. these cannot not examine medallions, presumably military Indeed, decorations in be the modern sense. medallions were not classedas military readily 91 however, Toynbee does that many medallions refer point out, given only to military men. found higher in in inscriptions, their concentrations to the army and that they are often

87Jocelyn M. C. Toynbee, Roman Medallions, Numismatic Numismatic Society, 1986), pp. 73-111.

Studies, S (New York: The American

88Toynbee, p. ii8.
89Walter Pohl points out the importance of Roman influence and interchange with Germanic elites, barbarian Romanate Szilgy-Sonly finds Roman impressive treasures at sites such as and of and the in Romania, in `The Barbarian Successor States', in Transformation of the Roman World, ed. by Webster and Brown, pp. 33-47 (p. 34 and plate 7).
126-31. 9oGad Rausing, `BarbarianMercenariesor Roman Citizens?, Fornvnnen,82 (1987), 91SeeToynbee, pp. 116-I7.

58

92 Medallions importance. around placesof considerablemilitary evidently are not military decorations,but they were particularly frequently given to membersof the military. Given Germania, decorations in phalerae that military such as occur very probably as the result of their being given to a Germanic warrior who had achieveda relatively exaltedposition in imperial military service,one should at least considerthe possibility that such personsmight 93 have equally receivedmedallions. If the finds of medallionswithin Germaniaare to be ascribedto legionaryservice
diplomatic (not links, however forgetting is than that the two rather a mixture of equally for Germanic has how implications this these possible), recipients understood the medallions. Rausing points out that `in the Roman army, not only did every soldier in the legions and in the auxilii have to be able to speak Latin, he also had to be able to read it and, if an officer, to write iti94. Given that the period of service was over twenty years, we may expect that any Germanic legionary receiving a medallion would, by the end of his period of service, be well (often its inscription partly religious) significance of the able to read and well aware of the Roman It likely, cultural seems given this understanding of the motifs on the medallions. from jewellery was simply copied contexts of medallions, that the use of the medallions as 95 full in importantly, Roman that they were used cognizance of their use as such, and, more Germanic This is iconography. in their not to say that a returning the religious elements legionary would necessarily have `converted' to Roman religion, or even that he would equate Roman deities with heathen deities, but he would very probably have recognised that there his iconography in the medallion. of was some religious significance

An examination of the evidencefor particular Germanic individuals in Roman Scandinavia, fourth during the century, when the medallionsreached military service Germanic known individuals in A of the reinforces this position. very useful prosopography

92Toynbee, pp. 113and 117respectively.

93 Rausing, p. 12-7-

94Rausing, p. 130.

59

Roman military serviceduring the fourth century is given by Waas in his Germanenim Rmischen Dienst im 4. Jh. n. Chr..96This contains sixty-one individuals in total, of whom been Roman four before The following have in known based to timeline, only service are 350. data Waas's in individuals (insofar the the tribal these on study, shows origins of asthey can be determined) againstthe date at which they are first evidenced in Roman service(in the date known) is cases where a precise : Date:
350/1 352/3 354/5 356/7 358/9 6o/I
3

Gmc.
2

Gmc. EGmc.

Frank
2
I 6

Goth

amann
2

Uncertain

I
2 I

I
I

6213 641
66/7 368/9
70/I 7213

2 I
3

/ 7q.
376/7

I
I

I
I

I I

378/9
$O/I $2%3 $4/ 1 I

I
I

386/7 388/9
390/1

I I
I

92/3 94/ 6/7 I

9sSee Toynbee, pp. ii8-zi; even if one prefers Toynbee's view that the medallions entered Germania been by Romans how have diplomatic the conditioned gifts, the way that they were used may well as used them. Chr., Habelts 96Manfred Waas, Germanen im rmischen Dissertationsdrucke: Dienst im n. .jb. Reihe alte Geschichte, 3 (Bonn: Rudolf Habelt, 1965),pp. 81-134-

60

398/9
400
Total: 11 23 15 82

This table givesan idea of the relative importance of different Germanic tribes in the Roman Clearly Franks Alamanni important half in military. the and the playedan of role the second 97 fourth East Germanic The the tribes are alsorelatively well represented. century, while
however, does The largest in table, clearly not provide a complete picture. single grouping the table is of individuals who are thought to have been of Germanic origin, but whose tribal is It is individuals be identified being West origin that unknown. striking some can as of Germanic or East Germanic extraction, but no individuals can be identified as being North Germanic. This might be thought to bear out Rausing's suggestion that it might not be distinguish Romanised North forms Germanic from Germanic to possible of names other The fact being West Germanic be identified East that some names can names. as and Germanic, however, suggeststhat this is not the case.It seemsvery possible, nevertheless, East Germanic North Germanic be that names could readily and confused at this time; North features East Germanic dialects West certainly, share a number of against and Germanic dialects,98and may, therefore, be more closely related to each other than either is Black linguistic Goths Sea in West Germanic. The to the area was, to migration of the Moreover, from fourth in the amber routes the terms, a relatively recent event, the century. Baltic to the Black Sea and eastern Mediterranean provided a direct link between the North West Germanic In by-passed fact, East Germanic the tribes. the areas,which effectively and dialects have led North Germanic Gothic between some scholars to prefer and the similarities

97On which see Max Martin, `Alemannen im rmischen Heer eine verpate Integration und ihre Folgen', in Die Franken und die Alemannen bis zur "Schlachtdei Zalpich" (496/97), ed. by Dieter Geuenich (Berlin: de Gruyter, 1998),PP. 407-7-2.Also Stroheker's articles `Zur Rolle der Heermeister frnkischer Abstammung im spten vierten Jahrhundert' and `Alamannen im rmischen Reichsdienst', in Karl Friedrich Stroheker, Germanentum nd Sptantike, ed. by Olof Gigon, Die Bibliothek der alten Welt (Zrich: Artemis, 1965),pp. 9-z9 and 30-53" E. Prokosch, A Comparative Germanic Grammar (Philadelphia: Linguistic Society of America, 1939), p. 30.

61

to view the largest dialect groupingswithin Germanic asNortheast Germanic (comprising East Germanic, North Germanic, and, accordingto some,Anglo-Saxon) and South Germanic (comprising most of the dialectsnormally referred to asWest Germanic),99 There is, then, the possibility that some,or evenall, of the nameswhich Waas could
identify East Germanic fact, in Scandinavians. More only the as are, than a third of names of individuals Waas identifies, furthermore, cannot be classified more specifically than as the being Germanic Some Scandinavian. be One deny probably of these extraction. of may cannot the importance of the Franks and Alamanni, but one would expect that the tribal origins of bordering Empire be likely known Roman be members of tribes the to on to would more historians, and, therefore, more likely to be known to us. Of course, the fact that their lands bordered on the Empire might well have resulted in a larger number of Franks and Alamanni than members of other tribes entering Roman service. The considerable proportion of the known individuals whose tribal origin is not known, however, is very likely to include at least Scandinavians. some It is also worth considering the size of the sample with which we are dealing. These

individuals known because high to they achieved sixty-one are mostly us only ranks high usually exceptionally ranks; several were magistrimilitum in the Roman military. 100

From the careerof Vitalianus, which appears have begun in 363,we can seethat it was not to impossible for these individuals to rise from very lowly positions; Vitalianus beganhis Roman 13 however, far have A in the the must greaternumber, auxilii. military career ranks of it is become in relatively unimportant officers; usually only the generals remained the ranks or Illyrica, Nectaridus, Buthericus hear, like in who was magister militum or men of whom we

99Winfred P. Lehmann, 'The Grouping of the Germanic Languages', in Ancient Indo-European University Linguistics Held California, Conference Indo-European Dialects: Proceedings the at of of the on Los Angeles,April zy z7, x963,ed. by Henrik Birnbaum and Jaan Puhvel (Berkeley: University of California Press, 1966), pp. 13-z'. "' On this point see also the chronological tables of `generalissimos'in John Michael O'Flynn, Alberta Press, Western Empire (Alberta: University Roman Generalissimos of 1983),pp. 2-30-31of the 101 Waas, pp. 133-34.

62

102 Britain. In Shore Saxon in Count individuals these are the some sense, who was of of the bracteates, importance for interest it is in the to such men that to us examining most of have been likely it is have had to the medallions are most given, and such men who would Roman Such most causeand opportunity to assimilateto social and religious practices. men homeland; Vadomarius, influential for in instance, their were also often powerful and was an Alamannic king. 103 It is important to remember,then, that thesesixty-one individuals are almost
Roman Germanic larger in certainly the tiny visible part of a very much warriors number of fourth Germani in Romanised have left indeed, no military service the century; will many failure because destruction to generate trace, either of of archaeological evidence, or through documentation during their lives, or, often enough, becausethey adopted Roman names 104 It likely least from is that at some of this modern scholars. which conceal their origins The fact Scandinavians. large, but hidden, individuals that were probably very number of Germanic kings Roman became, individuals in also suggeststhat service were, or some Scandinavia both did in military rewards and, at the same as arrive medallions very probably in iconographic have In diplomatic to think terms the time, these circumstances, we of gifts. (and, therefore, religious) content of medallions being well understood by their Scandinavian badge distinction We of may also suppose that they were widely recognised as a owners. if female in bracteates frequency graves,one worn as much, not more, of perhaps, given the by the wives of the owners than by the owners themselves (not surprisingly, given the In dealing in discussed for with the section z. 3.i, above). gynocentric social patterns evidence bracteates, one must work from the undoubted fact that they are modelled mainly on Roman Romans Germanic for The the suggests service under military evidence other medallions. in Scandinavia in terms of a socio-military were probably understood that medallions

'0' Noy's list of Germanic immigrants to Rome who are evidenced in inscriptions amply demonstrates Roman far (in fact, Rome frequently in Germani than took not) names; more often with contact that

102 Waas, pp_ 94-95 and p. 117. 103 Waas, pp. 1? -8-30-

63

Roman If their significancewhich recognised origin and significance. they were understood hold it is to probablethat that significancewas understood as relating a religious significance, them to Roman paganreligion, not autochthonousheathenism.This suggests that we may have Roman in Germania, to the well underestimated extent which religion waspractised and the extent to which Germanic heathenismsabsorbedelementsof Roman religion in the late 105 classicaland early medievalperiods.

The archaeology of the bracteatesprovides further evidence that Roman paganism, heathen in Germania than as the religious context of the rather cult, was understood bracteates. Gaimster provides a useful summary of the find contexts of medallion imitations 106 her Seventeen latter known bracteates. in to 1998,although the present were of the and bracteates, Goldbrakteaten der imitations in the catalogue of author counts nineteen medallion Vlkerwanderungszeit.Of the seventeenknown to Gaimster, ten were from graves,while the 107 five female finds. Of the ten graves,one was clearly contained weapons and rest were stray 108 CE). On (and two of the latter could be dated on archaeological grounds to around 400 `early heathen Gaimster association with the warriorthese grounds mythology's states that in Scandinavian is in the testified medallion-imitations, which are well represented a cult male context"'. This position is, however, plainly not substantiated by the evidence which

Gaimster presents. Over half of the medallion imitations to which she refers cannot be finds Certainly, to their clues owners' the provide no as stray seven sex. with either associated Gaimster does from four imitations not state graveswhich medallion sexes,and there remain

Germanic have handful individuals, hundred list includes clearly names of whom only a this around a (David Noy, Foreignersat Rome: Citizens and Strangers(London: Duckworth, zooo), pp. 303-06). iosWallace-Hadrill suggestsa speciesof interpretatio Germanica, in which `some Franks had found for Roman-Celtic had like Romano-Celtic their own names their own that they accepted the gods so (Oxford: Church Oxford The Frankish (J. M. Wallace-Hadrill, keeping their attributes' gods while University Press, 1983;repr. 1985),p. i8). 106 Gaimster, p. 2,15107 Gaimster, 108 Gaimster, p. p. zis. m5.

109 Gaimster, p. eis.

64

Given, least for the then, that at to provide any evidence the gender of occupant. one female be be imitation the with a medallion can associated grave,while majority cannot it that they were particularly associated associated with either gender, seemsrash to assume with male-oriented warrior-cult. The bracteates, Gaimster, She to according are very clearlyassociated with women.
burials bracteates known, four notes that over seventy containing are of which only are 11In hoards bracteates found do beads brooches, male. with are often and although they also finds occur occasionally with such as sword sheath mounts; there are, moreover, some "l These bracteates. line in necklaces containing with the ways that uses are all very much Toynbee, Roman Empire. According to medallions were used within the medallions were hanging loop, being frames frequently in jewellery, or set and provided with a very used as 112 Toynbee for pierced also notes that medallions were sometimes pierced suspension. for fixing hammered the purpose of through the centre and up round the edges, apparently them `securely in some such object as a box or vessel,or in a piece of armor -a scabbard,

Germanic like'113. Again, practice with regard to medallions, corslet or the we can see that medallion-imitations Roman bracteates practice with regard to accorded closely with and

female It is gravesamong the graves containing medallions. possible that the predominance of bracteates has something to do with the fact that they were most frequently used asjewellery It hard is be hung the to neck. to about the person, presumably on a cord or chain around know which items of jewellery would have been considered masculine, and which feminine, by early Germanic peoples, but we have little evidence for the uses of necklaces by men, and This bracteates have it on necklaces. would worn so we may suppose unlikely that men would Lamm Axboe, bracteates from fact by be the that, according to and the confirmed appear to

110 Gaimster, p. m5. ... Gaimster, p. 2,15-

112 Toynbee, pp. 118-19. 1'3Toynbee, p. izo.

65

114 Charon's This been Obol. have in to ties well with male gravesappear used specificallyas Germania for Roman influence fourth in the evidence the military serviceasa major on It for century. also providesa starting point examining the possibility that returning Germanic legionariesbrought back not only Roman medallions,but also at least some Roman paganobservances. The following table of finds of Charon's Obols in Germaniais basedon the list given in Lamm and Axboe's article.115 `M', `F' and `U' indicate gravesof females The headed `Notes' individuals column males, and of unknown genderrespectively. Lamm Axboe's find. provides context of the and comments on the archaeological

114 Jan Peder Lamm and Morten Axboe, `Neues zu Brakteaten und Anhngern in Schweden', Frhmittelalterliche Studien, 23 (1989), 453-77(pp" 466-7)"' Lamm and Axboe, pp. 466-71.

66

Find Site Salands, rchspiel finde, Gotland

Klder,

Type of Obol Notes Classedas a bracteate,but Der kleine `Brakteat'lag umittelbar am looks more like a rough Schdelin einem beraubten .mitation of a coin (cannot Mnnergrab lkerwanderungszeitlichen identify the type of coin at Waffenbeigabe' (p. mit 466)
resent)

Aus einem doppelten Mnnergrab mit

Kirchspiel
Linde,

Coin- or Medallion'imitation

der Kaiserzeit; jngeren affenbeigaben

die Medaillonimitation lag im Mund des

Gotland
gom,

(. Bestatteten' westlich 467)

`In diesem reichen Kammergrab mit Waffen, Glsern, Bronzegefen und vielen Beigaben Kopf lag Toten des anderen am Goldscheibe' eine mnzhnliche, schlichte (. 469) Kaiserzeit, der jngeren aus mit einem

edelpad,
Sweden Hgerup,

* Gold disc

Fnen , Denmark
Hol, Kirchspiel ustad, rondela

* Di nar

Denar im Mund desToten' (P"469)

`In einem Steinkistengrab [... ] lag ein Goldstck [... ] dicht beim Schdel, "ohne Zweifel in den Mund des Toten gelegt"' (" 469) flachgehmmertes Goldstck ein von o, z, o

* Piece of gold

Sotvet, Telemark

*P iece of gold

Gile, Toten, Oppland Berlin-Britt bermllern Grave IV

* Silver disc

Gold bracteate * B-bracteate

Perlen Goldbrakteaten gr. mit und zwei zusammengefunden,wahrscheinlich Schdel' (PPdem ebenfallsnahe 469-70) In einem kaiserzeitlichenWaffengrab [...] de eine runde, durchlochte Silberscheibe Zhnen den des zwischen Skeletts gefunden' (p. 470) im Mund der Toten gefundenwurde' and hanging-loop broken off (. 470)
`a uf dem Mund der Toten liegend

hanging-loop broken gefundenwurde' and (. . 470) ofl

found in The list given in the table aboveonly includes examples the the where piecewas Lamm Axboe list finds, it; to many more and mostly of pieces corpse'smouth, or very close

67

found burials. In in hanging loop these cases, without any cremation or piercing, and some Charon's Obols. intended by be one can no means certain that the pieceswere as Lamm and Axboe may be wrong to suggestthat most bracteates, coin- and do discs, hangingmedallion-imitations, and metal a which are not piercedand not possess loop, are Charon's Obols; one must still recognisethe possibility that they constitute some function funereal They had have other species of gravegoods. within may an unconnected have indicate individual in been intended they the to the rites, or may even status of simply do, Lamm Charon's Axboe however, the the grave. and presentaround ten clearcases of Obol phenomenon. On the basisof these,one must agreewith them `dadie mediterrane Sitte, den Toten im Mund Fhrgeld fr die letzte Reise mitzugeben,in Skandinavienin der
If Lamm and Axboe jngeren Kaiserzeit und in der Vlkerwanderungszeit bekannt wari116. Obols, dealing Charon's finds identify then we are with a many unclear are right to as from from England list Lamm Axboe and two several and widespread phenomenon. The value of this evidence is questionable, but the table above does include two Germany. 117 This Obermllern. Berlin-Britz from from Germany, finds suggests that those and other Charon's Obol may have been fairly widespread, geographically, in the Germanic area. It does if demographically have been however, widespread; even one accepts all of seem to not, Lamm and Axboe's examples, and assumesa fifty percent rate of loss, the number of these largely likely It is to relatively restricted that the was practice objects very small. seems been have individuals, able to afford the custom. since they would wealthy

The most curious aspectof this practice,however,is the fact that genuine Roman 11'The Germanic for fit the to evidence with well seems custom never used. coins are
have Auxiliary Army. Roman in consisted almost units would generally the warriors serving have from their therefore, used only well, and might area, a particular entirely of warriors legionaries, became Germanic dead burying burial warriors who comrades. rites when native Roman burial becoming for had have however, would acquainted with ample opportunity

116 Lamm and Axboe, p. 471117 Lamm and Axboe, p. 468.

68

These customs. warriors -

became officersand particularly those who

have might well

brought back Charon's Obol to Germania,if they returned there after their service.They are likely been in Their have to wealthy comparisonto their neighbours,on their return. also however, Roman have largely lack wealth, must consisted of coins and medallions;the of any Charon's Obols is, the then, rather surprising. signs of use of such artefactsas The gold bracteates The lies. indicate may where the solution to the problem bracteates in fifth likely have been the to production of stimulated and sixth centuriesseems by the end of the influx of solidi and medallionsinto Funen and surrounding areas.Roman further Scandinavia, but focus the east,around coins continued to enter of activity was
Gotland, Oland and mainland Sweden (seetable above). With the end of this external supply, it became necessaryto start manufacturing imitations of Roman coins and medallions in situ. Clearly, there was considerable demand for solidi and medallions. Given that the bracteates hanging-loops, it had demand seemsprobable that which sought to satisfy that almost always indicate demand based to prestige; they were the was upon a use of solidi and medallions Their have high rarity must also social status. probably worn to show that the wearer was of a increased their value, and this rarity probably increasedsteadily after the supply from outside Charon's Roman fact This that genuine coins were not used as ceased. perhaps explains the Obol in Funen and the surrounding area; they were simply too rare and valuable to be used in fact by fit Such interpretation the that the means no all of with also this way. would an Germanic Charon's Obols are bracteates; as the table above shows, medallion-imitations, and bracteates for If discs too the were the models used. are also and pieces gold and silver bracteates for be too would only rarely that might expect this to one purpose, used valuable be used in this way.

The problems with this explanationare twofold. Firstly, it plainly doeslittle to do bracteates Secondly, Funen and the surrounding area. explain the situation outside frequently appearin graves,ashas been mentioned above.The secondobjection is perhaps bracteates however, than numerous more very considerably are since objection, not a strong

118 Lamm and Axboe, p. 469.

69

been have nearly asvaluableas the genuine solidi and medallions,and may not, therefore, One cannot, after all, placetoo much reliance on a sampleof only around genuine articles. ten Charon's Obols. In such a small number, the distribution of different artefactsusedas Charon's Obol may be misleading.Nevertheless, the bracteates few highly-clustered localities a phenomenon of a separate, largely which are, after all,

do clearly seemto fit best

Roman Roman within a model of transference religion, to of customs,and perhapseven Germania Migration in Age. the northern In examining the bracteates, then, it is crucial that we recognisethat they aroseout of
Empire, between Scandinavia Roman and one which a very strong connection and the Roman Scandinavians for developing allowed the a understanding of the ample opportunity Roman artefacts which we find in Scandinavia. The development of the bracteates clearly Hauckian but bracteates from the this understanding, which shows a gradual move away Roman have Wodan models, and, are often very close to scholars attempted to relate to For it Roman now, remains safest to crucially, to a understanding of their significance. depictions have been bracteates of a understood as are unlikely to acknowledge that these Germanic heathen deity, since they were almost certainly understood as Roman medallions; Emperor. deity deity, if there was, was the the

The evidencefor Germanic warriors servingin the Roman army, and for Germanic fits Obol, Charon's Graeco-Roman well within the clearly religious practices,such as usesof larger context of a `smallworld' model of early Germanicsociety.In this case,asin later `small keep in do literary developments, the to world' mind well would we religious and far for greatermobility of characteristicsof medievalsocial networks, which clearlyallow in been has assumed studying pre-christian than generally religious and cultural expression Germanic religions and their literary reflexes.

70

Conclusion 2.4
A number of important basic assumptions and have been in methods established this In chapter. any examination of pre-christian Germanic religions it is necessaryto allow for fact the that much about these religions is irrecoverable. The extant evidence probably refers to only a tiny fraction of the deities worshipped by Germani in the late classicaland early medieval periods. At the same time, the evidence that we possessdoes seem to indicate a localised, We little general pattern of tribal gynocentric cults with tell or regional centres. can more than this about such cults, but the strength of the evidence for them clearly indicates Germanic `great that we must reassess the our understanding of gods' as relatively small parts larger of a much and more various range of recipients of cult. The strength of the evidence for Wodan and Obinn, in particular, is substantially reduced when seen within this context.

It is also clear that exotic influenceson Germanicpre-christian religion were

considerable, and the possibility of mediterranean Finnic Celtic, indeed, BaltoSlavic and, and

Germanic into is their elements making way pre-christian religious practices a very Obinn, In both Wodan literary the real one. considering cults of and and their reflexes, the importance of such exotic features must be examined; DuBois has done much to point out "' Minn, Smi Balto-Finnic figure influence deal possible on the and/or and this thesis will heretofore Wodan influences with some on unnoticed or underestimated mediterranean and Qbinn.

The problematic evidenceprovided by toponymy, iconographyand philology has been been As has been it have in than should many previousstudies. accordedgreater weight handling, and may reflect not pre-christian shown above,such evidencerequiresvery careful later This its but thesis reflexes,or evensocial or political practicesrather than religion. cult,

119 DuBois,

pp. 134-37"

71

will assume that such evidence requires appropriate contextualisation, whether within the literary its landscape the surroundings of production, or within the social, political, or era of in it At iconography be to which the the time, caseof place-names. attaches, as same must be by based the read within such contexts only when context can established a reading on the image alone; in other words, the image must be sufficiently clear and distinctive to allow us determine by its image itself, by to the than context examining rather allowing a by influence is depicted to the predetermined context our understanding of what actually image. A double process of non-contextual reading leading to contextual reading is therefore in dealing iconographic necessary with evidence. Finally, the most crucial assumption - and it is a plausible assumption - set out in O Wodan linguistic indeed, be is inn this chapter that need not cognates, and, and need not be cognate figures at all. There is no reason to assume,as many scholars do, that Wodan was Germanic Germanic deity in tribes the primitive period, and a culted amongst the majority of differentiate the early medieval whose name therefore reflects the sound-changes which Germanic dialects, and in particular the North and West Germanic dialect families. In fact, it O Wodan independently likely inn of one another, or at that either arose seems quite and least developed quite rapidly in very different directions, as the evidence for their cults, WodanThe idea indicate. in of a pan-Germanic subsequent chapters, seemsto considered Obinn is, Scandinavian Anglo-Saxon christian and as we shall see,probably a creation of literatures, and of the efforts of modern scholars in the interpretation of these literatures.

72

Heathen: The Invention Classicising Wodan in the 3. of Eighth-Century Intellectual Milieux

Introduction 3.1
It has been suggested figure Wodan be largely in that above may a constructed
is important deity, christian contexts, and not necessarily an as usually assumed. pre-christian Just as it is necessaryto question previous assumptions, however, it is also impossible to A have therefore proceed without some assumptions. number of methodological assumptions been put forward, which attempt to place the evidence for Wodan within appropriate and late As basic the part of this process, a nature and structure of model of plausible contexts. heathen been developed, has cults within which the cult of classical and early medieval Wodan may be more usefully contextualised. Armed with these tools, let us now turn to the he in Wodan, in for the was portrayed and constructed ways which evidence and consider This (comparatively) first he literature, extensive textual notice. attracts when eighth-century literary lead back in to the seventh-century sources of these eighth-century turn, us will, depictions, and, ultimately, through reading past the distortions inherent in these christian literary sources, to an understanding of what can be known of the cult of Wodan.

Diaconus Paulus hoc loco Refert antiquitas ridiculamfabulam: 3.2,

Antique Heathen and the
Paulus Diaconus has the unusual distinction of being the earliestauthor to mention Wodan in more than one of his extant works. In his Historia Langobardorum,Paulus in Langobard two the which appears the ethnogenesis narrative of of version recounts a

73

Origo Gentis Langobardorum Fredegar's in in and earlier, seventh-centuryversions the Charlemagne, in Cbronicon.Paulus also mentions Wodan in a verseepistle addressed to Wodan Danish Sigifrid. These king he the to the to attribute worship of uses which appears divergent. The former, Wodan in hardly straightforward prose,presents seemmore could of Wodan aspart of the past, asa deity of Paulus'sown people,the Langobards,and asa human being. The latter, in ornate verse,portraysWodan asbelonging to the present,asbeing Danes, by distant deity. This divergence and simply asa and worshipped the appearance of Wodan however, The in differences is, illusory. in these are, usesof contradiction apparent Wodan, fact, simply different aspects which of a classicisingre-imagination and re-use of Wodan heathen deity. idea of as an archetypal,representative exists alongsidean
In book one, chapter eight of his Historia Langobardorum, Paulus gives the following Langobardi: version of the ethnogenesis of the Refert hoc loco antiquitas ridiculam fabulam: quod accedentesWandali ad Godan victoriam de Winnilis postulaverint, illeque respondit, se illis victoriam Gambara ad daturum quos primum oriente sole conspexisset.Tunc accessisse Fream, uxorem Godan, et Winnilis victoriam postulasse,Freaque consilium dedisse, ut Winnilorum mulieres solutos crines erga faciem ad barbae Godan similitudinem conponerent maneque primo cum viris adessentseseque fenestram ille orientem versus erat solitus per videndas pariter e regione, qua Godan Quas feisse. factum Atque ita oriente sole cum aspicere, conlocarent. Tunc Frea longibarbi? `Qui isti dixisse: subiunxisse, ut sunt conspiceret, Winnilis Godan Sicque quibus nomen tribuerat victoriam condonaret. Victoria habenda. digna Haec sunt et pro nihilo risui victoriam concessisse. l de hominum, sed caelo potius ministratur. enim non potestati est adtributa At this point antiquity recounts a preposterous tale, that the Vandals, Winnili, he Wodan, demanded that the replied and over victory approaching he would give victory to whomever he saw first at sunrise. Then Gambara for Winnili, demanded Wodan, Frea, the victory and the wife of approached loosened Winnili Frea their should arrange advised that the women of the and hair in front of their faces to look like beards, and should stand together with in in the area on which themselves the morning, and should place their men, Wodan was accustomed to look through his east window. And so it was done.

G. Waitz, Monumenta by VI-IX, Italicarum, Saec. ' Scriptores Rerum Langobardicarum ed. et My (book 8). GermaniaeHistorica (Hannover: Hahn, 1878), translation. i, chapter p. Sz

74

When Wodan saw them at sunrise,he said: `Who are theselong-beards? Then Frea addedthat he should grant victory to those to whom he had given Wodan Winnili. And These granted victory to the so a name. claims are be For is worthy of scorn and should accountedworthless. victory not ascribed humans, but bestowed is from heaven. to the power of rather
Paulus appearsto have known and used both the Origo Gentis Langobardorum and Fredegar's Chronicon in writing his Historia Langobardorum. He often contradicts Fredegar, however, his in be Origo, in this to and case narrative seems the modelled on the version given rather than on the Chronicon's version. These two seventh-century versions of the narrative will be dealt with in due course (section 3.6.z, below), but at present we are concerned with how and Paulus his in his why makes use of sources presentation of the narrative.

Paulus follows fairly closelythe narrativeasgiven in the Origo, but he doesmake

frames Firstly, Paulus the story with sceptical some significant additions to the narrative. loco fabulam'; `Haec `Refert hoc digna comments: antiquitas ridiculam risui sunt et pro nihilo habenda. Victoria enim non potestati est adtributa hominum, sed de caelo potius ministratur'. 2 As Wodan Paulus implicitly euhemeristically, understands we can see,

because humans bestow his bestowal of victory as ridiculous cannot presenting supposed Paulus frames his What however, is God is interesting, that version of more can. victory; only fact belongs draws in to that the story the narrative attention not only to the a manner which from Paulus fact but that antiquity. receivesthis mythological narrative the past, also to the We know that Paulus had written sources for the narrative, but he names instead `antiquitas' itself as his source. He also leavesthe christian god out of his condemnation at the end of the `de in caelo', a manner which could as well narrative, carefully presenting victory as granted Paulus discussion Secondly, historical Virgil in in adds a scholarly work. as a christian appear Wodan development Langobard, of as a religious and of the of the etymology of the name figure, in the chapter immediately following that in which the ethnogenesis appears:

Certurn tarnen est, Langobardosab intactae ferro barbaelongitudine, cum Nam illorum iuxta fuerint, Winnili dicti ita postmodum appelatos. primis

2 ScriptoresRerum Langobardicarum, p. sz (book i, chapter 8).

75

linguam `lang' longam, `bars'barbamsignificat. Wotan sane,quem adiecta littera Godan dixerunt, ipse est qui apud RomanosMercurius dicitur et ab Germaniae deus haec tempora, gentibus ut adoratur; qui non circa universis 3 Germania, Graecia Longe in in fuisse sed anterius, nec sed perhibetur.
It is certain, however, that the Langobards, having originally been called Winnili, were afterwards named in this manner due to the length of their beards, untouched by iron. For according to their language `lang' means long Wodan `bars' beard. indeed, letter, and means they whom, with an added Godan, is individual Mercurius is the called self-same who called among the Romans and is worshipped by all the peoples of Germania as a god; he is have lived, Germania, in but long these times, reported to not around nor before, in Greece.

This not only servesto point up Paulus'sdistrust of the Langobard ethnogenesis narrative, but also reinforceshis presentation of the narrative,and of Wodan, asbelonging to a classical Isidore's (although not Roman) past; that is, to antiquity. This chapter echoes and amplifies fertur his Etymologiae: `Langobard' in `Langobardos vulgo nominatos etymology of the name ('it Langobards by is barba the people said that they are named prolixa et numquam tonsa' 4 It beard'). is identification because lengthy of possiblethat the and never-shorn of their Mercurius's placeof origin derivesfrom the samesource,although Isidore doesnot present Mercurius in an explicitly euhemeristicmanner, nor doeshe depict Mercurius asbeing ' be De Natura Deorum Cicero's Graeco-Roman. Greek might rather than specifically five Cicero's different enumerationof thought to provide a more plausiblesource,since Mercurii, eachwith different parents,clearlysuggests a euhemeristicunderstandingof Mercurius, evenif that was not Cicero's intention. On the other hand, Cicero also fails to ' Deorum Natura De It is Greece. Mercurius also quite possiblethat explicitly with connect in Paulus, this work the eighth century are at since tracesof the use of was not availableto

4 Isidori HispalensisEpiscopiEtymologiarum sive Originum Libri XX, ed. by W. M. Lindsay, Scriptorum Classicorum Bibliotheca Oxoniensis, z vols (London and New York: Clarendon Press, My book translation. 9, chapter 2, section 95. 1911),i,

My (book Rerum Langobardicarum, Scriptores translation. i, chapter9). pp. 52-53

5 SeeEtymologiarumLibri, ed. by Lindsay.

76

7 The best scanty. for detail Martin Braga's Correctione is De most obvious source this of Rusticorum,which statesthat Mars, Mercurius, Jovis, Venus and Saturnus`fuerunt homines Graecorum' (`were in the most evil and wicked men amongst the pessimi et scelerati gente ' Greeks'). It Paulus drew his that the translation of the name elements race of seemsprobable `Langobard' from knowledge, his Langobard of personal and observationon the Wodan's be from It difficult is to the samesource. pronunciation of name probably stems in for be, but the certain, massof manuscript variants,what the original reading this should Monumenta Germaniae Historica lectione `e the edition statesthat codicum vetustiorum Wotan Uuotan litteram patet, primam vocabuli sive cum sequenticonjunctam vocalisvim habere',and this seemsmost likely to be true.9 The Etymologiae knowledge important pagan about classical was an sourceof
Wodan West, in the early medieval and the use of this text, and the equation of mythology Paulus Frankish for Mercurius, circles and the with contribute to the creation of a sensethat, in which he worked, Wodan forms part of a mythological past which is not dissimilar from Wodan Mercurius, The with the mythological past of the classicalauthors. equation of (on Corpus Glossary, England, in Anglo-Saxon in the which see section which appearsalso Isidore, from but derive does late in below) not the eighth or early ninth century, 3.4, Sancti Vita Jonas Bobbio's Paulus's knowledge seventh-century of of probably reflects Columbani, in which an equation is made between Wodan and a figure called Mercurius (see below). section 3.6.3,

is not, however,simply a Paulus'suse of Wodan in the Historia Langobardorum Paulus heathen Wodan that tradition. also emphasises classical as part of a presentationof

6 See De Natura Deorum, ed. by O. Plasberg and W. Ax, M. Tulli Ciceronis Scripta quae Manserunt Omnia, 45, znd edn (Stuttgart: Teubner, ig8o), p. 140 (book 3, chapter zz). See M. Tulli Ciceronis De Natura Deorum Liber Primus, ed. by Arthur Stanley Pease(Cambridge, MA: Harvard University Press, 1955),PP- 57-58. De Correctione Rusticorum, in Martini EpiscopiBracarensisOpera Omnia, ed. by Claude W. Barlow, Papers and Monographs of the American Academy in Rome, il (New Haven: Yale University Press; London: Oxford University Press, igso), p. 189(section 8). Martin probably took this information from Caesarius of Arles (Martini EpiscopiBracarensisOpera, ed. by Barlow, p. 164). My translation. ScriptoresRerum Langobardicarum, p. 53, n. i.

77

Wodan is a falsegod, incapableof granting victory de hominum, adtributa sed caelopotius ministratur' -

`Victoria enim non potestati est importantly, he and, most statesthat

10 This Germaniae deus is important Wodan `abuniversis gentibus ut adoratur'. an extremely Wodan for is Or least, deity. it is that the earliestextant claim statement, a pan-Germanic at however, finds On it it is that so appears. closer examination, one not entirely straightforward to determine what Paulus meanswhen he claims that Wodan is culted asa deity by all the Germania. Firstly, Wodan he does peoplesof we should note that and not consider Mercurius to be separate, he Mercurius be so could referring to culting aswell as to culting Wodan. Secondly,we may havea rather broader definition of the geographical limits of Germania than did Paulus; we tend to think of Germaniaasincluding all the areas Paulus, Germani, England Scandinavia. by inhabited including thus on and predominantly Scandinavian England being have hand, thought of the other peninsularas and the may well have from Germania; he may, therefore, meant only the peoplesof continental separate Germania.
The problem is compounded when one examines Paulus's reference to Wodan in a Paulus Charlemagne. was a member of the court circle verse epistle to scholars mostly composed of

he lively learned in Charlemagne, the took part and and as such surrounding

Some of this correspondence survives, correspondence amongst members of this circle. including a number of letters written in Latin verse, some by Paulus, and some by other Wodan letter from Paulus is This one such poem referencing members of the court circle. " Diaconus, apparently addressing itself to Charlemagne. Although addressing Charlemagne, 12 Neff explains Petrus Diaconus. however, it clearly responds to a verse epistle attributed to

10ScriptoresRerum Langobardicarum, pp. 52-53 (book i, chapters 8-9).

" Edited in Poetae Latini Aevi Carolini, ed. by Ernest Duemmler, Monuments GermaniaeHistorica, Poetarum Latinorum Medii Aevi, i (Berlin: Weidmann, 1881), pp. 5I-52. 12 Latini Aevi Carolini, ed. by Duemmler, pp. 50-5!. Neff placesthe Petrus'sverseis edited in Poetae Sigfrid, he basis to which relatesto an of the reference composition of these poemsaround 783,on the Charlemagne in Sigfrid from to 78z: embassy ] Karl [... Lippspringe, Im Jahre 782,Mitte Juli, schickte Sigfrid eine Gesandtschaft zu nach bei Karl erschienene Wenn er in unseremGedicht erwhnt ist, so magedie im Jahre 782,

78

Charlemagne for from Petrus, seeingPaulus'spieceas answering'[...] this as a commission dasvon Petrus im Auftrage Karls gegebene Rtseli13. The piececan essentiallybe viewed, then, aspart of a learnedcourtly dialogue, at leastin part playful. In this verseepistle, Paulus writes of the Danish king Sigifrid that he `adveniat
/ Nec illi auxilio Thonar et Waten erunt' (`should come with manibus post terga revinctis, his hands bound behind his back, nor will Ponar and Wodan help him'). 14Although Paulus doubt believed Wodan that no genuinely was a pan-Germanic deity, he does not seem to have known about the Danish Obinn, for he here ascribesto the Danish king Sigifrid a belief in `Waten'. Neff suggeststhat this form of the name Wodan, and the form Thonar' for ponar, are not the correct Danish forms, but reflect his knowledge of the forms used by the Saxons: neighbouring Paulus bertrug jedenfalls aus Unkenntnis der nordischen Gtter Thbor und Odhin die Namen der bei den Sachsenverehrten und ihm bekannten auf die Dnen. 15

Gesandtschaft die Veranlassung gegebenhaben und man kann mit Recht folgern, da die Entstehung des Gedichtes zeitlich diesem Ereignis nahe liegt und jedenfalls ins Jahr 783fllt. (Die GedichtedesPaulus Diaconus: Kritische und Erklrende Ausgabe, Karl by Neff, ed. Quellen und Untersuchungen zur lateinischen Philologie des Mittelalters, 3.4 (Munich: C. H. Beck, i9o8), p. zoo, note to line 17) This dating would fit well with the dating of Paulus's time at Charlemagne's court, but Neff himself ] `[... Jahre Heide, blieb Sigfrid im Briefe Alkvins that out 789 noch points auch wie aus einem litteras N M. G. Epp. de Danorum (p. ist: per ersichtlich 31)IS mandat mihi si spes ulla sit conversione' .. Neff The Alcuin's lines letter does ioo, note to i8-zo). of quotes not explicitly state that passage Sigfrid was heathen at this time, but the crucial importance of royal conversions in establishing Germanic nations as christian strongly suggeststhat we must understand this passage as referring to (see Richard Fletcher's discussion New Constantines' king `The the to the as much as of people (chapter 4), in The Barbarian Conversion:From Paganism to Christianity (New York: Holt, 1997), pp. [originally United Kingdom The Conversion Europe: From Paganism in the to published as 97-1z9 of Christianity, 37i-r386AD (London: Harper Collins, 1997)]). While the occasion of an embassymight likely to evoke such a response among the versifiers of the court, we cannot establish with seem date best it these poems were written so close to the simply to certainty that of the embassy; seems in that they the were written sometime the 7805. strong probability state 13Die GedichtedesPaulus Diaconus, ed. by Neff, p. io5 (note to lines 47-49). 14PoetaeLatini Aevi Carolini, ed. by Duemmler, p. Sz. My translation.

15 PaulusDiaconus,ed. by Neff, p. 104(note to lines 35-36). des Die Gedichte

79

Neff is undoubtedly correct that Paulus'sforms are not the correct Danish forms. The form Ribe Waten' is quite unlike the Danish form `upin' which appears in inscription already the (on On below). CE hand, Neff hardly be the which seesection 5.6, can other of around 725 forms Paulus Saxon for `Waten' is was using of the names, correct that certainly not a normal Old Saxon form of the name Wodan (for a probably Old Saxonform, seethe Old Saxon Abrenuntiatio Diaboli, discussed in section 3.3,below). Paulus'scommentson, and spellings (insofar as they can be reconstructed)of the nameWodan in his Historia Langobardorum he Langobard both deity, `Godan', the thought of suggestthat god as a specifically and as a Germanic Wotan': littera deity, `Wotan more generallyworshipped sane,quernadiecta
Godan dixerunt, [... ] ab universis Germaniae gentibus ut deus adoratur'lb. For the latter he fortis first in medial stop syllable, an unvoiced or seems to use a name showing o-vocalism the dialects in in the second syllable, although one might expect some and perhaps a-vocalism due form be This `Waten', in to a scribal then, could easily schwa an unstressed syllable. `t' His probable use of rather miscopying of a spelling reflecting these phonic characteristics. This High German fortis `d', sound changes. than stop, reflects suggesting an unvoiced or Danes, form Paulus or of the name appropriate to the was not using a strongly suggests that forms, have known Paulus Saxons; indeed, these and may well not even to the neighbouring If Paulus knew Wodan by for being have had these peoples. worshipped any evidence may not Denmark, Wodan in believed O and was that the cult of was present sinn and therefore about have he Germania, in given a more nearly correct version of the name would thus universal Obinn in his poem. The fact that he does not do so suggeststhat he ascribesbelief in Wodan to Sigifrid not becausehe knew that Sigifrid did believe in Wodan (or rather, in

O inn), but because he believedthat Wodan, asa universalGermanic deity, must be culted by Sigifrid. Paulus'sclaim for a pan-Germanic cult of Wodan may, then, be basedon the Wodan if Germanic, Paulus's Wodan in were culted context; even own, southern ubiquity of

16ScriptoresRerum Langobardicarum, p. 53 (book i, chapter 9).

80

Langobards, from Langobard Alamanni this would, only among the and a perspective such as " Paulus's,seemto be a very widespreadcult.

The idea that the Historia Langobardorumand Paulus'sverseepistleprovide evidence for Wodan being culted over a wide areaof Germania,by a wide variety of tribal groupings, is, therefore, probably incorrect. Indeed, the idea that Paulus'sverseepistleprovidesclear O for Denmark being know in is highly evidence culted also problematic; although we sinn Obinn from the Ribe inscription that Denmark in in the early eighth was alreadyculted Paulus Paulus's does knew by indicate that report that this verseepistle not seemto century, Paulus Wodan, indicate it that correctly, viewed was the case;rather seemsto not necessarily
Wodan Paulus's ponar deity. Indeed, in heathen the pairing of and as a typical and ubiquitous Sigifrid Paulus heathen by is the use of two characterising as a epistle strongly suggests that Germanic least because heathen deities, they are the partly who are archetypal at archetypal Bible. Roman (in Paulus's for deities the archetypal pagan of the cultural sphere) equivalents Jovis and Mercurius, the equivalents of ponar and Wodan, are the only Roman pagan deities believing Actus Apostolorum in portrays pagansas who appear christian scripture; due to

Paul's miraculous healing of a congenitally lame man -- that Paul and Barnabas are Jovis and Mercurius:

[Paulus] dixit magnavoce surgesuperpedestuos rectus et exilivit et ambulabat Paulus levaverunt fecerat vocem suam turbae autem com vidissent quod lycaonicedicentesdii similes facti hominibus descenderunt ad nos dux Mercurium ipse Paulum lovem Barnaban erat quoniam vero et vocabant verbi
ianuas lovis qui erat ante civitatem tauros et coronas ante sacerdos quoque adferens cum populis volebat sacrificarelg

17If Wallace-Hadrill is correct in his argument that at least some of the pre-christian Franks in Wodan, impression increase the the this the of then universality cult of would worshipped Frankish orbit (The Frankish Church, pp. zo-zz). His arguments are not particularly convincing, however, and, even if they were, the more nearly-contemporary pagansof the Suebic Frankish Frankish been have important doubt in informing models eighth-century more peripheries would no heathenism. of

81

Paulus said in a loud voice, `standup straight on your feet', and he got up and had But Paulus had done, lifted then the crowds, they walked. seenwhat up Lycaonian, in `gods in forms have down their voice, saying the of men come Barnabas And Jovis Paulus, indeed, to us'. they called and they called Mercurius, because he was the leaderof the word. Also the priest of Jovis who front bringing bulls doors, town, the the was at of and crowns outside the wanted to sacrificewith the populace. This passage links Jovis and Mercurius very specificallywith sacrifice,and certainly resonates 19 Indiculus Superstitionum in item Paganiarum Jovis'; `De Mercurii the with the et sacris as vel below, Indiculus Saxon Old Abrenuntiatio Diaboli, in the the we will see and which appears the samemanuscript, do seemto make use of thesearchetypaldeities (seesection 3.3,below).
Mercurius also appearsin scripture in Proverbia, which refers to the ancient GraecoRoman custom of throwing stones from the road onto piles of stones at crossroads,in honour of Mercurius: `sicut qui mittit lapidem in acervum Mercurii ita qui tribuit insipienti honorem' ('anyone who honours a fool is just like someone who throws a stone onto Mercurius's pile'). 20It is striking, then, that Paulus should choose ponar and Wodan as look for help. Paulus's Thunor deities Sigifrid typical to whom to might reference and Wodan clearly reflects his ideas as to the deities most important to the Danish king, but Wodan Langobards, by importance ideas the of these among the were probably conditioned Mercurius deities, Jovis importance as christian archetypes of pagan rather and of and the At Paulus drawing Danish heathen is knowledge by the same time, than of cults. on a actual Wodan functioning in deities, again presenting as some way christian tradition about classical deity. Germanic classical as a

This impression is reinforced by Paulus'suse of the phrase`manibuspost terga Duemmler (line correctly notes that this echoesa very 35) at this crucial point. revinctis'

18Acts 14.9-12, in Biblia Sacra luxta Vulgatam Verszonem, Weber Robertus by and others, 2 vols ed. (Stuttgart: Wrttembergische Bibelanstalt, 1969), 11,1722.My translation. 19See the transcription of the Indiculus in Konrad Haderlein, `Celtic Roots: Vernacular Terminology Carlomann's Draft Capitulary of A. D. 743, Codex Vat. Pal. Lat. y77', The Ritual in Pagan and Canadian journal of Irish Studies, i8 (1992), 1-30 (pp. 29-3o).

82

line Given book Aeneid for Virgil's inspiration in this classical the 57.21 i,, similar phrase bonds, and the mention of ponar aswell asWodan, there seemsno reasonto suspectthat Paulus had in mind a specificmythological attribute of Woden asa bond-loosening deity. It is more interesting, and certainly more fruitful, to remark that evenhere in his verse,full of Paulus draws heathen in in texts, gods a passingreferencevery similar to echoesof classical 22This demonstrates just the passingreferences not an of classical verseto pagangods. interest in classicalliterature, but, more importantly, an engagement with the paganismof Historia Langobardorum. literature here in just classical as the It is appropriate,then, that Paulus Diaconus should be the earliestauthor to mention Wodan in more than one extant work, for he clearly had a vigorous and productive personal from basic idea of Wodan. He usesWodan in contrasting wayswhich nevertheless spring a As have Paulus's deity. Wodan important we universal seen, as an understanding of
importance Wodan's of cult was probably skewed, resting principally upon the assessmentof figure Wodan, however, is The literary Langobards. not so tribally- and this cult among the in from Frankish intellectual his but the circles way outwards makes geographically-confined, in (in less Paulus exalted) contiguous circumstances perhaps rather worked, appearing which Germanic areas, such as Kent and the Saxon periphery of Francia.

Saxons in Imagining Saxnote: Uuoden Thunaer the ende ende 3.3 Frankish Circles

It seemsprobable,then, that Paulus Diaconus waspreparedto useWodan as an (of deities deity deity, heathen whom the specific or of regardless representative archetypal,

20Proverbs 2,6.8, in Biblia Sacra, ed. by Weber, ii, 980. My translation. ). 21PoetaeLatini Aevi Carolini, ed. by Duemmler, p. Si, (note z, ' For a useful summary of the classicalinterests of the scholars of Charlemagne's court seeF. J. E. Ages (Oxford: Middle Close Beginnings Poetry: From Christian-Latin to the Raby, A History of the of the Clarendon Press, 192.7), pp. I54-2oI.

83

Paulus was probably unaware in any case)actually worshipped by the heathens he happened to be writing about. He also seemsto have seen and used Wodan as a figure of classical Germanic status, and in this he seems to have been working within the wider classicising In Carolingian light tradition of the the court-circle. of this, we should perhaps reassess our for Wodan in the supposed the eighth evidence understanding of some of other a cult of The Old Saxon Abrenuntiatio Diaboli has been thought to provide useful century. often for This Abrenuntiatio Diaboli evidence cult. catechumens formulae baptising in one of the used

is found on folio 6" of Vatican Library manuscript Codex Palatinus Latinus

23 follows: 577,and reads as Forsaichistu diobolae.

forsacho diabolae et respondeat. ec diobolgelde end allum

fsacho dioboles respondeat. end ec allum uuercum and uuordum thunaer ende hira genotas sint24 uuoden ende saxnote ende allvm them unholdum the Do you renounce the devil?

And let him/her reply: I renouncethe devil. And all idols?

Let him/her reply: And I renounce all the devil's works and words, ponar and Wodan and Saxnot, and all those scoundrels who are their companions.

formula is language On linguistic grounds it has beensupposed this that the early of 25Krogh, for instance,has argued is not alwaysaccepted; Old Saxon, although this assessment 26 Krogh himself Unfortunately, is dialect as of this text necessary. that a reassessment of the date, in German Low linguistic is insufficient material of a similar and a points out, there linguistic dialects, to allow us to construct a reliable context within wide enough variety of

24Kleinere altschsische Sprachdenkmler:Mit Anmerkungenund Glossar,ed. by Elis Wadstein, Niederdeutsche Denkmler, 6 (Norden: Diedr. Soltan, 1899),p. i. My translation. 25For an identification of the text as Old Saxon see Old-Saxon Texts, ed. by J. H. Gallee (Leiden: Brill, 1894), pp. 246-47" Altertum, 26Steffen Krogh, `Zur Sprache des "Schsischen Taufgelbnisses"', Zeitschriftflir deutsches (1995), 143-50. 124

23See Old Saxon Texts, ed. by J. H. Gallee (Leiden: E. J. Brill, 1894),p. 245.

84
27 Saxnot Given however, locate in it the the to text, text. the this use of name which seems for Saxons, intended language is Old Saxon, the that the text that therefore probable was and be Saxons. believed intelligible dialect to to easily or a The text is noticeably lacking in featuresshowing High German influence, such as
for Primitive Germanic long /o/ (the `Uuden' in is best the digraph <uo> sequence <uuo> by forms long /o/; the considered as a semi-vowel represented with <o)representing the <uu> `uuercum' and `uuordum' clearly show that the writer of this formula was using <uu> as a be be semi-vowel, since <ue> would an unprecedented vowel spelling, and <uo> would inappropriate in `uuordum', which has short /o/ rather than long /o/). The form `hira' for the Helland is `ira' `iro' being forms the usual genitive plural personal pronoun of the unusual, and 28The forms High German initial influence, manuscripts. without while <h> probably show English Old form initial line is in the with with usage,and probably reflects the original <h> Old Saxon form. It is interesting that this formula clearly does not conform to the relatively Helfand, language is in the the of which now often thought regular spelling system evidenced 2' The Saxon for Frankish in be to spelling aristocratic audiences. circles a grapholect created The lacking internal in decidedly in consistency. erratic and contrast, appears of this text, forms <the> and <them> appear in the Helland, but the various forms of the word `and'

('end', `ende',`and') found in this text are not known in the Helfand, which consistentlyuses 3The form distinctively, Helland, in the to <ic> the and, very <ec>corresponds <endi>. 31 Helland often givesthe dative plural ending of nouns as <-un>, not <-um>. The use of <Helfand Given Helland. is the the grapholect, the nature of the of usage not normal ch-> also

27Krogh, pp. 146-49Handbibliothek, 4 (Halle: Waisenhaus,1878), ` SeeHeliand, ed. by Eduard Sievers,Germanistische passim.
for 29See, for instance, The Saxon Genesis, Doane, by a grapholect was pp. 43-47; the argument ed. Fragen und Forschungen im Bereich und in Heliandsprache', Rooth, die `ber Erik in proposed f r Theodor Frings zum 7o. Geburtstag, Philologie: Festgabe Umkreis der germanischen Verlag, (Berlin: Akademie Literatur, Sprache Instituts fr deutsche des 8 Verffentlichungen und 1956), pp" 40-79. 30Heliand, ed. by Sievers,passim. 31Heliand, ed. by Sievers,passim.

85

High Old German did impression influence this tends to confirm the that not substantially These Abrenuntiatio. Old is the writer of the considerationssuggestthat this text an early Saxon text, probably of the eighth or ninth century, but do not allow us to placeit more precisely. The manuscript confirms and narrows this dating; savefor the last item, it is written in one hand, which, accordingto Galle, can be dated to the secondhalf of the eighth 32 Dierkens century. arguesquite convincingly that the manuscript wasprobablywritten 33 CE Fulda Mainz. The goo around at or manuscriptcontains a variety of ecclesiastical texts, including some relating to eighth-century church councils, and severalrelating to pastoral The Abrenuntiatio itself has been thought to accordwell with the contents of the care. following folio (7`), namely a Creed and the IndiculusSuperstitionum Paganiarum; Galle et believesthat not only these texts, but the entire contents of the manuscript,form a unitary handbook for Fulda Saxons, from He the whole, compiled as a mission to the 772to 779.34 believesthat the main manuscript hand corroboratesthis view, being, in his view, certainly of 35This view is, however,rejectedoutright by Boudriot, who Fulda provenance. arguesthat Paganiarum, for Indiculus Superstitionum is the with the conversionof et one, not associated the Saxons: Es ist ganzdeutlich, da der Indiculus nicht ein Instrument der Sachsenbekehrung ist [...], sondernein Werkzeug der bonifatianischfrnkischen Kirchenreform. Das Bild, welchesman ausdem Indiculus erhlt, ist das gleiche wie ausder KorrespondenzdesBonifatius mit den damaligen Ppsten:daseinesverwilderten gallisch-frnkischenChristentums, in dem Art, dem Aberglauben dem die Priester aller gelegentlichsogar reichlich auch Gtzendiensthuldigten.36

33Alain Dierkens, `Superstitions, christianisme et paganismea la fin de 1'epoquemerovingienne: A Sorcellerie, Herve Parapsychologie, by Magie, in 1'Indiculus de ed. superstitionumet paganiarum', propos Hasquin, Lalcite Serie `Recherches', S (Brussels: Editions de 1'Universite de Bruxelles, [no date]), pp. (p. I2). 9-i6 34Old Saxon Texts, ed. by Gallee, p. 245. 35Old Saxon Texts, ed. by Ga11ee, p. 2,46.
36 Boudriot,

32Old Saxon Texts, ed. by Gallee,p. 24g.

Religion,p. i8. Die altgermanische

86

The contents of the Indiculus certainly bear out Boudriot's interpretation better than Boudriot's Gallee. has from Dierkens that of recently advanced position, arguing that the Indiculus is closely linked with the three related sets of acts of church councils (Concilium Germanicum, Council of Estinnes and Council of Soissons) which precede it in the manuscript, and which all reveal a concern with the practice of correct christianity, or, as the Council of Soissons puts it, `ut populus christianus paganus non fianti37. Dierkens is no doubt in Boudriot Indiculus `revelent la the correct that agreeing with the articles of mostly vivacity de coutumes que l'Eglise ne pourra jamais extirper et qui, relevant dune <<religion intemporelle, 1'Eglise tolerees populaire seront par comme pratiques folkloriques', important indicate do by the that these proviso toleration church councils not seem to `' Eglise', even if an individual ecclesiastic or community might be tolerant. 38Unfortunately, Dierkens also agreeswith Boudriot in assuming that some articles of the Indiculus reflect Germanic in continuance of pre-christian religious practice christian or semi-christian Jovis Mercurius Indiculus in be intended 8 that the contexts, suggesting to and article of may 39 Wodan ponar. There is for identification, to this refer and absolutely no evidence and there Given is intended. that the names are good reasons to suppose that such an equivalence not Wodan and ponar are used immediately prior to the Indiculus (in the Abrenuntiatio Diaboli) in the manuscript, and given that non-Latin words for non-christian practices are used in Latin Indiculus 's it be that names would several of the articles, would seem unlikely used to Germanic far be This Germanic likely. deities; the names would more refer to these manuscript reveals at several points, moreover, a concern with addressing unchristian practices in the vernacular of their practitioners. The Old Saxon Abrenuntiatio Diaboli is itself evidence baptismal in to the vernacular of such a concern, whilst a canon requiring priests use It (quoted be if, below) this concern. would strange also reflects given abrenuntiatio-formulae with

37Capitularia Regum Francorum, ed. by Alfredus Boretius, Monumenta Germaniae Historica, Legum Sectio II, Capitularia Regum Francorum, I (Hannover: Hahn, MA p. 30 (article 6). Dierkens, `Superstitions, christianisme et paganisme' (see above).

38Dierkens, p. 2,5" Dierkens, p. i9.

87

Mercurius Jovis Indiculus in this concern, the names the and were not the vernacular names by Indiculus is directed. This is those the used practices at whose unchristian suggestion Haderlein's by Indiculus convincing argument that the practices to which the strengthened refers origin. particularly those given vernacular names Celtic Germanic are of rather than

The Indiculus Superstitionum Paganiarum, Concilia, have, the et and then, associated
do Saxons. The Old Saxon Abrenuntiatio Diaboli, however, was almost to the nothing with for intended Saxons, it certainly use with since, as mentioned above, supposesits intended be Saxnot, deity `companion Saxons', to recipients worshippers of a whose name means of the have been imagined by Germanic Christians as being closely associated and who would clearly Saxons. Indiculus Although is connected with the Concilia in the manuscript, the with the then, the Abrenuntiatio may have little to do with any of them.

A closer examination of the languageof the Abrenuntiatioconfirms this suggestion.

The variation between `end', `ende' and `and' in this formula could be thought to be the lack first textual result either of corruption, or of a of certainty on the part of whoever wrote down the formula about exactly how to represent the language in writing. The latter might likely former, however, is loss in there the than seem more since no of sense this text, as one Haderlein Although fairly in textual corruption. might expect a caseof argues convincingly Paganiarum Indiculus Superstitionum that the shows substantial textual corruption, the et Diaboli Abrenuntiatio in suggestseither a very careful manuscript preservation of sense the 40 The is transmission, or even that the extant manuscript the original manuscript of this text. formula be but in this to these thought this militate against view, might spelling variations first The fact, in than they might at appear. considerably more regular spelling variations are, The first `and' is frequently in the the word syllable of one <e>. vowel which appearsmost instance of <a> occurs immediately after a syllable whose vowel is <u> ('uuercum and'), while (<e> following front `ende' A `end' initial instances vowels the and or and <ae>). of are either following in to vowel <diobol->, which appearsas <diabolpreceding <u> also seems affect a

88

following Latin but in following its with source, <o>, >, as <diobol-> <-a-> as <u> ('forsaichistu' and `allum'). It would seem, therefore, that the sound represented by <u>, in in immediately dialect following back the the this text, preceding syllable, could, of causea be by The to to the vowel pronounced closer sound represented <u>. writer's attempts to in these represent variations pronounciation appear erratic to modern readers, who are used to do standardised spelling systems which often not closely reflect the pronunciation of their language. Similarly, the choice between `end' and `ende' appearsto be conditioned by whether following (`end front final in is the syllable contains a vowel ec'; which caseno <-e> required) ('ende back final in vowel or a uuoden ende saxnote ende allvm'; which casethe <-e> perhaps between front back The difference in between the transition smoothes and vowels speech). `forsacho' from influence `forsaichistu' the second syllables of the to of and also seems result following `forsacho' the vowel; seemsto show the normal second syllable vowel, with <ai> `forsaichistu' It due following i-mutation. in to an assimilation to the arising <i>, not unlike formula in to then, that the writer of this would seem, was making a careful attempt record language did have for he not a spelling system; the writer therefore writing a or she which formula, in in this their phonological context rather than spells the words as they sound This also attempting to standardise the spelling of each word on one of the possible spellings. language he fluent knew well, since or she speakers suggests that the writer was not writing a language will recognise words and their relationships to each other, and thus tend not to of a between fine distinctions etymologically-related words, as the writer of of phonology notice Haderlein's This formula `forsacho'/'forsaichistu'. formula has in claim that this supports this `his he familiarity language lack or she was attempting to write: with the of shows the writer's Haderlein's habits firm choice of the term of a native speaker'41. phonetic are not the `phonetic' is, however, unfortunate, for it is precisely the writer's phonetic, rather than is in language the apparently erratic reflected which phonemic, approach to the source habits firm hers Abrenuntiatio: his phonemic of a native speaker, are not the or spelling of the

4oHaderlein, pp. 9-zo. 41Haderlein, p. 8.

89

but the firm phonetic habits of someonedeeplyconcernedwith accuratelyrepresentingan language. unfamiliar for The formula itself suggests an additional reason the writer's carefulrecording of
(for `et directions for it indicate example respondeat') that clearly that sounds, provides stage formula be be it important, intended the to therefore, that the spelling spoken; would was how formula. This be to to the should give an accurate guide pronounce would particularly important, moreover, if the formula was to be administered by a priest (perhaps even the language his himself) interlocutor; is, the writer who was unfamiliar with of unfamiliar, that formula formula. language The in in the the title given the with represented of the baptismalesi42, `Interrogationes indicates formula that the manuscript, et responsiones also formula, flock; Latin be by in dealing to a similar also with was used priests with their stage 43 directions, exists in a probably Frankish dialect, and, indeed, later in Palatinus Latinus 577 44 there appearsa canon which states: Nullus sit presbyter qui in ipsa lingua, qua nati sunt, baptizandos, interrogare intelligant non studeat, ut abrenuntiationes vel confessionesaperte dedignatur, taliter agere quibus abrenuntiant, vel quae confitentur; et qui 45 de secedat parochia. May there be no priest who does not take care to conduct the renunciations in language into be baptised the same clearly and or confessions of those to born, so that they may understand what they are renouncing, which they were let in or what they are confessing; and whoever refuses to act this manner, him retire from his parish.

42Dierkens, p. iz.

43 (Berlin: Weidmann, Elias Steinmeyer by Sprachdenkmler, Die kleinerenalthochdeutschen von ed.

1916), p. 2-3-

" On the characterof this manuscriptand the canonsand other materialscontainedtherein, see Historical Royal Rosamond McKitterick, The Frankish Churchand the CarolingianReforms, 789-895, Society Studiesin History (London: Royal Historical Society,1977), pp. 33-34.
45Patrologia Latina, volume 89, columns 82,2D-823A (<http: //pld. chadwyck.co.uk/pld/fulltext? action=byoffset&warn=N&offset=9z33i586&div=5&file=.. /se [accessed My December ). translation. 5 zooz] ssion/IO391075o6_27323>

90

The Old Saxon (if it is indeed Old Saxon)AbrenuntiatioDiaboli clearlyrepresents a response This that the writer was,as one might expect,a cleric, and to canonssuch asthis. suggests Christianity. one concernedwith the reinforcement, and perhapsthe extension, of The Abrenuntiatiois, then, a text which reflects the concernsof the Indiculusand the Concilia and which respondsto thesetexts. This doesmuch to explain the coexistence of these texts in Vatican manuscript Palatinus Latinus 577.It is also clear, however,that the Abrenuntiatio hasnothing to do with the Indiculusand the Concilia at the stageof
latter in in the composition; were composed 743-744, responseto poor observanceof christian behaviour in Francia and its dependencies,whereas this copy of the Abrenuntiatio is probably, its highly idiosyncratic the given accuracy of reproduction of a spelling system, the original It directly the text. that the copy of seemsprobable writer of this manuscript copied the text into the manuscript from an oral source such as an interpreter or a priest who was already formula It but had down. it is using the also possible that the writer was the not written individual already using the formula, although this seemsless likely as such an individual likely It be less formula in is than to record the would a phonetic rather a phonemic manner. difficult, then, to determine exactly when and in what context the text was composed, but we be in baptising Saxons. it intended be that to was would probably used correct to suppose Given the references to ponar, Wodan and Saxnot in the formula, moreover, it seemslikely Saxons for its be intended formula to whom author perceived use with that the was originally from do; (not the quotation as that this author necessarilyperceived paganism as we pagan Soissons Council the given above shows, eighth-century ecclesiasticscould understand of `pagan'). Christians as poorly-behaved

While the attempts of the writer of Codex PalatinusLatinus 577to representthe however, is been have his her there some scrupulous, target audiencemay or vernacularof formula in include he doubt the the nonthat or she made any seriousattempt to reasonto (if by deities who actuallywere worshipped this audience this audiencewere actually christian As discussed Christians bad or semi-Christians). non-Christians at all, rather than simply Wodan, Jovis Mercurius ponar Paulus Diaconus's in were and and use of above, relation to

91

(section for deities ). Not it the archetypal pagan early christian authors 3.2, surprisingly, then, is this pair who go on to become the prototypical pagan deities for the christian Franks and Anglo-Saxons. Thus the Indiculus Superstitionum et Paganiarum, as mentioned above, refers Jovis Mercurius, Mercurius Jovis form that specifically to the worship of and and, given and for immediately Christians West the most the recognisable representatives of paganism of the in the eighth century, the use of ponar and Wodan in the Old Saxon Abrenuntiatio Diaboli be learned but by can very plausibly the writer to provide a understood as a misguided attempt for heathens just vernacular abrenuntiatio-formula or semi-christians, as the canon which later Codex in Palatinus Latinus (see The ponar instructs appears above). choice of and 577 Wodan could reflect his or her knowledge of the learned equation of ponar with Jovis and Wodan with Mercurius which circulated in Francia and England in the eighth century, as Mercurius deities. It knowledge Jovis well as seems of the use of and as prototypical pagan likely that he or she also possessed the understanding Diaconus (see section 3.2,,above) Paulus in evidenced the writings of

Mercurius deities Jovis that were universal pagan and

(although known to Germani by different names). If the writer did possess any real knowledge of the specific religious affiliations and observancesof those for whom the formula formula. Saxnot her by his indicated is intended, to the this probably addition of or was Saxnot's relevance to the Saxons, which is made obvious to us (and no doubt also to the likely, it discussed Diabob) his Abrenuntiatio by above, that the as name, renders writer of the

Saxons did, it Saxons; it for did intend this text alsomakes plausiblethat the usewith writer in fact, worship a deity or ancestorfigure known asSaxnot prior to their conversion.This in localised fit and ethne-specificcults proposed section 2.3.1, well with the model of would Abrenuntiatio based is however, One the that this suggestion not on should stress, above. Diaboli, which only revealsthe ideasof a christian writer about his or her audience(who may but he heathen by been have the on only preservation time was writing), the she or not even Wodan, Saxons. ponar Saxnot the and with the connection evident the name's and name of deities form Saxons by been have these part of a the at all, as worshipped not need moreover, formula before being into heathenism Frankish idea learned this was which probably came of

92

Langobard in the context of and written, semi-heathenisms.

aswe shall see-

Alamannic heathenisms and

The learnedconception of Wodan as a universalGermanic deity, equivalentto the Roman Mercurius, circulated, then, not only within Carolingian court-circles, but alsoacross Continent Nor his he Frankish in is the this the stageof confined to areamore widely. Corpus Glossary, be development,for he also appears, in which may around this time, the English seenasan reflex of this continental scholarlytradition.

Frankish-Kentish Woden: Glossary Mercurium The Corpus and 3.4 Classicism

The CorpusGlossary is a large, alphabeticalglossaryof the late eighth or early ninth 46The Latin-Latin, but is there are a considerable number of century. glossary mostly lemmata glossedin Old English, one of these pairsof lemma and glossreading`Mercurium be In to a much-extendedversion of woden'47. terms of textual tradition, the glossaryappears Epinal-Erfurt Glossary. Woden This with the constitutes the earliestextant equation of Paulus have been Mercurius in England, and appears to written around the time of Diaconus'sdeath. At the time of his death, Paulushad not completedhis Historia Epinal-Erfurt (and, following i it, the Corpus Glossary Langobardorum,which, like the 48 Herren finds (see Etymologiae Isidore's Glossary), influence above). section 3.2-, of shows the Epinal-Erfurt `mini-renaissance' in Glossary Corpus Glossary in the evidenceof a and the Wessex Kent in England in in Graeco-Roman particular and mythology -of studies pagan,

4bSee Michael W. Herren, The Transmission and Reception of Graeco-Roman Mythology in Anglo-Saxon England 670-800', Anglo-Saxon England, 2-7(1998), 87-103(p. 103). 47An Eighth Century Latin Anglo-Saxon Glossary,ed. by J. H. Hessels (Cambridge: Cambridge University Press, 189o), headword M197. Epinal Erfurt Glossary,ed. by J. D. Pheifer (Oxford: Oxford University 48Old English Glosses in the Mythology, Graeco-Roman Reception Transmission Herren, `The liv; Press, 1974), p. pp. of and 9o-9i.

93

4' The Corpus Glossary, however, beginning is the the eighth century. of generally -- around dated to the end of the eighth century, and therefore cannot be seenashaving beencopied as development, its have been in this part of antecedents although may compiled the context of this mini-renaissance.
The precise origins and significance of the Corpus Glossary'suse of this equation are difficult to ascertain. Herren's thesis assumesthat the alphabetical glossarieswith which he deals were compiled, or at least modified, in response to immediate pedagogical requirements. The mythological comparisons of this group of glossarieswere written in order to help both Latin This in texts. students understand mythological references pagan and christian fits idea in the that seems a very plausible assumption, and with general alphabetical glossaries for help intended in to answer a need with the were general works of reference order to Latin The interlinear translation of various texts. glossing, widespread use of marginal and however, as well as of glossae should perhaps caution us against assuming that collectae, It help intended to with standard curriculum texts. seemsmore alphabetical glossarieswere likely that they were intended to aid more advancedstudents or scholars with texts which English is Old in In glosses were not regularly studied. either case,the assumption that Latin lemmata. intended to translate and explain their alphabetical glossaries were

however,is that one cannot rule out the possibility of The difficulty with glossaries,
Woden-Mercurius It is equation possible, therefore, that the use of the mechanical copying. in the Corpus Glossaryis due to mechanical copying of an exemplar produced much earlier in Herren's if in the context of mini-renaissance; this were the case,then the eighth century, the equation need not despite its appearancein this glossaryhave been in common

Paulus England in Anglo-Saxon around the end of the eighth century. scholarly use Diaconus's use of the equation in his Historia Langobardorum, however, seems to reflect a Frankish in the understand use of probably we should circles, and common scholarly notion living Glossary Corpus in scholarly tradition which the as very much part of a the equation Kent Francia Kent. The influence Francia in both in this renders on of and was available

49Herren, 'The Transmission and Reception of Graeco-RomanMythology', p. 103.

94

5The equation of Wodan with Mercurius have been may, then, connection very probable. late English Frankish in nothing more than a scholarlycommonplace eighth-century and Wodan Germanic figure, the scholarly circles, reflecting re-use of asa classical mythological figures Latin The literature. Corpus Glossary, like mirroring the mythological the of classical Historia Langobardorum,indicatesnot that eighth-century authors sawWodan asbeing in Mercurius in his but to some way similar mythological characteristics, rather that eighthin Wodan Mercurius, focus for century authors saw, a pre-existing equation of with a Germanic. their creating a classical past of own, a past at once prestigiousand unmistakeably It is in this context, moreover, that we should probably placethe developmentof the
Germanic vernacular names for the days of the week. These loan-translations of the Latin have for days been by the the thought to names of of week many scholars provide evidence 51 fourth Wodan CE. The the existence of a pan-Germanic cult of as early as the century be however, it here lies in dating this, the these to problem with names, so will of useful Germanic Latin briefly development the counterparts. sketch names and their of the

Romans The seven-day from borrowed by the easterntraditions. planetaryweek was According to CassiusDio, writing early in the third century CE, the tradition had been begun by the Egyptians, and wascommonly usedby the Romans in his day,although it was, 52 Bickerman him, seesthe planetaryweek as a relatively recent adoption. according to Augustus, West in time the and receivingofficial recognition under of entering the 53 Stern Persian in (following Celsus) it Constantine, but suggests tradition. that originated de les 'qu'elle dans Egyptian in milieux syncretistes est nee origin, particular prefers the

5oSee Wilhelm Levison, England and the Continent in the Eighth Century (Oxford: Oxford University Press, 1946), pp. S, 95 and io6-07. The importance of ties between southern English kingdoms and Francia prior to the eighth century is discussedin Vera I. Evison, The Fifth-Century InvasionsSouth Franks Sutton Hoo', People Wood, `The Ian in (London: Athlone, in Thames and 1965),and of the Hayes Sawyer, Wood Peter Ian by Europe Essays Honour in Northern Places in ed. and soo-rs-oo: of and Niels Lund (Woodbridge: Boydell, 1991),pp. 1-14" Helm, for instance, in his Altgermanische Religionsgeschichte, loan-translation the sees of the names as (11.1, Anglo-Saxon migration z31-3z). taking place prior to the 52Cassii Dionis Historiarum Quae Supersunt,ed. by Boissevain, c, 405-o6 (book 37, chapters i8-i9). 53E. J. Bickerman, Chronologyof theAncient World (London: Thames and Hudson, 1968), p. 61.

95

1'EgypteheNnisee, dansle cours desdeux dernierssieclesav. J. i54.This is not -C. implausible, although, as Stern himself notes, the planetaryweek's`temoignages les plus de 1'Occident le dernier J. ', anciensviennent et ne remontent pasavant siecleav. -C. 55 Italy influenced by Greece. suggestingthat the planetaryweek prior to was perhaps The planetaryweek must havebeen known to, and usedby, Germani who servedin
Roman lived in Roman Empire (on the army, or who or spent substantial periods the whom, From became Empire see section 2.3.z, above). this whenever the system common within (certainly by Cassius Dio's day, and probably somewhat before then), individuals such would have used the system in their daily lives. Most importantly, however, they are unlikely simply to have forgotten and abandoned the system if they returned to their homes in Germania. We do not know if any Germanic groups had a week-system before the introduction of the (a seven-day week planetary week in most areas,but a numbered week among the Goths), but it seems that any prior systems were thoroughly eclipsed by the seven-dayweek by the time for Germanic of our earliest written sources the use of the seven-day week. One might tend to assume that the loan-translation of the day-names into Tiw's

11

day, `Wodan's day, and so on, took placeaspart of the initial process of reception of the This borrowing does likely, however; in foreign seven-dayweek. not seem chronological a liable foreign least The is initially. terminology, at to use the system, one exigencies of trade different languages for that of use common names speakers and military servicerequire likely important lexical elementssuch as the daysof the week; a Germanic trader is scarcely `Martis his have dies' into to to wanted to translate something more appropriate own religious Latin he learn he have the name so that could make would wanted to affiliations, rather Bavarian Romans. This by form is borne dealings it his in the out practical use of with `Eritag' for `Tuesday; here we havea planetaryday-namebasednot on Latin `Martis dies',

illustrations, Institut Franais Henri Stern, Le Calendrier de 354. Etude sur son texte et ses d'Archeologie de Beyrouth: Bibliotheque Archeologique et Historique, SS(Paris: Imprimerie Nationale, 1953),P. So. 55Stern, p. 50.

96

but on Greek "Ap&og `71p. pa' -

56 has loan-translation. As which not undergone one might

for Germanic from interactions Mediterranean (see the the expect with evidence pre-christian in that this the the section 2--3.2,, above), suggests processesof reception of planetary week Germania were various and localised, and did not necessarilyinvolve immediate loanday-names. We doubt translations of the the existence of universal might also reasonably `great gods' among the Germani around the first centuries BCE and CE; the model for late heathenisms The in classical and early medieval proposed section z. 3.1would support this. in Germanic day-names is, the the relative uniformity of gods referenced then, a calqued if loan-translations day-names to that the considerable problem we wish suppose were of the development first BCE fifth CE. This in the through to the of sometime a century century familiar fit deities however, very well with the uniformity, and the referenced, would, development of the loan-translations in christian Germanic contexts sometime in the seventh The heathen deities CE. day-names or eighth centuries are essentially a catalogue of the familiar to Carolingian and early Anglo-Saxon intellectual circles; Paulus Diaconus, for instance, mentions Wodan, Frea and ponar Tiwaz is only missing.

Although it is possible, then, that the names of the days of the week were calqued in Germanic languages as early as the first century BCE, as a result of trade and military It likely. does seemsconsiderably more plausible that these calques not seem contacts, this Graeco-Roman deities in in the with which equation of were created a scholarly context heathen deities was of particular interest. Eighth-century scholarly circles in Francia and in England provide, as the foregoing discussion suggests,the most plausible context for this day-name calques. creation of the

The scholarly understanding of Wodan around the late eighth and early ninth Although Frankish idea Wodan is this of availableat this time. centuries not the only view of due be in Wodan -- whose seventh-centurysources course- was clearly will considered important, and was availableat least by the early ninth century in England, there was a rather

16See Helm, AltgermanischeReligionsgeschichte, ILz, z38.

97

different use of Wodan (or rather, Woden) known to English circles.This wasWoden's use in Anglo-Saxon lists. This as a royal ancestor-figure royal genealogies tradition and regnal has its origins at latest in the early eighth century, when the first extant referenceto Woden in Bede's Historia Gentis Ecclesiastica Anglorum. as a royal ancestorappears

Uoden, de 3.5 cuius stirpe multarum prouinciarum regium genus duxit: Use Bede Woden Anglo-Saxon in originem and the of Genealogies

Bede'sHistoria Ecclesiastica GentisAnglorum is the earliestof the eighth-century

literary sources for Wodan. Bede mentions Woden as the progenitor of the royal house of

Kent:
Duces feisse perhibentur eorum primi duo fratres Hengist et Horsa; e quibus Horsa postea occisus in bello a Brettonibus, hactenus in orientalibus Cantiae filii Erant Uictgilsi, insigne. habet autem partibus monumentum suo nomine Uoden, de Uecta, Uitta, cuius stirpe cuius pater cuius pater cuius pater 57 duxit. multarum prouinciarum regium genus originem Their first leaders are said to have been two brothers, Hengest and Horsa; of Britons, has battle by killed in Horsa the and a was afterwards whom Now Kent day. his in in to this very name the eastern parts of monument Wecta, father Witta, father Wihtgils, was was whose they were sons of whose family Woden, from father of many whose stock the royal was whose has its traced origin. provinces

in 731or Bede, who died in 735,seemsto havecompletedhis Historia Ecclesiastica Woden failure Bede's Harrison Kenneth has that to mention was a arguedthat 732.58

57VenerabilisBaedae:Historiam Ecclesiasticam GentisAnglorum: Historiam Abbatum: Epistolam ad Ecgberctum una cum Historia Abbatum Auctore Anonymo, ed. by Charles Plummer, 7,vols (Oxford: Clarendon Press, 1896), 1,31-32 (book i, chapter is). My translation. 18On the date of Bede's death see Walter Goffart, The Narrators of Barbarian History: AD 5To-8oo: Jordanes, Gregory of Tours, Bede, and Paul the Deacon (Princeton: Princeton University Press, 1988), `Hic Ecclesiastica have Historia in himself Bede to the 731: est claims completed 242. p.

98

heathen deity provides evidence for the view that Woden was in fact a historical ancestral figure, who only later came to be seen as a deity. 59Leaving aside the fallacy of supposing that deity be thought of as an ancestor, the extant references to Wodan simply do a could not also bear Harrison's discussed Wodan is (rightly not out position; as above, clearly understood or `great in Jonas Bobbio's the wrongly) as a god' works of other eighth-century writers. of Vita Sancti Columbani, moreover, explicitly describesWodan as a deity (this seventh-century is discussed in section 3.6.3, below). Eric John has suggested,on similar grounds, that text Bede's genealogy tracing the Woden-descent of Hengest and Horsa is effectively an interpolation, having been foisted on Bede (he believes) by a Kentish nobleman who was too important for Bede to refuse.6 He interprets Bede's silence on the divinity of Woden as an indication that Bede did not want to mention Woden at all, since he knew that he was a heathen god. There are some obvious problems with this. Bede does mention other heathen deities for instance Eostre and Hreda in De Temporum Ratione is there and no

Bede Woden's did know it highly evidence that of status, although one might think probable 61 he did. that

John is right, nevertheless, does fit to note that the passage given above well with not Bede'saccount of King iEthelberht of Kent's genealogy in book two, chapter five:62
Erat autem idem Aedilberct filius Irminrici, cujus pater Octa, cujus pater Oeric cognomento Oisc, a quo reges Cantuariorum solent Oiscingas Cujus Hengist, filio Oisc invitatus qui cum suo cognominare. pater a Uurtigerno Brittaniam primus intrauit, ut supra retulimus. 63

s9Woden', in Famulus Christi, ed. by Bonner, pp. 35I-56. 60Eric John, `The Point Studies Woden', Archaeology History, (1992), Anglo-Saxon in and s of 127-34 (pp. 12-9-30)-

inpraesentiarumuniuersaestatusBritanniae, [...] dominicaeautem incarnationisanno DCCM' (Venerabilis Baedae: Plummer, (book Historiam Ecclesiasticam, by c, 351 S,chapter2.3)). ed.

" On Hreda and Eostre seeBedaeVenerabilis Opera:Pars VI7Opera Didascalica, Corpus z, Temporum (De Ratione, Christianorum SeriesLatina, 123B (Turnholt: Brepols, 1977), p. 331 15)62 John, p. 119. 63Venerabilis Historiam Ecclesiasticam, Baedae: (book by Plummer, My i, ed. 2, chapterS). go translation.

99

However that sameRthelberht was the son of Eormenric, whose father was Oeric, surnamed Oisc, after whom the kings Kent are accustomed to style [i. Whose Hengest, Oiscings. Oeric's] father first themselves was e. who Britain Vortigern's invitation his Oisc, described entered at with son as we above. John seesthis as evidence for the previous passage Bede's being `He have not own work: must known the contents [of the earlier passage]but in II, S he deliberately ends the genealogy with Hengest'64. This argument ignores the fact that the passage in Book z, chapter 5 clearly refers to the previous passage `ut supra retulimus' ignore it; rather than trying to or gloss over

fact, is, in Hengest hingethe reader encouraged to connect the two passages, and to use as a figure to link up the two halves of the genealogy. Yet, asJohn correctly notes, Bede does

Kentish Oiscingas, `descendants that to themselves state contemporary royalty referred as of Oisc'. Given Bede'sattempt to connect the two halvesof the genealogy, however,we should Bede deliberately understandthis not as contradicting the earlier portion of the genealogy, he Kentish but how which wished to suppress, simply asa statementof royalty referred to Bede's Woden-Hengest If is Bede's in day. it themselves genealogy which we anything, not Kentish Hengest-Aedilberct but his should assignto a genealogy. source,
This presents a considerable difficulty for the assumption -- made by many scholars, including Eric John Woden Anglo-Saxon in that was a royal ancestor pre-christian

England, whose divinity conferred special authority and prestige on those claiming descent from him. If the royal house of Kent referred to themselves as Oiscingas in the early eighth Oisc figure looking to a named as their ultimate century, this suggests that they were ancestor, the source of their royal status and authority.

What, then, hasWoden to do with Oisc?Bede'sgenealogyof King YEthelberht, WodenIf his back Oisc, internal to tracing sense. one considersthe ancestry makesgood Hengest genealogy,however,it becomesapparentthat this genealogyis lessstraightforward; Oisc, in fact, may havevery little to do with Woden. A careful considerationof the Woden-

" John,p. i29.

100

Hengest genealogymay, then, provide us with some indication of how and why Woden Kentish feature Bede's in comesto genealogy. Woden-Hengest Bede's The completesequence Voden Vecta is: of genealogy > > Vitta > Victgils > Hengest. The name Vecta is identical with the name Bede gives(in the 65 Hengest-genealogy) his Isle Wight. This lead to the samechapter as of might one to Vecta his Isle Wight, but fact for Isle Wight that to the the that the suspect gave name of of 66 Vecta fourth by Eutropius, half in the second was alreadycalled writing century. of the Clearly, Vecta's namewas createdfrom that of the Isle of Wight, not vice versa.That the Vecta Latin based Anglo-Saxon being is name on a place-namerather than name a real early Vitta, form does in in may also render us suspiciousof which not appear this all manuscripts 67 Historia but Ecclesiastica. One dismiss this asscribal error, one should of the might easily Vitta fact insertion is in the that a post-Bedan which perhapsconsider alternativepossibility in to the was added genealogy order to createtwo name-pairswhich exhibit the two-stress Vitta Vecta, Victgils. The English Voden found Old in alliterating pattern and and poetry: latter Vitta, in Vecta the of some manuscriptsof similarity of the names and the absence and Bede, may, in fact, be a symptom of Vitta having been invented in post-Bedantradition on by Vecta. That if the model of the addition, we acceptthat there was one, wasnot made includes `cuiuspater' Bede can be evidencedfrom the fact that Bede'sLatin genealogy between eachname, which would break the rhythm. The use of alliterative half-lines in As Chambershasremarked,however,the West Latin would also be unlikely in any case. 68 Chambers Old English Woden forms from Cerdic to Saxon royal genealogy verse. regular be because in is would such versegenealogies arguesthat the genealogy constructed this way bear in This is but in true, mind that the we should easierto remember pre-literate societies. influence in tend to which were verse,or were verse-like, would pre-existenceof genealogies

6sVenerabilis (book Plummer, by Baedae: Historiam Ecclesiasticam, r, 31 t, chapter is). ed. 66 Eutropii Breviarum ab Urbe Condita, ed. by Carolus Santini, Bibliotheca Scriptorum Graecorumet Romanorum Teubneriana (Leipzig: Teubner, 1979),P. 47 (book 7, chapter i9).
67VenerabilisBaedae:Historiam Ecclesiasticam, (book Plummer, by 1,31 i, chapter is). ed.

101

if the way that later genealogies were compiled or composed,even they were compiled or Bede's Kentish in literate for in two composed a context. genealogies are a perfect case point, both could form verse-like genealogies in Old English:
Aedilberct Irminricing Irminric Octing Octa Oiscing Victgils Vitting Vitta Vecting Vecta Vodning

If, as seemslikely, the Oisc-genealogy in was existenceprior to the compilation of the Woden-genealogy,then we havehere a clear exampleof the influence of the verse-like in literate contexts. It is, then, quite traditional genealogyon the compilation of genealogies possibleevenprobableVitta is insertion into this genealogy. that a post-Bedan

The picture which beginsto emergeis, asone might expect,of a legendarygenealogy.

More specifically, the genealogy seemsto reflect legends about origins. This may seem but here Hengest for dealing obvious, one should note the plural, we are not only with as an figure for Anglo-Saxons, first Anglo-Saxon, but the the originator also with and original Vecta as a figure probably involved in a lost aetiological narrative for the name of the Isle of Wight. An example of a very similar narrative, creating an Anglo-Saxon incomer with a Anglo-Saxon be in seen the similar name to explain a pre-Anglo-Saxon place-name, can Chronicle's explanation of the name Portsmouth. 69At the same time, we should also he is `Uoden, Bede Woden figure very explicitly connects with origins; whom consider as a de cuius stirpe multarum prouinciarum regium genus originem duxit' (`Woden, from whose drawn its has house origin'; my translation). of many provinces stock the royal

It would seem,therefore, that this genealogy to a constitutes a constructedresponse Kentish desire for origins; regional origins in the caseof Vecta, whose presence may reflect (Hengest), English Isle Wight, and continental origins origins of and wider claims on the is essentiallyan attempt to genealogise (Woden). Bede'sWoden-Hengest genealogy several England Hengest, first Saxon in the arrival origin myths. in the therefore some sense and

69 Two of the Saxon ChroniclesParallel, ed. by John Earle and Charles Plummer, 2 vols (Oxford: Clarendon Press, 189z-i899), I, 14-I5.

Stories 68R. W. Chambers,Beowu An Introduction to theStudyof thePoemwith a Discussion of the of Offa and Finn, 3rd edn (Cambridge:CambridgeUniversity Press,1963),pp. 316-17.

102

Anglo-Saxon original -

is re-imagined asa royal ancestor,although he may not originally

have been thought of asanything other than the first Saxonto arrive in Britain. It is hardly houses if Anglo-Saxon be indeed, to that would wish most not all royal associated surprising, This Hengest Anglo-Saxon. first within genealogies use of probably with the archetypal developedin a southern context, given the connection of Vecta to the Isle of Wight, and in Kentish Bede's fact likelihood that version of the tradition, the all a context, given the Kentish. is earliest extant version, One might wonder, then, how Woden comesto form part of this genealogical
is function Woden's As indicated as a within this tradition above, conflation of origin myths. figure predating the adventusSaxonum, and therefore relating to the continental Germanic figure Wodan Just Paulus Diaconus later in of antiquity, the eighth century uses as a as area. however, Paulus, in from Bede differs belonging Bede Woden here to the past. as so presents figure, fully-euhemerised deity, but Woden Germanic as a version of a classical not as a using imagined simply as a man, with no suggestion of divinity. At the same time, Bede's Woden is Similar but he belongs genealogical also to another place. not only to the past, continental; (and, Hauck by did Wodan not exist on the continent, and the attempt traditions concerning following him, John) to find evidence for traditions of Wodan-descent in Gothic, Saxon and 70 Gaut, Langobard ethnogeneses tracing descent from figures named or compounds of -Gaut, 71 'man', does not address the simple but insuperable difficulty that Gaut may simply mean and in independently, have been as contexts an numerous possibly therefore, used, may, 72 find decidedly be It for legendary to strange also would progenitor. a appropriate name Langobard in distinct distinct Wodan featured in two ethnic names, the roles, with two

Families, Noble Medieval Kindred Associated 70Karl Hauck, `The Literature of House and with Medieval Nobility: The in Nobility, Satires Illustrated from Eleventh- and Twelfth-Century on the Century, Twelfth Sixth f Germany France the ed. and to the Studies on the Ruling Classes i om and of North(Amsterdam: Studies, Selected Ages: Middle Timothy Europe in Reuter, by 14 the trans. Woden', Point `The John, (pp. 68-69). p. i2,8. of Holland Publishing Company, 1978),pp. 61-85 University (Berkeley: Dunlap, J. Thomas by Goths, " Herwig Wolfram, History of the Ind edn trans. Press, California 1988),p. Zi. of Gothic `the Gaut, Wodan 72Indeed, Wolfram rejects the supposedequivalenceof arguing that and (p. different Odin-Woden Gaut and III). gods' were

103

history, once as Gaut the royal ancestorand once asWodan the tutelary deity. This strongly Woden English, is is, that this use of peculiarly genealogical and moreover,a useof suggests Woden as a figure characteristicallyboth continental and ancient, that is, pre-christian. Bede, Woden has then, as part of a christian re-imagining of the precompletely euhemerised focussed (contemporary) Saxons is on the christian past which specifically continental -- or Antiqui' ('Old Saxons'),as Bede calls them73 `Saxones _ asnon-christians and asancestors, Wodan. Paulus As have in we and on their supposed connection with seen relation to
Diaconus's uses of Wodan, and the Old Saxon Abrenuntiatio Diaboli, such a use of Wodan Wodan be based idea than of on anything more as one of the typical need not a scholarly heathen deities (see sections 3.2 and 3.3, above). Indeed, it seemsscarcely credible that Bede indeed deity he Woden's if fail heathen the object of to comment on was status as a could Bede fact is in It England. is difficult heathen in to escapethe conclusion that significant cult Saxons imagining the old concerned rather with an act of Saxons the contemporary continental and his ancestors who are at once

like deriving the their origins, rather as indeed, and, is, in

Langobards, from the paradigmatic heathen deity, Wodan. Bede's Woden -

Woden as he appearsin subsequent Anglo-Saxon regnal lists and royal genealogies-

fact, firmly rooted in continental christian understandings of Woden, and need not owe heathen English cult. anything to

Woden The evidenceof Bede, then, suggests was not well-known asa not only that deity in Anglo-Saxon England, but also that the figure Woden could, in Bede'sday, be used framework; Woden learned Bede does figure. within a not attempt to place as a royal ancestor he do him (nor Mercurius Woden does he to so, since expect one would with not equate he deity heathen Bede does Of deity). Woden does not present not equateany course, as a in eighth-century there early was already clearly although equivalent, mentions with a classical
73He calls them this in the specific context of discussing the adventusSaxonum in book one, chapter discussing in Plummer, by Historiam Ecclesiasticam, Baedae: fifteen (Venerabilis 1,31), as well as ed. (Venerabilis five, book Saxons in ten and eleven chapters the among continental missionary activity ); Plummer, Ecclesiasticam, by the Historiam continental the of Baedae: connection 302. 1,2,99 and ed.

104

Anglo-Saxon England Kent in interest in particularly the classical authors and --- some 74 The figures. interest in classicalmythology in early eighth-century Kent their mythological is perhaps important for understanding the development of Woden as figure in an ancestor Kent, for it shows an interest in the past and in origins which might well motivate the use, as figure heathen deity In an ancestor, of a who was a elsewhere. the context of the AngloSaxons' interest in their continental origins (fostered by Bede's treatment of the not only in Historia Ecclesiastica, but also by Kent's strong links with the continent, the subject Francia), by Christians deity particularly with this the one might see as adoption of a who by probably was not worshipped their ancestors, but who was thought by the Franks - and, indeed, by Anglo-Saxon ecclesiasticsin Francia be by `contemporary to their worshipped Old This Saxons. insertion Woden into the ancestors', the adoption, and associated of a royal Woden, is he have been genealogy, at once proves the non-divinity of to since supposed a human being, and at the same time appropriates the authority inherent in him as a figure of antiquity.

The eighth century usesWodan in differing but interconnectedways.The FrankishWodan English Woden centred tradition of asa great god contrastswith the use of asa royal Yet English is dependent its despite the tradition the tradition, ancestor. continental on Woden different deity. These in desire to traditions to refusal portray also share, ways,a asa figure Wodan him in to the past, to use who, variousways,providesauthority or relate asa legitimation, whether in terms of ethnicity or royalty, or through the construction of a form have Clearly, been Germanic heritage. too readyto acceptsuch modern scholars of classical The heathen direct in cult. eighth-century usesof and simple way to evidenceas relating a Wodan are not simple reflections of his importance and usewithin pre-christian cults and

Saxons with heathenism and with the ancestry of the Anglo-Saxons could hardly be expressedmore clearly. 74On this see Herren, `The Transmission and Reception of Graeco-Roman Mythology, who points Southumbrian in in Northumbria, interest `Bede, despite the working classicalmythology, out that, (p. by 103). all this' was unimpressed

105

by interests but of eighth-century authors shaped the concernsand societies, are necessarily With in intended this mind, an examination of the pre-eighth-century and their audiences. Wodan help for development to the establishnot only something of may actual evidence of Wodan, but how for it, influenced the the cult of cult, and the seventh-centurysources also how in they this eighth-century christian authors, and misunderstoodor misrepresented cult; how Wodan in wasre-invented the eighth century. short,

Wodan Vodano Suebi, Cult Columbanus, the of 3.6 and the nomine:
The eighth-century evidencefor a cult of Wodan is by no meansasstraightforward as
it might at first be thought to be, and it seemsthat the picture of a widespread eighthfalse be England Anglo-Saxon Wodan Continent in one. may a and on the century cult of Paulus Diaconus uses Wodan as a paradigmatic heathen deity, whose worship can be ascribed At known heathen, be the to anyone to without regard to their actual religious convictions. Paulus's fulfil deity in Germanic Wodan he classical who can as a same time, appropriates Bede Mercurius Mars in Jovis fulfilled by classicaltexts. or or own writings the roles Wodan as part of an ancient undertakes a similar re-appropriation of pre-christian and

Diaboli does Abrenuntiatio Old Saxon Even Anglo-Saxon not relate the past. continental -Continental Saxons; once as simply as one might supposeto the practicalities of converting Anglo-Saxon Frankish demonstrates paradigm of and the power of the again, this text heathen cult centering around a pan-Germanic Wodan (and ponar), but we might very Continental by deities indication it doubt the the culted that of gives a reliable reasonably Saxons in the eighth century.

If eighth-century authors constructedWodan as a great god, a figure of antiquity and derive did heathen they whence the remains: question religious cult, of object pan-Germanic What in Wodan Were Wodan? idea this way? was their they wrong to view their of our and

decidedly As is, Wodan? The for scarce. evidencewas, and evidence the actual cult of

106

have been by the to we shall see, moreover, evidence seems the radically misinterpreted did the eighth-century authors whose reading of so much to shape subsequent evidence both in the the medieval period and up to the present day. readings of evidence,

Logaj Wodan: The Nordendorf Brooch Alamannic Cult 3.6.1 ore and

The earliest pieces of documentary evidence for the existence of Wodan date to the CE. These Jonas Bobbio's Vita in Sancti Columbani, seventh century to are a reference, of Suebi worshipping Wodan, a runic inscription on a brooch found near Nordendorf, and two Langobard found Fredegar's Chronicon Origo Gentis in versions of the ethnogenesis and the Langobardorum. As mentioned above (section 3.2), Paulus Diaconus used both the Chronicon Origo Gentis in his Historia He Langobardorum. doubt the and composing was also no aware Vita Columbani, Wodan Mercurius. him the of which provided with with the equation of The one source which we have and which Paulus did not is the inscription on the Nordendorf brooch, and it is this source which probably provides the earliest reference to Wodan. Krause dates the inscription on the Nordendorf brooch to the beginning of the 75The inscription is, `logapore wodan wigiponar awaleubwinix', and reads seventh century. 76 Krause's Not Krause, dialect. in Alamannic all scholars would agreewith an according to identification of the dialect, however; Helm, for instance, has suggestedthat the name Wodan should not show <d> in an Alamannic dialect, but <t>? It is clear, nevertheless, that improbable it is Germanic individual, by brooch that and not the a was owned at some point it was also buried with a Germanic individual. Clearly, the runic inscription on the brooch Germanic by least Germanic for intended speaker a made speaking owner, or was at a was

75Die Runeninschriften im lteren Futhark, ed. by Wolfgang Krause, Abhandlungen der Akademie der Wissenschaften in Gttingen, Philologisch-Historische Klasse, series 3,65, z vols (Gttingen: Vandenhoek & Ruprecht, 1966), t, 2-94(no. 151).See also illustration i below, basedon Krause's (I, 65). plate photograph 76Die Runeninschriften, ed. by Krause, i, 29z. 77Helm, AltgermanischeReligionsgeschichte, ii. z, 277.

107

Figure i- Nordendorfbrooch

108

literacy. Given Nordendorf location it the who also possessed runic of seemsmore probable brooch by Alamann, Helm's (our despite than not that the was owned an misgivings knowledge of early Alamannic dialectsis, after all, very limited). It is also probablethat the individual or individuals involved in the creation and use of the inscription were heathens,as we shall see. Krause divides the inscriptions into two sections,A and B, the former constituting first lines division, inscription, latter fourth This line. is three the the the of the a reasonable fourth line in is in the since written the other way up relation to the other three, a strikingly different style, and evencollideswith the third line of sectionA. It seemsquite clear that B is A, from inscription, judge it the way that collideswith section was a separate which, to
be (although itself later A than section would scratched a close examination of the artefact his It interpretation definitely). is in of the text, odd, then, that necessaryto settle this point Krause seems to connect the two sections, rendering them as an invocation of Wodan, Thunor (and perhaps Logathore) by Awa and Leubwini: `Logathore ("Rnkeschmied"), Wodan, Weihe-Donar! B. Awa (und) Leubwini (wnschen Glck, oder: schenken))78.It does not, however, seem likely that A and B were both scratched at the same time, as a single inscription. While this suggestsstrongly that Awa and Leubwini were not originally intended inscription, figure in the to one might also note that the presenceof two names suggeststhat The inscription. later declaration possibility of ownership was not the purpose of the a simple it intended inscription later to connect the names contains with was remains, then, that the brooch. deity-names the the on

Krause also notes another feature of the brooch which might be significant in any `Man Germanic Wodan; trinity: of a namely the possiblepresence examination of the cult of Zeile Gtter Nennung der drei fr die jedes A Abschnitt dem da in beachte, eine eigene This he connectswith the three godsof the Old SaxonAbrenuntiatio benutzt wurdei79.

78Die Runeninschriften, ed. by Krause, 1, Z94" 79Die Runeninschriften, ed. by Krause, [, 294.

109

80Adam Bremen's Gesta Diaboli (on which seesection 3.3,above). Hammaburgensis of Ecclesiae 81 important but also mentions a cult grouping of three particularly the gods, identification of a species Germanic heathen of trinity either here or in the Old Saxon Abrenuntiatio Diaboli seems The discussionof the Abrenuntiatioaboveshows that unjustified.
this formula probably reflects christian assumptions about Saxon pre-christian cult (see section 3.3, above). Equally, in the caseof the Nordendorf inscription, Krause's argument for based layout line for divine be trinity a on a planned of a new each name must rejected, since logapore runs right up to a raised area on the brooch (part of the fastening mechanism), line for different from is in necessitating a new wodan, while wigiponar written a style again that used for the previous two lines. The o-runes of logapore and wodan show a marked tendency towards continuation of the upper limbs beyond the point where they meet the lower limbs, while that in wigiponar does not. The w- and p-runes of the first two lines, have in is high in moreover, the rounded pockets, and that of the w-rune wodan up on stem, contrast to wigiponar, which shows angular pockets in both p and w, and a pocket running from the top to the bottom of the stem of the latter. For these reasons, it seemsmore likely brooch but inscription, that the two, three actually contains not separatestages of one Thunor, Leubwini, Awa logapore the naming one and word or name and one giving followed by the name of Wodan. It has been suggestedthat logapore is a form of the name 82The L6 urr, a deity known from much later Norse mythographic sources. difficulty main forms Lo''urr logapore from is development that the and of the with this suggestion a single Germanic be linguistic by known developments among the early root cannot supported any dialects. As Dwel points out, `eine Anknpfung an den nord. Gtternamen Lo''urr < *LogJ,orr, von dem wir kaum mehr als den Namen wissen, bleibt hypothetisch'83. Another interpretation `schemer, is loga]iore is that cunning one' or possible an epithet meaning

80Die Runeninschriften, Krause, by 1,294. ed. al Quellendes Jahrhunderts, Bresslau Pertz, (book Schmeidler, and p. 470 4, 9. und ii. ed. by Waitz, chapter z6). 82Die Runeninschr f en, ed. by Krause,t, 294. 83Klaus Dwel, Runenkunde, Sammlung [1968]), Metzler (Stuttgart: Metzler, ist edn p. 40.

110

Wodan. This based here being is is `wizard', view appliedto on the perhapseven which Glossary logeper logor in Cleopatra in English Old the word or and the appearance of the (Cleopatra) Harley Glossary (Cleopatra for and cacomicanus and asan equivalent marsius Harley).'4This is linguistically more plausible,although the infrequencyof the word, and its 85 Old English, in do leave for doubts its in appearance only room about precisemeaning Germanic It dialects. likely, however, it does here imply that that the continental seems Wodan inscription the writer of saw power or expertise, as possessing somespecialised he have being in this although or she may not seen as magical character. Klaus Dwel, however,has arguedthat logapore is a pejorativeterm, and that the
inscription in fact represents a sort of epigraphic abrenuntiatio-formula which identifies 86This is Wodan and ponar as deceivers. interesting certainly an and plausible suggestion, but, if correct, it does not necessarilyfollow, as Dwel has suggested, that the inscription is `nicht mehr ein Zeugnis fr [... ] einen heidnischen Gtterglauben der Alamannen'87. The be inscription back brooch, it position of the on the where would not seen when the of the brooch was worn, is a common feature of runic inscriptions of this date in the southern 88 Germanic area, has it been for that this the and may reflect plausibly suggested use of runes

84See the Dictionary Old English Corpus, //ets. umdl. umich. edu/cgi/o/oecof <http: idx?type=bigger&byte=36z977oo&qi=logeT&qz=&q3=> (Cleopatra, `marsius') [accessed z6 November zooz) ; <http: //ets. umdl. umich. edu/cgi/o/oecidx?type=bigger&byte=36os7679&qi=logT&qz=&q3=> (Cleopatra, `cacomicanus') [accessed z7 November zooz] ; <http: //ets. umdl. umich. edu/cgi/o/oecidx?type=bigger&byte=36678133&qi=logeT&qz=&q3=> (Harley, `cacomicanus') [accessedz7 November zooz]. 85See Dwel, Runenkunde, p. 40. 86See Klaus Dwel, `Runen und Interpretatio Christiana: zur religionsgeschichtlichen Stellung der Kraft, ed. by N. Kamp and J. Wollasch Bgelfibel von Nordendorf I', in Tradition als historische (Berlin: de Gruyter, i98z), pp. 78-86. Dwel replies to criticisms of his original article in `Runeninschriften als Quellen der germanischen Religionsgeschichte', Ergnzungsbnde zum Altertumskunde, S (1991), 336-64 (pp. 356-59), and repeats the claim in Reallexikon der germanischen `Frhe Schriftkultur bei den Barbaren: Germanische Runen, lateinische Inschriften', in Die Alamannen, ed. by Karlheinz Fuchs and others (Stuttgart: Theiss, zooi), pp. 491-98 (PP. 494-95). 87Dwel, Runenkunde, znd edn, Sammlung Metzler (Stuttgart: Metzler, 1983),p. iz8; quoted in Dwel, `Runeninschriften als Quellen', p. 359" Dwel, `Frhe Schriftkultur', p. 491.

111

inscriptions of a private or personal nature; given that female graves are far more likely than has it been inscriptions, have held that to also male graves contain runic argued runes may 89The domestic inscription back brooch placement of the some significance. on the of this indicate dedication, denigration, or then, than may, personal worship, propitiation rather as for Denigration inscription. but the motive the the remains a possibility, nevertheless,

domestic or private characterof the inscription is hardly consonantwith the very public It inscription that the nature of an abrenuntiatio-formula. seemsmore plausible representsa personalresponse whether positive or negativeto non-christian cults with which the

brooch's owner had a personalconnection. Dwel could be correct, therefore, in reading the inscription asan attack on Wodan and ponar, but it seemsunlikely that he is right to reject the inscription asevidencethat thesedeitieswere worshippedamong the Alamanni. The Nordendorf inscription clearlydoesnot provide much information, in itself,
Wodan If Dwel Alamanni. it is the the about cult of right, suggeststhat the cult of among Wodan was already losing adherents in the early seventh century. This is hardly certain, however. We can at least state with reasonablelikelihood that Wodan was worshipped by Danube Alamanni just in the region around the some north of the western end of the beginning of the seventh century CE. It remains to consider the evidence of the Langobard Sancti Vita Columbani, which, as we shall see, may provide us with a ethnogenesis and of the Nordendorf inscription. wider context within which to understand the

89See Giorgio Ausenda, `Current Issues and Future Directions in the Study of the Merovingian Ian by Period', in Franks and Alamanni in the Merovingian Period: An Ethnographic Perspective, ed. Wood, Studies in Historical Archaeoethnology, 3 (Woodbridge: Boydell, 1998),pp. 371-453(p" 42-w).

112

Quemfanatice Wodano: The Langobard Ethnogenesis 3.6.2 nominant as Evidence for Cult

We now come to the Langobard ethnogenesis. Already in the seventh century this 90 in Gentis Origo Langobardorum, Frankish history two texts, the appears anonymous and the 91A known as the Chronicon,and attributed to Fredegar. version of the narrative also appears in Paulus Diaconus'sHistoria Langobardorum, (on in the eighth century which waswritten The found Fredegar's Chronicon follows: in is which seesection 3.2, above). version as
Langobardorum gens, priusquam hoc nomen adsumerit, exientes de Scathanavia, que est inter Danuvium et mare Ocianum, cum uxores et liberis Danuvium transmeant. Cum a Chunis Danuvium transeuntes fussent bellum Chuni, inferre, interrogati conperti, eis que Bens eorum conarint a At introire ille terminos praesumerit. mulieris eorum praecipunt comam legarint, habitum capitis ad maxellas et mentum quo pocius virorum hostium simulantes plurima multitudine ostenderint, eo quod erant mulierum longa. barbae Fertur desuper instar valde coma circa maxellas et menturn ad falangiae dixisse: Langobardi', his `Haec vox sunt quod ab gentibus uterque fertur eorum deo fuisse locutum, quem fanatice nominant Wodano. Tunc

90The Origo Gentis Langobardorum is extant in three manuscripts. The Maritensis manuscript dates Cavensis Mutinensis tenth the the the to to the tenth century, the century, and manuscript end of beginning itself, however, have been to to the the the text appears of eleventh century; manuscript (Scriptores Rerum during half the second of the seventh century composed sometime Langobardicarum, p. i). 91According to Ian Wood, the Chronicon appearsto have been completed in 659 or shortly thereafter, from Columbanian be the the work of a monk within monastic tradition; certainly, this and may individual knew Jonas of Bobbio's Vita Sancti Columbani at a remarkably early date (Ian N. Wood, desInstituts fr sterreichischen Geschichtsforschung, (1994), `Fredegar's Fables', Verffentlichungen 3235966 (p. 360)).

113

`Qui instituerat nomen, conciderevictoriam'. Hoc Langobardi clamassent: Pannoniae invadunt92 Chunus prilio superant,partem The race of the Langobards,before it took up that name, leaving Scandinavia (which is betweenthe Danube and the Ocean),crossedthe Danube with their When Huns learnt the that they were crossingthe wives and children.
Danube in order to wage war on them, the Huns asked what people dared to borders. But he hair fix heads their enter the advised their women to of their to their cheeks and chins, by which stratagem, simulating the appearanceof more men, they might appearmore numerous to the host of enemies, because the women's hair around their cheeks and chins appearedto be very long beards. It is reported that a voice from above on each side of the troop said: `These are the Langobards', which was said by these peoples to have been their god speaking, whom they fanatically name Wodan. Then the Langobards shouted: `Let him who appointed the name grant us victory. In this battle they overran the Hun and invaded the area of Pannonia.

This seventhcentury version of the Langobard origin myth is very different from the
Origo Gentis Langobardorum Paulus Diaconus's Historia in the and versions given Langobardorum. The versions given by Paulus and the Origo are not identical, but they are The be in the to tradition. considered as representing same version given sufficiently similar Origo Gentis Langobardorum follows: is the as Tunc Ambri et Assi, hoc est duces Wandalorum, rogaverunt Godan, ut daret dicens: `Quos Respondit Godan Winniles sol surgente victoriam. eis super Gambara duobus filiis Eo ipsis dabo tempore cum antea videro, victoriam'. Winniles, Ybor Agio, id rogaverunt qui principes erant super suis, est et Fream, [uxorem Godam], ut ad Winniles essetpropitia. Tunc Frea dedit Winniles et mulieres eorum crines solutae consilium, ut sol surgente venirent Tunc faciem barbae in similitudinem et cum viris suis venirent. circa luciscente sol dum surgeret, giravit Frea, uxor Godan, lecturn ubi recumbebat Et ille faciem fecit aspiciens vir eius, et eius contra orientem, et excitavit eum. faciem; Winniles habentes ipsorum et ait: crines solutas circa vidit et mulieres `Qui sunt isti longibarbae? Et dixit Frea ad Godan: `Sicut dedisti nomen, da illis et victoriam'. Et dedit eis victoriam, ut ubi visum essetvindicarent se et

93 Langobardi Winnilis haberent. Ab illo tempore vocati sunt. victoriam

92Fredegarii et Aliorum Chronica. Vitae Sanctorum, ed. by Bruno Krusch, Monumenta Germaniae Historica, Scriptorum Rerum Merovingicarum, z (Hannover: Hahn, 1888),p. no (book 3, chapter 65). My translation. 93ScriptoresRerum Langobardicarum, pp. 2--3.My translation.

114

Then Ambri and Assi, the leaders of the Vandals, asked Wodan to give them Winnili. Wodan `whosoever I see first at victory over the replied, saying: I At the sunrise will give the same time Gambara, with her two sons, victory. that is, Ybor and Agio, who were the chiefs of the Winnili, asked Frea, the Wodan, be favourable Winnili. Then Frea gave counsel that to wife of to the Winnili the should come, at sunrise, and their women, with their loosened hair around their faces in the likeness of beards, should come with the men. Then, at sunrise, Frea, the wife of Wodan, turned bed in the around which her husband was lying, and causedhis face to be towards the east, and she him And he, looking Winnili woke up. the out, saw and their women with loosened hair around their faces; and he said: `Who are these longtheir beards? And Frea said to Wodan: `As you have given them a name, give them And he the also victory. gave them the victory, so that where their troop was defended they seen, themselves and had victory. From that time on, the Winnili have been called Langobards.

Paulus Diaconus'sversion can clearlybe seento derivelargely from this, or a very similar, version of the narrative:
Refert hoc loco antiquitas ridiculam fabulam: quod accedentesWandali ad Godan victoriam de Winnilis postulaverint, illeque respondit, se illis victoriam daturum quos primum oriente sole conspexisset.Tunc accessisse Gambara ad Fream, uxorem Godan, et Winnilis victoriam postulasse,Freaque consilium dedisse, ut Winnilorum mulieres solutos crines erga faciem ad barbae Godan similitudinem conponerent maneque primo cum viris adessentseseque fenestram ille videndas pariter e regione, qua orientem versus erat solitus per fuisse. Atque ita factum Quas Godan aspicere, conlocarent. cum oriente sole longibarbi'? Tunc Frea dixisse: `Qui isti conspiceret, suns subiunxisse, ut Godan Sicque Winnilis quibus nomen tribuerat victoriam condonaret. Haec habenda. Victoria digna sunt et pro nihilo victoriam concessisse. risui 94 de hominum, enim non potestati est adtributa sed caelo potius ministratur.

Chronicon. He Fredegar's Paulus appears Origo have known both to the and often and used Fredegar, has however, he here in that the version given we can see preferred contradicts and The fact, logically Origo. Chronicon is, in in the version of the myth given the more fake beards Fredegar identifies than the the other versions. aswhat they clearly satisfactory be must -a has been lost, In military stratagem. the other versions,their military purpose

115

inserted into the narrative merely to provide the and, as a consequence,they seem awkwardly Langobard; Wodan have Winnili first if had had would still name seen the they even not their women impersonating men. Fredegar's version, in which the use of a purely military happens Wodan's better The difficulty lies in to excite stratagem attention, makes sense. deciding why it makes better sense.It is possible that this version is closer to the original form of the myth (which we must supposeto have made some sense).It is equally possible, however, that Fredegar knew the myth in a confused form, and attempted to rationalise the The latter for fact Fredegar's narrative sequence. explanation would not account the that Huns WinnililLangobards, the version gives as the enemy of the Paulus Origo while and the

identify them as Vandals. It is difficult to seehow this could be a rationalisation on Fredegar's part. This suggeststhat the Chronicon more or less faithfully reproduces an (possibly for is it this unknown oral) source also possible that this represents an narrative; form Langobard Paulus Origo. The by the earlier of and the origin myth than that given latter sources could, however, merely represent a garbled version of a separateearly form of the myth. In these circumstances, dating the origin of the Langobard ethnogenesis becomes

The Chronicon 's Langobards statementthat the were attempting to crossthe problematic. Danube at the time of their battle with the Huns would placethe origin of the narrativeto from River Elbe Langobards the upper reaches of the and migrated south sometime after the Huns in Danube, were activearound the crossedthe and, of course,to the period which the 95 Origo (as Gentis in The Vandals Danube. the asenemies western reachesof the

95The Langobard migration from the Elbe occurred sometime after the third century CE, when the Langobards are mentioned by Ptolemy as inhabiting the region around the northern reachesof the Elbe (Claudii Ptolemei Geographia,ed. by Carolus Mullerus, i vols (Paris: Didot, 1883),i, 258-59 (book z,, chapter ii, section 8)). The Chronicon's description of their original homeland as being between the Danube and the Ocean fits well with this, although it names this region Scandinavia. The Huns were active around the stretch of the Danube contiguous with Pannonia in the early- to Europe The Peoples Heather, Peter Huns, by (E. Thompson, The A. of rev. century mid-fifth (Oxford: Blackwell, 1996), p. 30).

94Scriptores See in (book Rerum Langobardicarum, 8). translation section3.2. i, chapter pp. 52--53

116

Langobardorumand the Historia Langobardorum), however,would indicate an earlier dating, Langobards' in movement southwards. at a more northerly point the Both of these datings are perfectly possible,and there is little to choosebetween
Nordendorf In light Wodan it find inscription, is interesting is the the them. to that of very Langobards Alamanni, Alamanni by the since the are, associatedwith the as well as the Strabo Suebi, Langobards by known identified seventh century, otherwise as the and the are 96 Suebic It Chronicon is tribes. the as one of certainly suggestive,moreover, that the Wodan's Langobards dramatic with the region of associates onomastic and military aid to the Nordendorf brooch Danube, is in the western end of the which the which the same region (see It Wodan, buried is possible that the cult of was above). and the associatedethnogenesis, Elbe, before Langobards' from the the migration and that this cult was practised not arose Suebi in Langobards but in the area. the this region, also more widely among only among the Equally, it is possible that the ethnogenesis, and perhaps also the cult of Wodan, arose among Danube, in Suebi Langobards in sometime the area around the western end of the the and Wodan for The CE. lack fifth the existence of the to seventh centuries of any evidence does Danube the second support not the region the than or outside seventh century earlier Danube is Wodan for in for the so scarce the around century seventh the evidence possibility, Elbe, from from it is the time, around and an earlier that evidence that all conceivable quite has perished.

Wodan is Wodan? Langobard does What, then, tell us about the ethnogenesis bestowal with the certainly portrayed asa god associated least with the achievement or at

The Langobards. he is told as the episode with connected clearly of military victory, and

96As Quast points out, `Nach wiederholten Kmpfen gegen die Goten in den Jahren 466-47o wurden Whrend den Alamannen Jahre Im verbndet. die Sueben vernichtend geschlagen. 470 waren sie mit Stmme Jh. des Ende bis Schriftquellen Sueben und Alamannen in den stets als eigenstndige S. zum Jh. die Beginn des 6. Mal letzte das Donausueben die mit werden explizit 477 genannt werden Einzelgrab Quast, `Vom (Dieter Volk fr dasselbe Suebi zum Begriffe Alamanni und gebraucht' Fuchs, by Alamannen, Die in Jahrhundert', pp. 171ed. Friedhof: Beginn der Reihengrbersitte im 5. The Loeb Jones, Leonard Horace by Strabo, Geography The See (p. trans. ed. and of 183)). 9o (book Putnam, York: New Heinemann; (London: 7, 111,156 1917-1932-), Classical Library, 8 vols chapter i, section 3).

117

in the Historia Langobardorumand the Origo could well be thought to relate to this tribe's Wodan, but the version given in the Chronicondoesnot original reception of the cult of development. The Chronicon, in to seem reflect such a religious narrativewhich appears the in fact, does not involve Wodan to nearly the samedegreeasthat given in the Origo and the Historia Langobardorum.The Chronicon's Langobards `fanatice that the statement nominant Wodano', however, doessuggestthat Wodan is to be understoodasbeing or at least, --having been importance Langobards deity. the to of considerable as an ethnic

As suggested above, this centrality of Wodan in Langobard pre-christian society was key factor in development Paulus Diaconus's Wodan idea the very probably a of as a major of Germanic deity (see section 3.2). Taken together with the evidence of the Nordendorf brooch, the Langobard ethnogenesis clearly suggestsa cult of some importance amongst the Suebic tribes, and since such tribes were, by Paulus's day, identifying themselves as `all men', Alamanni, looking it is Frankish the out at this tribal perhaps not surprising that circles, Frankish deities be heathen to grouping of the periphery, took their characteristic of all nonGermani. however; There for do likewise, fact little is the that what christian no reason us to Langobards Wodan Alamanni for the and exists all convergeson seventh-century evidence Jonas Bobbio's Wodan likely indicate to that of was specific to these related tribes. seems Vita Sancti Columbani, moreover, not only reinforces this position, but also offers an indication of how the eighth-century equation of Wodan with Mercurius arose, and how Wodan was re-imagined from a local, tribal recipient of cult to the christian Frankish deity. heathen archetypal

118

Vita Identifying Mercurius: The Sancti Columbani Cult 3.6.3 and the of Wodan

The Vita Sancti Columbani-

Susa, by Jonas Bobbio, written who of a native of

Bobbio foundation by its 617, the entered monasteryof a relatively short time after around Columbanus be interesting because it thought might particularly seemsto contain the 97 Wodan: description in earliest extant of a ritual used the worship of
Sunt etenim inibi vicinae nationes Suaevorum. Quo cum moraretur et inter habitatores loci illius progrederetur, repperit eos sacrificium profanum litare XX modia amplius velle, vasque magnum, quem vulgo cupam vocant, qui Dei Ad in accessit minusve capiebat, cervisaplenum medio positum. quem vir Deo Vodano illo Illi de fieri nomine, sciscitaturque, quid vellint. suo aiunt se Ille Mercurium, litare. pestiferum opus ut alii aiunt, autumant, velle quem frustra fragore dissolvitur insufflat, et per audiens vas miroque modo vas cum dividitur, visque rapida cum ligore cervisaeprorumpit; manifesteque datur intellegi diabolum in eo vasefuisse occultatum, qui per profanum ligorem caperet animas sacrificantum98 For there are Suebic tribes in that locality. While he stays there and goes finds he inhabitants that they want to perform of that place, about among the in is large called a cup the vernacular vessel- which a profane sacrifice, and a [a is and contained around twenty modia modium an ancient corn measure; full in is fluid the capacity of this vessel substantial] -- was placed the middle, God At beer. went up and askedwhat they might want which the man of of They done be say that they want to sacrifice to their god, called to about that. Wodan, whom, as others say, they call Mercurius. He, hearing this appalling design, blew on the vessel, and, in a wondrous manner, the vesselbroke up beer liquid flowing force in irrecoverably, the the of the and was split and broke through it; and he was clearly given to understand that a devil had been

9' On Jonas's life seeJonasdeBobbio,Vie de Saint Columban et de SesDisciples,trans. by Adalbert de Abbaye de Bellefontaine, Vogue and Pierre Sangiani, Vie Monastique, 19 (Begrolles-en-Mauges:
1988),

98Passiones Vitaeque Sanctorum Merovingicarum, ed. by Bruno Krusch, Monumenta Germaniae My Hahn, (Hannover: Rerum Merovingicarum, Scriptorum translation. Historica, 1902), p. ioz. 4

pp.

19-20.

119

hidden in that vessel,who, by meansof the profaneliquid, would steal the souls of those sacrificing.
Nor is this passageinteresting only as a description of a ritual; it may also be the earliest Wodan Mercury. It is however, extant source which equates with that not universally agreed, this equation is to be traced back to Jonas. According to Krusch's edition, the phrase `quem Mercurium, ut alii aiunt, autumant' is omitted in manuscript A3 (Codex Mettensis n. 523, a " De Vogue Sangiani `ces the manuscript of eleventh century). that and suggest mots PAUL etre inspiree de DIACRE, semblent une glose Hist. Lang. I, 9i100. If this is the case,

it merely shifts the problem of the origin of the equation of Wodan with Mercury into the Historia Langobardorum. In the eighth century and some respects, this seems a plausible Corpus Glossary is take, to the position since also eighth-century, and also makes this English Frankish in interest that the equation, suggesting eighth-century and revival of Graeco-Roman pagan mythology, argued above to be a plausible cultural context for the development of the names of the days of the week, may, as part of that development, underlie development Germanic Graeco-Roman (see deities the of this and other equations of and This be however, is correct, view unlikely to since the numerous section 3.4, above). Vita Columbani include the equation, and although none of the almost all manuscripts of 101 families these manuscripts predates the ninth century, the early separation of manuscript of likely is it that this text makes original, or entered very early a phrase so uniformly-witnessed 102 At be if it in the same time, would the manuscript transmission. strange eighth-century on Wodan Graeco-Roman mythology chose, on mythographical grounds, to equate scholars of Jovis, ponar in Mercury; the with make good sense equation of with other equations, such as deities (in this case,thunder-wielding) of the equated terms of the mythological attributes

ioo Ve de Saint Columban, trans. by de Vogue and Sangiani, p. 159(note 9). lo' Passinnes, Krusch's for least Krusch, is, dating by Rohr has one at pp. 36-45. ed. pointed out that (Christian Quelle: late Rohr, historische `Hagiographie als manuscript, somewhat too Ereignisgeschichte und Wunderberichte in der Vita Columbani des Ionas von Bobbio', Mitteilungen Geschichtsforschung, 0995), (PPdesInstituts fr sterreichische 103 243-44))" 229-64

99 Krusch, by Passiones, ed. pp. ioz and 42.

120

Wodan, however, have This is with who, aswe with one another. not the case seen,seemsto be related with victory in battle in the Langobard ethnogenesis, indeed, his through and, him These fury. name, which seemsto present asthe controller of angeror are not the Mercurius, have to they were; attributes proper nor would eighth-century scholars supposed it seemsprobable, therefore, that the equation did appearin the Vita Sancti Columbani from the start; eighth-century scholarswere perhapsmerely following the lead of this influential seventh-century text. A careful considerationof exactlywhat the Vita Columbanistatesat this point
The Vodano `Illi Deo is: this confirms position. crucial sentence aiunt se suo nomine, quem Mercurium, ut alii aiunt, autumant, velle litare'. De Vogue and Sangiani seem to take this like `They to phrase mean something say that they want to sacrifice to their god, called Wodan, whom others, as they say, call Mercurius'. This is a strained reading of the sentence, taking `alii' as the subject of both `aiunt' and `autumant'; this involves understanding `ut alii (as be), but it aiunt' not as a self-contained adverbial phrase one would expect to as introducing a new subject for the relative clause introduced by `quern'. The sentence can be 'They Wodan, read much more naturally as say that they want to sacrifice to their god, called Mercurius'. De Vogue Sangiani, in whom, as others say, they call and other words, Wodan, Suebi have deity understand the sentence as claiming that the a called whom other Mercurius. It likely, however, is that the to people refer as seemsmuch more sentence stating Wodan Suebi deity (Jonas knows be have that the this to the case),whom they also a called It Mercurius (Jonas Jonas has from `alii'). is intends this call sources, possible that a distinction between his own correct understanding that the Suebi have a deity called Wodan, Suebi Mercurius. deity This incorrect is the this understanding of others that call and the Jonas have `alii however, incorrect, since claimed that what could easily aiunt' was unlikely, but does not do so; the most straightforward reading of this passage is that Jonas understood Wodan Suebi have deity both Mercurius. the to a whom they called and

102 On the considerable variation between manuscripts of the Vita in different families see Rohr, p. 244.

121

The phrase`ut alii aiunt', then, may refer to oral or written transmission.The use of `alii' suggeststhat other writers were intended, sincewere the transmissionoral, one would individuals `ut homines It lines is the the or something along of aiunt'. namesof expect either just possiblethat Jonashad Tacitus in mind, who claims of the Germani that `Deorum [... ] Sueborum Mercurium Isidi (`they Mercurius colunt maxime pars worship et sacrificat' [... ] Suebi Isis'), but have the most greatly of gods some of the also sacrificeto we no 103 Jonas Tacitus's Ages. Germania, knew Middle little-known in that the evidence which was This passage from the Germaniagives,moreover,no clue asto why Jonasshould understand Tacitus's Mercurius asbeing Wodan. This is probablynot, therefore, one of Jonas'ssources, Jonas's in have and no other texts are extant which could servedas source this matter.
In the absenceof such texts, it is difficult to determine why Jonas ascribesto the Suebi the use of Mercurius as an alternative name for Wodan. Herren has discussed the Irish interest in classical Latin literature, which may also have influenced the monasteries set up by

Bobbio long Saint Columbanus.104 The importance of the classics has, been moreover, at 105 for favourable development This interest the environment of would clearlyprovide a noted. but is hard Roman deity, it learned between heathen deity to see pagan a and a a equation for Wodan. If Jonas be how Mercurius might haveappeared to to an appropriateequivalent Wodan did not differ significantly betweenthe Langobardsand the Alamanni in the seventh Langobard Wodan in in the asrecorded the ethnogenesis, century, then the presentationof Origo GentisLangobardorumand in Fredegar'sChronicon,would seemto indicate that Wodan was a deity concernedwith war and tribal identity. This would hardly make Mercurius, asmentioned above,an obviouschoice for an equivalentfor Wodan. There do, however,exist some indications of the nature of Wodan's connection with Mercurius -- although not the Mercurius one might expectin the seventhcentury. The

104 Michael W. Herren, `Literary and Glossarial Evidence for the Study of Classical Mythology in Ireland A. D. 6oo-8oo', in Text and Gloss:Studiesin Insular Learning and Literature Presentedto JosephDonovan Pheifer, ed. by Helen Conrad-O'Briain, Anne Marie D'Arcy and John Scattergood (Dublin: Four Courts Press, 1999), PP-49-67.

103 Cornelii Taciti OperaMinora, Germania,chapter9, sectionsi-z. My translation.

122

Langobards battle, heathen in in Langobard is Wodan. After the the patron of ethnogenesis, their conversion, however, such a patron would no longer be appropriate, and it is to Saint Mercurius that the Langobard upper echelons appear to turn for a patron saint; and a patron distinctly It decidedly saint, moreover, of a martial character. would seem strange that an Egyptian martial saint known but little in Occident, in the then almost exclusively a -and Vita in Sancti Basilii his Vita focus become the short episode rather than own a of - should Langobard aristocratic cult. This is not simply learned whimsy on the part of some Langobard scholar, choosing a saint whose name connected him Wodan. Paulus with Diaconus, for instance, demonstrates a very hostile attitude to the story of Wodan's granting indicates Langobard be Christians. of victory, which that such a choice would unlikely among The clear importance of Saint Mercurius to the Langobard nobility is amply demonstrated, 106 in his literary it. the translation of moreover, output associatedwith relics and the Furthermore, the influence of this saint is apparent in one of the Langobard battles described by Paulus Diaconus, when they claim to have cynocephali fighting for them, just as Saint Mercurius himself is accompanied into battle by his two companion cynocephali in some 107 Vita. Friedman his has Historia in this the eastern versions of already noted narrative Langobardorum as `a variant form of Mercurius' dog-headed helpers', but fails to make the between Saint Mercurius's Langobard in this narrative and connection special role some circles. 108

It would not, then, seementirely fanciful to suggestthat Wodan's role astutelary deity of the heathen Langobards (or at leasttheir nobility) cameto be filled, subsequentto depended Saint Mercurius. This have by simply upon a might conversionto christianity, learned equation of Wodan with the pagandeity Mercurius; certainly one could conceiveof a

Mercurii `translationes s. plures exstant narrationes' notes that (P. 573), of which it prints an example on pp. 576-80. In addition to his activities in the translation itself, moreover, Arechis may also have had a hand in the composition of a PassioSancti Mercurii (ScriptoresRerum Langobardicarum, p. 574). 107 See John Block Friedman, The Monstrous Racesin Medieval Art and Thought (Cambridge, MA: Harvard University Press, 1981),pp. 71-72.

iosPassiones, Krusch, by ed. p. ii. 16ScriptoresRerum Langobardicarum

123

in by situation which acceptanceof christianity was strengthened the provision of a patron saint who his Roman Mercurius the as namesake of counterpart provided a good

for Wodan. Not is because incompatibility this the only substitute unlikely of relative of Wodan with the god Mercurius, it also seemsoverly simplistic, for two reasons. Firstly, Jonas Bobbio be least his Suebi to that, seems of claiming at according to some of sources, the (otherwise known as Alamanni; see section 3.6.2, above) called Wodan by the name Mercurius. This indicates that the equation of Wodan and Mercurius (whether saint or deity) was known to heathens or semi-christians as well as learned christians. Secondly, the hallmark Saint Mercurius depicted who are a cynocephali of are on a sword-sheath and a (not forms stamped metal sheet unlike the one which part of the sword sheath), usually identified as Alamannic work of the seventh century, found at Gutenstein (Baden, Germany) 109 Obrigheim (Pfalz, Germany). Langobards The Alamanni and of the seventh century and Wodan, deity but deity identification to the that appear share not only of with a also an This connection is also, figure called Mercurius, and a connection of battle with cynocephali. Mercurius. legend Saint It in of would not seem entirely as mentioned above, present the implausible, then, that the Mercurius-figure identified with Wodan by heathen or semiMercurius. Mercurius deity, Saint Roman but Suebi the pagan was not christian It may be, therefore, that the efforts at accommodation of missionaries and

Alamanni Langobards createda popular equation among these and christianisersamong the Indeed, Saint Mercurius, Saint Mercurius. Wodan asa martial saint, clearly with peoplesof Wodan does Langobards' for battle-deciding fit better than the messenger-god the makesa have Saint Mercurius Mercurius. One could go further, and speculate that could actually been the sourceof the deity Wodan; this saint wascapableof deciding the outcomesof battles, he would havebecomeavailableto the Alamanni and Langobardsat around the time Elbe from dating him Wodan have if the to after the migration as arisen we see would when

iosFriedman, p. 72-. For a photograph of a replica of the Gutenstein sheath,seeDieter Quast, `Opferpltze und heidnische Gtter: Vorchristlicher Kult', in Die Alamannen,ed. by Fuchs and others, pp. 433-40(P. (figs Gaimster, Gutenstein finds, Obrigheim both drawings For 14). i8 13 and the p. see of and 437)"

124

(seeabove),and his power to control and unleashthe savagery his of companion cynocephali have bestowal justified him `controller frenzy, Wodan. the would the on of epithet of the This possibility fits well with the idea, expressed by Quast, that theseAlamannic-Langobard Wodan: to the cynocephali are related cult of
Die Wolfskrieger in der Literatur werden nordischen als Ulfhednar (Wolfshuter) bezeichnet und im Zusammenhang den bis heute mit (Brenhutern) [... ] Ihren sprichwrtlichen Berserken genannt. Ursprung haben Berserker und Ulfhednar in den Odin kultischen geweihten Kriegerbnden. Sie werden in den Quellen als Odins Mnner Krieger oder bezeichnet. 110 This idea stems from the work of Hfler, 111 from its and reaches conclusion very bases; doubt, for instance, that a simple, pre-christian unsound we might very reasonably between Alamannic-Langobard Wodan the connection exists the cynocephali of and the fhe6nar Snorri Sturluson's Prose Edda. It have of u seems,nevertheless,that poor methods here produced a final result which has some validity, at least in that Wodan can plausibly be dog-headed Langobard-Alamannic his in Hfler doubt associatedwith men cult. would no have been horrified, however, to find that this association may indicate that Wodan developed from an Egyptian martial saint. Such a conclusion is, of course, speculative, but it is no less plausible than many common assumptions about the cult of Wodan.

Apart from this surprising indication that the equationof Mercurius with Wodan has but in its roots not in early classical a practicaland appropriateconnection of scholarship, Wodan with Saint Mercurius, Jonastells us little enough about Wodan. There is no here, beer-sacrifice described Wodan human bears the and asa ancestor small suggestionof (on later A human sources which, seesection i. i). sacrificesof certain resemblanceto the Wodan. however, beer is in the seekingto associate ritual with amount of caution required, Jonas, disruption describes by The Vita Vedastis, of a probably also a rather similar miraculous by in a mixed group of christian and very similar gentilis ritus, although this casecarried out heathen Franks:

' lo Quast, `Opferpltze', in Die Alamannen, ed. by Fuchs and others, pp. 437-38.

125

domum introiens, [Vedastis] conspicit gentile ritu vasaplena cervisaedomi Quod ille adstare. sciscitans, quid sibi vasain medio domi posita vellent, inquirerit, responsum est, se alia christianis, alia vero paganis opposita ac gentile ritu sacrificata. Cumque ita sibi denuntiatum fuisset, omnia vasade industria signo crucis sacravit, ac omnipotentis Dei invocato, fidei nomen cum adminiculum, caelitum auxiliante dono, benedixit. Cumque benedictionem cum crucis signo super vasa, quae gentili fuerant ritu sacrificata, premisisset, legaminibus, mox soluta cunctum cervisaeligorem quem capiebant in 112 deiecerunt. pavimentum house, Vedastis full beer in the entering that saw a vessel the of was standing house according to pagan ritual. When he askedwhy they might want a vessel in house for themselves, it was replied that the vessel the placed middle of the before was placed some who were christian and others, indeed, who were pagan, and this vesselwas sacrificed according to the pagan ritual. And when he was informed thus, he busily hallowed the vesselwith the sign of the cross, blessed it, and calling on the name of omnipotent God with the support of faith and the aiding gift of heaven. And when he had spoken forth the blessing, with the sign of the cross over the vessel,which had been sacrificed immediately to the according pagan ritual, the vesselloosened in its fixings, forth liquid beer it floor. the poured and all the of which contained onto It is possible, then, that the beer ritual is not a ritual genuinely used in the worship of Wodan, but a general-purpose heathen ritual used by Jonas to add colour, whether it or not One however, Jonas was appropriate. that could also argue, was aware of a genuine ritual used in the worship of Wodan. He may then have used the ritual indiscriminately in his writings, heathen for, he have it in two caseswhere he whenever a ritual was called or may used only Wodan deity These being was aware that was the worshipped. are not the only possibilities, for moreover, we need not assumethat a given sort of sacrifice was only appropriate to one heathen deity; a beer sacrifice may have been used in the worship of a variety of heathen deities. The variety of possible interpretations of this sacrifice should caution us against Wodan. What it assuming too readily that was a genuine custom of the worship of we can

112 VitaequeSanctorumAevi Merovingici et Antiquorum Jonas of Bobbio, Vita Vedastis, in Passiones Aliquot, ed. by Bruno Krusch, Monumenta Germaniae Historica, Scriptorum Rerum

Hfler, pp. 55-68.

126

literary for Jonas's is is source this custom outside of say that there no obvious writings, and be that this suggeststhat the custom may well a genuine non-christian religious practice, Wodan, In Germanic. this and not evencertainly although not certainly one connectedwith finds Rhenish-ware from large drinking the of extremely vessels connection one should note Roman population centresin Gaul, dating to around the fourth century CE; Symonds have been have if that these may plausibly suggests usedasritual vessels, and, so, they would 113 formed part of Roman or Gallo-Roman paganreligious practice. It is possible,although it be Jonas he that was awareof a paganreligious ceremony,which used as a cannot proved, for heathen his descriptions religious rites. of model The eighth-century usesof Wodan, then, may owe much to the Vita Sancti
Columbani, the Langobard ethnogenesis, and, indeed, the seventh-century cult of Wodan Columbani, Langobards. The Vita Sancti Alamanni reflecting the confusion and among the have brought heathen-christian liminal the equation of area,seemsto and manoeuvring of a Wodan and Mercurius into the literary world of eighth-century Francia. The equation was immediately misunderstood, however, and re-used as part of a strong classicising tendency. At the same time, the cult of Wodan had, by the eighth century, assumed enough distance for be Wodan for to start to christian purposes, re-used and re-imagined and unimportance from providing a Germanic classicalheritage, to creating a figure of the continental past and royal prestige.

have limited in been This is probably in part because the cult seemsto quite geographicalterms; all the evidencerelatesto a relatively small region around the western end frontier Pannonia, Danube Bregenz, Nordendorf, Danube. which playsa with and the of the It is fall Langobard in this this within region. all the ethnogenesis, the of of versions part one incorporation for Alamanni habitation key to their the within up areaof region which was a

Merovingicarum, 3 (Hannover: Hahn, 1896), PP. 399-427 (pp. 410-11).My translation. On the Vedastis, Vita seep. 399. authorship of the 113 See R. P. Symonds, Rhenish Wares: Fine Dark Coloured Pottery from Gaul and Germany, Oxford University Committee for Archaeology, Monograph 23 (Oxford: Oxford University Committee for Archaeology, 1992), p. 70.

127

fifth the Frankish empire, and it was here that the Langobardsfound themselves around the date CE, Wodan, they the took when up the cult of century perhaps and certainly a period been have issue identity, having they the particularly concernedwith would when of tribal from It hardly, Elbe. is it the then, that is this period and this region migrated surprising, be in least in some quarterswhich should associated the seventhcentury -- at with their

ethnogenesis,and, of course,with the cult which was still presentin the region, if not There have been but here it, is Wodan strongly. may never much of where the cult of was practised.

Conclusion 3.7
It is, in fact, in the imaginationsof christian writers of the seventhcentury and
Wodan Our the especially eighth century that chiefly existed. understanding of the cult of Wodan in this period has been guided very much by the writings of eighth-century commentators, whose use of the meagre seventh-century sources available to them was, as we have seen, enthusiastic but misleading. From Columbanus's encounter with a Suebic rite in honour of Wodan, and the confused and confusing versions of an old Langobard aetiological English Frankish scholars of the eighth century created a portrait of a great and narrative, Roman Mercurius. If Germania, such a god really equivalent to the god, worshipped across Wodan What have left indications he has however, we of a cult of clear no trace. existed, Langobards, Alamanni localised still practised and cult, proper to the show no more than a it Danube in the early seventh century, where attracted only around the western end of the Vita be in in from the missionaries of the period to passing the mentioned enough attention Sancti Columbani.

This is, after all, exactlythe sort of cult that we should expectfor a heathen deity. As likely heathen highly it that the vast majority of cults were section 2.3.1argues, seems have been localised, specific to regions or peoplesor both. In some cases cults may even

128

for Wodan The is, family. then, entirely consistent with such a evidence a cult of specific to a is Here have Alamanni to the a cult which very probably specific model. we and the Langobards. The cult was probably not linked to a region by any religious considerations,but it seemsto havebelongedparticularly to the region just north of the Alps, around the Danube. western end of the This is by no meansthe only interpretation possibleof this evidence,but it is at least
as probable as in and the present author's view, more probable than interpretation an

based on accepting that eighth-century authors such as Bede and Paulus Diaconus wrote Wodan from knowledge do than position about a of greater modern scholars, and that they Wodan. If knew to they simply attempted record as accurately as possible what we about accept by large, do and and we that these authors shaped and interpreted and presented

knew Wodan less is to serve their own ends, then this no true of their uses of than what they figure of their uses of any other of the past.

129

Ancestors Authority Figures: The Toden-Obinn 4. and Complex in Late Anglo-Saxon England

Introduction 4.1
The eighth-century intellectual elite, from Bede through to the scholars of Charlemagne's court, had, as we have seen, a profound effect on christian literary depictions Wodan. drew Anglo-Saxon and understandings of authors of the ninth to eleventh centuries heavily on Bede's Historia Ecclesiastica, and on the cultural and textual models of Charlemagne's circle. ' Not surprisingly, then, Bede's use of a Kentish royal genealogy Woden from Woden led key figures Anglo-Saxon becoming in to royal one of the stemming Historia Ecclesiastica It be played such a central proved absolutely that the genealogies. cannot The is far likely but it is development, in than not. alternative view this more certainly role been have important in Bede that was simply reflecting a widespread tradition which would later Anglo-Saxon discourse anyway. This seemsunlikely, since it fails to account for Bede's Northumbrian Bede brief genealogy, which gives a non-Kentish genealogical materials. Woden have been Woden, an may not and thus suggeststhat makes no mention of furthermore, is, This Bede's day. in strongly suggestion ubiquitous royal ancestor formed for houses Bede's by on the suffix -ing, such royal references to names corroborated 2 Such Wuf ingas. Oiscingas of understanding probably reflect an very names the the and as As has been from Woden. descent idea is pointed of antecedent to the royal origins which Oisc Bede to in the of claim recognising genealogy, composite uneasy gives an section 3.5, out

1 See,for example,Janet L. Nelson, `"... sicut olim gensFrancorum nunc gensAnglorum": Fulk's ... Occasion her Janet Bately Wue: Studies Honour in Alfred Revisited', in Alfred Letter to of on the the of Sixty-Fifth (Cambridge: L. Janet Godden Malcolm Roberts Nelson Birthday,ed. by Jane and with
Brewer, 1997), pp. 135-44-

130

be head of the Oiscing dynasty,aswell asWoden's claim. If East Anglia wasruled by kings however, likely it dynasties Wuffingas in that the eighth century, seems generally called (such Woden Oisc Wuffa) local progenitors as and at this time; the use of recognisedunique later be development. a as a master-progenitor must This genealogical tradition is witnessedevenby an insular sourcefrom outside into England, the Historia Brittonum, and the Anglo-Saxon genealogical tradition continued
his Woden illustrations the thirteenth century, when contemporary manuscript of and sons ' Snorri At it influenced this time also attest the continuing currency of the tradition. O inn as a king who migrated from Asia (see section 5.1,below). This Sturluson's idea of however, discussion dissemination itself to a and transmission of of the confine chapter will, England. Anglo-Saxon in this tradition

Nor is this by any meansthe only use of Woden in Anglo-Saxon England.

Scandinavian settlement in the northeastern areasof England in the later Anglo-Saxon Scandinavian figure Anglo-Saxon led increased to an writers of the awarenessamong period O Wodan, by was considered as such sinn, who, although probably not actually a reflex of

English authors. As we shall see in this chapter, the interplay of genealogical traditions -- in figure Woden had become of royal authority and prestige - combined with the a which Obinn deity, heathen Scandinavian created a complex set of as a understanding of the familiar interconnected personally, and responsesamong a group of authors who were often figure between The a of prestige and one of problem of mediating with each other's work. identical in Anglo-Saxon in origin authors, essentially evil who are, the understanding of Obinn Woden and as separatetextual entities. caused these authors to attempt to re-create Obinn in it, Woden, inherent This variety of uses of and the contradictions and indeed, Anglo-Saxon the modern scholar authors -- and, presented bewilderingly with a

Oiscingas) (book Plummer, by 2 Venerabilis Historiam Ecclesiasticam, Baedae: i, 9o z, chapterS; ed. Wuffingas). (book z, chapter is; and ii6

3 Nennius: `British History' and `The WelshAnnals', ed. and trans. by John Morris, History from the Joan (chapter Littlefield, Rowman Phillimore; Totowa: 67 & (London: Sources 31). i98o), p.

131

multiplicitous

figure. Cicero faced a similar problem in his De Natura Deorum, dealing with

the extraordinary and sometimes contradictory exploits and attributes of the Graeco-Roman his deities, Mercurius, to treat was pantheon; solution problematic such as as several separate figures conflated in popular discourse.4 Cicero invented, then, what we may term Mercurius a interrelated figures developed from, complex, a set of and representing aspectsof, a figure We identify different Mercurius otherwise perceived as unitary. might well a rather sort of Anglo-Saxon in England, Mercurius in different complex where appears contexts as a pagan deity, an antediluvian giant, and as a figure of medical magical authority and significance. In last he helps this manifestation, moreover, to create the magico-medical use of perhaps Woden in the Nine Herbs Charm (seesection 4.3.z, below). Woden and Obinn are perceived to be identical by Anglo-Saxon authors, and yet these same authors use and re-imagine these figures in very different ways, essentially re-creating a figure (whom they understand as being figures. In Wodenthis unitary) as a complex of connected sense,then, we can talk of a Obinn complex in late Anglo-Saxon literary culture. This Woden-Obinn complex is also reflected, and perhaps rejected, in a unique Woden idols, Maxims I(B) (see in presentation of which occurs as connected with section To below). different literary some extent this may reflect a 4.4, sphere or audience, especially Woden English in Old is The to two one of only since this references verse. other reference from different belongs in the prose material which appearsto a somewhat context certainly have circulated in learned and, crucially, very high status ecclesiasticaland aristocratic circles. This reference occurs in the Nine Herbs Charm, and can be seen, therefore, as representing a Woden depend him heathen ultimately on an understanding of which may magical use of as a deity. One could, however, argue that this charm occurs in a learned context (though of indeterminate status), and it is entirely possible that Woden's use here depends, as suggested Mercurius above, more upon a traditional equation with in in who appears some contexts

Turville-Petre, `Illustrations of Woden and his sons in English genealogical manuscripts', Notes and (1988), Queries, 2-33, 158-59" n"s" 35

4 De Natura Deorum, ed. by Plasbergand Ax, p. i4o (book 3, chapter 2,2).

132

Anglo-Saxon England asa medical figure -- than upon his being thought to be a heathen deity. Clearly, this complexity and variety in the usesof Woden and O Anglo-Saxon in sinn

England presents some problems of organisation for any discussion of the Woden-Obinn The both lines to these traditions present witnesses of connection and of complex. Obinn. The discussion discontinuity between the various representations of Woden and less different follows is, discrete then, arranged as a series or more or on uses of essays which Obinn Mercurius, figures: deity, his heathen the presentation of these and as a equation with Obinn Woden; Woden from Scandinavicise dissociate the the contrasting moves to and to development of a magico-medical function for Woden; and the synthesising use of Woden in Maxims I(B), which draws in aspectsof most of the other uses apparent in Anglo-Saxon England, and thus leads into a final examination of the connections and disjunctions apparent Woden-Obinn within the complex as a whole.

Oongeba'ten

4.2

Uses Anglo-Saxon Denisc: Late o$rum naman on

Obinn of
lists in The origins of the Anglo-Saxon royal genealogies and regnal from within a very similar cultural context to least or at

havebeen Bede'sHistoria Ecclesiastica

Woden (section As been has as a source use shown, such genealogies 3.5). consideredabove Woden's heathen by do but legitimacy, status as a they appealingto so not of authority and deity; rather they invoke him asa figure of antiquity who connectsthe Anglo-Saxons with their continental origins -a by It is Germanic precisely the originator. classical sort of

Wodan's is deity heathen Wodan's that this madepossible;certainly cult as a weaknessof his is into the within of positioning part necessary god a a man euhemerised as a status in have been but undertaken a at the sametime this positioning could not classicalmodel, It is Wodan's in clearly significant, status as a recipient of cult was still strong. context which little he England, in is Woden culted, where was probably used asa royal ancestor then, that

133

if at all, and is not used in this way among the Alamanni and the Langobards,who seemto have culted him as a tribally important deity. In late Anglo-Saxon England, moreover, the influence of Danish heathenismappears to haveimpacted on the use of Woden in Saxon West The noblemanthelweard genealogicaltexts. Latin his translation, made a

Chronicon,of a version of the Anglo-SaxonChroniclein the late tenth century.' Naturally, he his doing in he but, in translated the genealogies present source, so, madesome curious his alterations to source.
Baelda: Woden's Anglo-Saxon in is the g, who appearsas one of sons royal genealogies, Scandinavian deity Baldr, If despite the not cognate with their similar-sounding names. the Anglo-Saxon figure were cognate with Baldr for a -- which would constitute evidence lengthy tradition, as well as an Anglo-Saxon reflex of Baldr *Bealdor, we should expect

but this never occurs. The only instance where Baldr is more closely approximated is in Ethelweard's Chronicon, which gives 'Balder'6 It is interesting, in this connection, that ., . )Ethelweard also twice switches Woden's son and grandson in his version of the Hengest and Horsa genealogy, and gives the grandson's name as `Vuithar' and `Wither', forms which are Obinn's Old Icelandic Vibarr (one considerably closer to the of sons) than the more usual form Witta. 7 This would suggest that iEthelweard was aware of Scandinavian traditions and harmonise Anglo-Saxon The is them. to the genealogieswith sought suggestion supported, by fact the moreover, Meaney by pointed out iEthelweard `uses that although the

form Woddan once in his Chronicon,elsewhere he hasinvariably the compromiseform wotheni8. This is a particularly strangereshapingof the Anglo-Saxon royal genealogical it literary the the tradition, especially when one considers within context of vernacular O inn in AEthelweard's day. De Falsis Deis, by in its AElfric treatments of versions and

5 The Chronicon has been edited by Campbell, The Chronicle offEthelweard. On the identity of Ethelweard, see The Chronicle ofiEthelweard, ed. by Campbell, p. xiii. , 6 The Chronicle ofiEthelweard, ed. by Campbell, p. 33 (book 3, chapter 4). The Chronicle offf'thelweard, ed. by Campbell, pp. 9 and i8 (book r, chapter 4 and book 2, chapter
2).

134

Wulfstan, aswell as the anonymoushomily known asNapier 42, evidencea circle of literary interest in Obinn (and his equationwith Mercurius), in which)Ethelweard probably played a Deis Falsis De iElfric's providesperhapsthe earliestextant text producedwithin this part. (section below), Rlfric following have been may well 4.2.i, a circle, although, aswe shall see It be first. however, to considerthis text convenient, northern example. will JElfric's De FalsisDeis contains a lengthy passage dealingwith the Roman pagandeity O Mercurius, and making an equation betweenhim and the Danish inn: Mercurius life, Sum man waes gehaten on facenfull dxedum, se waes swipe and swicol on leasbregdnyssa. lufode eacstala and and
pa h4 him to maerangode, one macodan enan lac him and aetwega gelaetum offrodan, him beorgum heagum brohtan to onsxg[ed]nysse. and Des god ways[a]rwyr&e betwyx eallum hxpenum,

Obon Denisc. he is gehatenobrum naman on and Nu secgabpa Deniscan on heoragedwylde

hi pr htab, Pct se louis ware, Jae Mercuries sunu, Pe hi Orion hatab; bocum, hi for Pam Jae we raedabon ac nabbab na riht, Louis Cristenum, pat hetola hoepenum se ge on ge on Saturnes to so&an ware sunu,

beon Pa bec awxgede ne magon and Je Ja ealdanhxbenan be him awriton Puss;

9 Prowungum we gemetab swa awriten. and eac on martira

A certain man was calledMercurius in life, who wasvery sinful and deceitful in his actions, and who loved theft and treachery.The heathensmade him brought into a famous god, and presentedhim with gifts at crossroads and heathens, honoured This hills. high him among all god was on sacrificesto O Danes Now in Danish. in inn, by he is the say called another name, and Mercurius, prr, Jovis, was the son of whom they call their confusion that Obinn; but they (both books in because we read are wrong, whom they call heathen and christian ones),that the loathsomeJovis was in fact Saturn's son, discounted books be wrote about which the ancient pagans cannot and the
8 Meaney, Woden in England', p. no. De Falsis Deis in Homilies of1E'c: A SupplementaryCollection, ed. by John C. Pope, Early English Text Society, o.s. 159 and z6o, 2 vols (London: Oxford University Press, 1967), n, 684-85 (lines 133My translation. 49)"

135

in them; and also the passions of martyrs we encounter the same thing written.

Obinn in this text, it will be necessary In discussingthe presentationof to considerthe Anglo-Saxon It in England. be the text will useful, contexts of production and reception of briefly to sketch the writing and revision of the text, and to considerits possible therefore, According to Clemoes,iElfric's De FalsisDeis was originally intended usesand audiences. 10Pope between lines 99z and iooz. notes that 141-49 were addedat a composedsometime later date." Clemoes ascribes this revision to an authorial re-issueof the text sometime after his from Wulfstan he it text that this expanded worked when wrote ioo6, and notes that was De FalsisDeis.12He notes that this re-issue`promptsthe questionwhether Mfric himself issueda set of his homilies for unspecifiedoccasions and pieceson generalthemes,revising 13 he did. is but for there that that the no evidence concludes purpose', some
On the basis of manuscript evidence, Clemoes suggeststhat the original issue of De Falsis Deis was as an appendix to the Lives of Saints, with which it belonged, as they were `all liturgy, for but for intended pious reading at any reading as part of the not narrative pieces De (for discursive instance Falsis Deis De the non-)Elfrician pieces time'14. and various other XII Abusivis) appear to have been included with the Lives of Saints on the basis that they too " (probably Gatch, A According for intended the to thelweard non-liturgical reading. were iEthelmaer his Chronicon) )Ethelweard son and who wrote the same Catholic Saints, Lives have known an augmented of to commissioned the are Homilies I, presumably also a version or copy of Catholic Homilies II, and be library Such Genesis. that certain can reasonably one and a portions of long books had lords towards enabling way go a would these great other -

Studies in some 'o P. A. M. Clemoes,The Chronology of . Elfric's Works' in The Anglo-Saxons: (London: Bowes Peter Clemoes by Bruce Dickins, Presented History Culture Aspects to ed. and of their
z44). & Bowes, 1959), pp. zIz-47 11Homilies of1? ic, ed. by Pope, i, 147and ii, 673-74. 12Clemoes, pp. 145 and z39 (note z). 13Clernoes, p. 2-39. 14Clemoes, p. zzo. 's Clemoes, pp. zzo-zI. (p.

136

SEthelweardand his son to follow in their own devotions the observances of

monks16

The original version of De Falsis Deis, then, appearsto have been intended primarily as a text for more or less private pious reading rather than instruction of the laity. It may have been but it it for intended to small groups, read aloud seems unlikely that was general It is interesting Scandinavian deals that this text, which consumption. particularly with heathen deities and completely ignores the fact that they once had equivalents in AngloSaxon England, should have formed an appendix to a collection of texts intended for JEthelweard.. lfric's striking failure to mention the English equivalents for the Scandinavian

deities with which he deals is doubly striking in the context of iEthelweard's Scandinavicisation of Anglo-Saxon royal genealogicalmaterials, as discussedabove. The fact his Danish Elfric discussion heathen to the that gods, and makes no mention of confines 17 This Pope by being deliberate Anglo-Saxon is due their to as reticence. equivalents, seen It discern for is difficult be it is to possible such reticence. may a reason correct, although Anglo-Saxon England lies behind in Woden's in that prominent place royal genealogies iElfric's silence, but the fact that he does not mention the names of any of the Anglo-Saxon heathen gods suggests that a more general wish to avoid discussion of the heathen past of the English is the reason. It would also be strange if Elfric were more sensitive about the latter iEthelweard, Woden's in since the was use royal genealogiesthan political aspects of his day. important noblemen of one of the most

}Ethelweard's position is, in fact, a rather strangeone. As an ealdormanhe was a he figure of considerablepolitical power, but at the sametime, assuggested above, seemsto have taken a rather greater interest in the monasticlifestyle than was common in a nobleman patron even a major ecclesiastical '8 The Chronicon Latinity his day. of the unusual of

has also been remarked as reflecting a similar contradiction; while scholarssuch as Campbell

16Milton McC. Gatch, Preachingand Theologyin Anglo-Saxon England: zE (Toronto and Buffalo: University of Toronto Press, 1977),pp. 48-49-

Wu c and

an

17 Homilies of1Ec, ed. by Pope, ii, 716.

18The Chronicle of thelweard, ed. by Campbell, pp. xiii-xv. if

137

Winterbottom and

Latin as a form of hermeneutic Latin, have seen and as a -YEthelweard's highly personal response to this wider trend towards involved syntax florid lexis, Angelika and

Lutz has recently argued that AEthelweard's vocabulary and interest in syntax also reflect an Old English poetry, and an attempt to colour the Chronicon with 19 a vernacular poetic style. This accommodation of the contradictory within life thelweard's and work could account if for his willingness to Scandinavicise the Anglo-Saxon royal iElfric genealogieswhere prefers O-6inn On to keep Woden and clearly separated. the other hand, Frank's plausible claim for Anglo-Saxon attempts, in the ninth and tenth centuries, to construct Scandinavian dynastic links through the addition of Scandinavian figures to their lists, royal genealogiesand regnal 2 furnish Scandinavicising iEthelweard's may more a plausible explanation. tendency, seen in this light, does not reflect a lack of political sensitivity on his part, but the extension of a long-standing Scandinavicising tendency within English genealogical traditions. Perhaps, then, iElfric's De Falsis Deis reflects a sensitivity which is more religious than political. At the same time, the re-writing of De Falsis Deis noted by Pope (see above)

Obinn iElfric's indicates that Mercurius his developed to response and equationwith over Falsis The De Deis Mercurius Danish time. earliestversion of merely equates with the `Obon', taking most of the details about the deity from /Elfric's main source,Martin of 21The statementthat heathens`to heagumbeorgum him Braga'sDe Correctione Rusticorum. however,doesnot perfectlymatch this source,which does not brohtan onsaeg[ed]nyssei22, but Mercurius demons that the who reports connect specificallywith offerings on mountains, frondosis `ut in in sacrificiasibi excelsis montibus et silvis were the pagangods urged men 23 high leafy ('that in to them on they should offer sacrifices mountains and woods'). offerent'

Old English Poetry', Anglo-Saxon England, Chronicon Lutz, 'JEthelweard's Angelika 29 and jog-18;
(2,000), 177-2-14.

(1967), Medium 19 fEvum, `The Style Aethelweard', M. Winterbottom, Campbell, pp. xlv-xlix; 36 of

Chase (Toronto: University of Toronto Press, Ig81), pp. 123-39(PP-126-29). Zl See Homilies ofiE Martini Episcopi Rusticorum De Correctione is in Pope, by 684. it, edited c, ed. Bracarensis Opera, ed. by Barlow, pp. 183-203. u Homilies offE (line Pope, by 684 it, 138). c, ed. 23Homilies ofif lfric, ed. by Pope, 11,716.My translation.

20Roberta Frank, `Skaldic Verse and the Date of Beowulf', in The Dating of Beowuf, ed. by Colin

138

It has been arguedthat, prompted by the generalreferenceto mountain offerings in Martin, iElfric included this asa feature of the worship of Obinn of which he had heard, probably 24If, however,one from from Danish practices. excludes which was considerationthe passage lines `Obon' later, becomes him the of added 141-49, mention a mere equationof with Mercurius at the end of the passage Mercurius. This demonstrates JElfric's knowledge of on Mercurius, Woden have have traditions equating with which one might not expectedto
Corpus Glossary is English instance that the the only eighth-century survived, given of this iElfric, despite before Woden in Anglo-Saxon the equation all the referencesto genealogical It does Mercurius however, demonstrate JElfric that materials. not, was thinking of `Obon'; being if that were the case,one might throughout the passageas the same god as It iElfric be is by impossible that rather expect the equation to stated at the outset. no means Obinn, Danish first in but his involved the cult of given that was referring to a practice Woden Falsis Deis does De version of not seem to show any particular concern with Mercurius his at the end of the passage equation with merely mentioning fact that and the

(as Martin), Mercurius by he details portrayed gives are related to the cult of all the other iElfric based far Woden, it the reference to seems more probable that rather than to that of Danish Rusticorum Correctione De practices of rather than upon offerings on mountains upon been have he aware. may or may not which Mercurius between The later addition to this passage and expandsupon the equation Woden, and compares the Graeco-Roman divine family-tree with the Danish one. This it Danish )Elfric's knowledge need not mythology, although part of on must reflect some indicate that this mythology has an early origin.

His Circle Wulfstan and 4.2.1

24Homilies ofAic, ed. by Pope, ii,

715-16.

139
i

iElfric's use of O sinn in De FalsisDeis, however,is not an isolatedphenomenon. Wulfstan, as noted above,re-wrote the final, extendedversion of iElfric's De FalsisDeis in his own style, rendering the passage discussion follows: under as
Sum man eac waysgehaten Mercurius on life, se wes sw be facenfull j yeah full snotorwyrde swicol on daedumj on leasbregdum. Done macedon pa haebenanbe heora getxle eac heom to mxran gode, 7 aetwega gelaetumhim lac purh deofles lace, heagum beorgum him brohton offrodon oft 7 gelome 7 to loflac. Des oft mistlice gedwolgod wes arwur&e eac betwux eallum haebenum Pam dagum, he Orion is Denisce Nu 7 on gehaten o&rum naman on wisan. pa Denisce heora Joct louis Jae by i secga sume men on gedwylde se were or hatab, Mercuries sunu, Jae hi Obon namiacb, hi forhan Pe ac nabbab na riht, we bocum, haepenum Cristenum, Jaet hetula Iouis raedabon ge on to ge on se
's Saturnes is soban sunu.

There was also a certain man called Mercury in life, who was very crafty and, Him in deceitful his in in although very clever speech, actions and trickeries. heathen for themselves a great god according to their the also made devil, him through the teaching reckoning, and over and again, of the offered him brought sacrifices at crossroads and often various sacrifices of praise on This high hills. false god also was venerated among all the heathen in the Danish days, he in by Odin. is those the the manner and called other name Now some of those Danish men say in their error that Jove, whom they call Thor, was the son of Mercury, whom they name Odin; but they are not right becausewe read in books, both pagan and Christian, that the malignant Jove is in fact Saturn's son.26

Obinn Mercurius Wulfstan )Elfric One might supposethat in in simply copied equating and Wulfstan followed iElfric's De Falsis Deis in This in is true that this passage. clearly making Wulfstan's however, It that within circle this equation. seemspossible, Deis De Falsis his version of produced before he and

known, but familiar this equationwasnot only even

Obinn Latin be in for prr texts to usedasunexplainedequivalents translating and enough found homily for be The in Mercurius. is Jupiter the evidence this to and mentioning De Adso's largely This homily Napier known consists of a translation of 42. as commonly

25De Falsis Deis in The Homilies of Wulfitan, ed. by Dorothy Bethurum, (Oxford: Clarendon Press, 1957),p. 223.

140

Antichristo, which, echoing psalm 95:5 (see section 4.4, below in reference to Maxims I(B) deos Herculem `omnes Apollinem, Iouem, this gentium, and uidelicet, psalm), mentions Mercurium, quos pagani deos esseestimant' ('all the gods of the [pagan] peoples, that is to 27 Jovis Mercurius, Hercules, Napier Apollo, be whom the pagansthought to say, and gods'). list `ealle, deities Ja hxbene Jt be 42 translates this of classicalpagan men cwxdon, as godas

beon sceoldan on hxbene wisan; swylc swa wxs Erculus se ent and Apollinis, Pe hi marne leton; Owen, ]gor god eac and Pe hx6ene men herjab swipe' ('all those whom heathen men

Hercules be in heathen Apollo, to the the ent and stated gods, manner; such as was whom Obinn, held famous be Prr heathen they to a god; also whom and men praise very

28 greatly'). The translator appears have Graeco-Roman been he interested in to mythology, since Hercules does Adso, but information that or she not merely translate providesthe extra was 29This cent. her Jovis Mercurius his interest is in translationsof an same or and as reflected
Obinn. That he or she could do so without any explanation, and without signalling J brr and Obinn This her interesting. his prr is in that could or original, very were not the names and be becausethese equations were a commonplace of late Anglo-Saxon Latinity, perhaps as a Glossary Corpus in the the teaching result of the use of mythologically-rich glossariessuch as Corpus Glossary). The (see Latin learning early glossaries, section 3.4, above, on the and of

26Anglo-Saxon Prose, Library (London: The Everyman Michael Swanton, by ed. and trans. Vermont: Tuttle, 1975; repr. 1993), pp. 186-87.

Dent;

27Adso Dervensis:De Ortu et TemporeAntichristi, ed. by D. Verhelst, Corpus Christianorum, Continuatio Mediaeualis, 45 (Turnhout: Brepols, 1976),p. 17. My translation. 28Wulfstan: Sammlung der ihm zugeschriebenen Homilien nebstUntersuchungen ber ihre Echtheit, ed. by Arthur Napier, Sammlung englischer Denkmler in kritischen Ausgaben, 4 (Berlin: Weidmann, My translation. 1883),p. '97. 29One might object that calling Hercules a 'giant' shows ignorance or confused recollection of Graeco-Roman mythology, since Hercules was known as a giant-killer. The modem English term `giant', however, is probably not a precise translation of `ent'. The use of the Old English phrase `enta Roman has been describe buildings and similar constructions geweorc', and similar phrases,to Britain, Germanic in and the auxiliaries stationed thought to sit uncomfortably with the evidence of (see As Roman Anglo-Saxon suggestedabove, constructions section z. z. z, above). use of early however, the difficulty is removed if one thinks of the term `ent' as a more precise term than simply

141

with their mythological equations, probably owe their survival to their continuing utility Anglo-Saxon One large least, through a period. can, at part of the reasonably suppose that their use in the Anglo-Saxon curriculum was not limited to the generation of their by learning Latin the majority of those to read production, and that such glossarieswere used level in later Anglo-Saxon England. The importance of such glossariesin the to an advanced Anglo-Saxon curriculum may, moreover, even have been somewhat underestimated, since Latin-Old English glossaries might be of limited utility in the early Anglo-Norman period,

Old have Other English be likely been lost destroyed therefore, to time. and might, or at that texts for Old English Gregory's Cura Pastoralis instance, the translation of such as, -

have fared better, apparently becausepost-Conquest ecclesiasticsin some areasstill saw a use 30 for Anglophone for them, perhaps as materials audiences. preaching to That a few basic equations of heathen and classicaldeities were relatively common in late Anglo-Saxon scholarly discourse is supported by the use of heathen deities as straight Martyrology, punor Old English deities in in for the translations which and classical pagan " be Jove Mars. The for Tiw are used as equivalents to origin of this text appears and helps Wulfstan his to establish the currency of these and circle, and considerably earlier than 32 The in translation, moreover, although equations some quarters prior to the tenth century. The Martyrology, is in probably original. only one of the manuscripts of the occurring

former `Pone C. The has B in the readings occur only manuscripts and relevant passages Tiges deofolgylde'('the emperorDecius compelledSyxtus to Syxtum nedde Decius se casere lxdan deofulgeldum, hi pures Romeburge Necetius, `het Tiw's idol') to gerefa, and worship be Rome, ('Nicetius, hi Pxt het to the taken that they the ordered reeve of weorrian' and

`giant', having the force `mighty man of a former period'. This would also seem to be the meaning here. 30Wendy Collier, `The Tremulous Worcester Hand and Gregory's Pastoral Care', in Rewriting Old English in the Twelfth Century, ed. by Mary Swan and Elaine M. Treharne, Cambridge Studies in Anglo-Saxon England, 30 (Cambridge: Cambridge University Press, zooo), pp. 195-zo8 (pp. zo6-o7). 31See Das altenglische Martyrologium, ed. by Gnter Kotzor, Abhandlungen der bayerischen Akademie der Wissenschaften, Philosophisch-Historische Klasse, n. s. 88, nos i and z, z vols (Mnchen: bayerische Akademie der Wissenschaften, 1981), II, 172,, zoo-os, 47-z,42-3.

32Das altenglische Martyrologium, 1,143.

142
33 idols of prr, and he ordered that they worship them'). Contrastingly, in C Tiges deofolgylde' is replaced by `J t he gelyfde on hys deofolgyld' ('to believe in his idol'), and `pures' is replaced by `hys'.34For the scribe of manuscript B to have replaced something along punor Tiw C lines have him the readings with the names of and of the would required or her to return to the sources of the martyrology. 35It seemsfar more likely that the scribe of C, or of its exemplar, substituted these readings for translations which he or she felt to be According Ker, (Corpus C Christi College, to either meaningless or offensive. manuscript Cambridge, manuscript 196) is written in one or perhaps two Exeter hands, and probably formed part of Leofric's donation to Exeter. 36It seemsunlikely in the extreme that these Tiw hardly punor by distaste, in since the entries present and a alterations were motivated deities light, iElfric Wulfstan heathen positive shows any reluctance to name and neither nor in the process of condemning them; quite the opposite. Did the scribe of the Exeter Probably Martyrology find if incomprehensible? the not, and even manuscript of the names he or she did, he or she would undoubtedly still have recognised them as names (of false 0 left The have them. very act of substituting grammatically and gods) and would therefore contextually appropriate alternatives shows some measure of understanding of these names. Perhaps, then, we should consider a third possibility, that the scribe replaced the names for Tiw punor Roman deities he because that were not and or she recognised scholarly reasons, Roman Were by it have been the this a case, worshipped emperor. and therefore would not Exeter Roman deities, in interest in the tenth century, not at pagan would reflect a scholarly dissimilar from that evidenced elsewhere by iElfric, Wulfstan and his circle.

for The continuing use of Latin-Old English glossaries the ready may account Old English deities in in Graeco-Roman Germanic for translation, as the substitution of Martyrology, or, indeed, in Napier 42. In the caseof Napier 42, however,it doesnot explain

33Das altenglische Martyrologium, II, 172and zos. My translations. 34Das altenglische Martyrologium, II, 172and Zos respectively. My translation. On the spelling of `pures', see Campbell, Old English Grammar, p. i9o (section 474.3). Martyrologium, 11,337and 351" 35On the sources of the Martyrology, seeDas altengische

143

English Scandinavian JElfric's than the the translator's use of the rather names. use of Scandinavianrather than English nameshas beenseen,asdiscussed being due to a above,as 37One might, deliberatedesireto avoid mention of the gods of the heathenAnglo-Saxons. however, object that Danish heathenismwasactuallylikely to be far better known to )Elfric English for his former few distant than the wasthe variety, and audience was at most a hundred miles, the latter a few hundred years.This considerationappliesall the more forcefully to the author of Napier 42, if we acceptHollis's convincing argument that he was 38 Wulfstan. closely connectedwith According to Hollis, Wulfstan clearlyusedNapier 4z asa sourcein the production of Bethurum believes,however,that Wulfstan had himself alreadyusedAdso's his homily 5.39 De Antichristo asan outline for the preparationof his homily ia.40A comparisonof the first Adso does between it is the two, of part is with not show striking similarities although 41 follows found in Adso: apparentthat is an outline not unlike that
Wulfstan: Adso: Omnis qui secundum cristiane professionis rectitudinem auf non uiuit docet quam oportet, auf aliter Christo in quia scilicet, cunctis contrarius erit Anticristus est, quia secundum Christo (p. et contrariafaciet. 22) interpretationem sui nominis apellatur. Anticristus enim contrarius Cristi dicitur.

Multi xtiam tempora Anticristi non in tarnen membris eius uidebunt, sed

Hic itaque Antichristus multos habet sue

(p. 22) malignitatis ministros,

inueniuntur, multi Deinde Surgent legitur: in per uniuersum orbem nuncios mittet enim euangelio sicut
36N. Ker, Catalogue of Manuscripts Containing Anglo-Saxon (Oxford: Clarendon Press, 1957),PP.

75-76. 37See Homilies of yEqic, ed. by Pope, ii, 716. 38Stephanie Hollis, `Napier XLII and Wulfstan's Homily V, Notes and Queries,zoo, n. s. 42. (1995), (pp. 425-26). 423-2,6

39Hollis, pp. 424-25.Numbering following The Homiliesof Wulfitan. 4oHomilies of Wulfitan, pp. z82, -84. 41Pagereferences Wulfstan's Bethurum's De Verhelst's Antichristo edition of and edition of are to homily ia.

144

Predicatio pseudocristi et pseudoprophete, et et praedicatores suos. autem eius et dabunt signa magna,ita ut in errorem potestastenebit a mare usque ad mare, ab fieri mittantur si potest etiam electi. orienteusque ad occidentem, ab aquilone usque
Faciet ad septentrionem. quoquesigna multa, inaudita. Faciet ignem de miracula magna et florere terribiliter celo uenire, arbores subito et arescere,mare turbari et subito tranquillari, diuersis figuris in naturas mutari, aquarum cursus et ordinem conuerti, aera uentis et commotionibus multimodis agitari et cetera innumerabilia et stupenda, mortuos etiam in bominum ita in conspectu suscitari, ut errorem inducanur, sifieri potest,etiam electi. (pp. 2-4z5)

Intelligite ergo, quasidubitando dixit,

fieri Dominus si posest, non quod dubitet, aliquid qui nouit preterita et futura, sed si electi sunt fiere non Si fieri potest. autem potest, electi non sunt. Dicit ergo si fieri potest, quia trepidabunt propter multitudinem signorum;
dicuntur tarnen sed quia non cadunt, electi.

According to Bethurum, 'This is the only first in homily has the the that passage part of (p. Adso' in no parallel 283). Nam quando tanta ac talia signa uiderint etiam illi, qui perfecti et electiDei sunt, dubitabunt, (p. 2,5)

Erit enim tunc tribulatio et angustia fait die qualis non quo gentes esse a Et illud. tempus nisi ceperunt usque ad breuiati fuerint dies illi, non poterit salua esseomnis caro; sed propter dies breuiabuntur illi, ne electos inducantur in errorem. (p. 113)

fait tales tribulatio, qualis non erit superterrain ex tempore,quo gentesesse ceperunt usquead tempusillud. [... ] Tunc breuiabuntur dies Nisi enim Dominus abbreuiasset propter electos. dies, nonfuissetsalua omnis caro. (p. 25)

by Bethurum for this section of is seemto be a The other verbal parallelssuggested

42 from If Wulfstan however, Adso. did follow do deal Adso's than the passages great closer, borrow Adso's despite he have had little inclination borrowing to to wording, outline, seems 43 Vulgate. Isidore, Gregory from his Augustine, and as well as the wording much of

' Homilies of Wulfitan, p. 2,8343Homilies of Wulfitan, pp. z8z-83.

145

It is possible, then, that Wulfstan's use of, and acquaintancewith, Adso's De Anticristo was chiefly through a copy of the translation in Napier 4z (whether that in the in longer Whoever Napier extant manuscript, or a manuscript no extant). produced 42 (in his her first instance, it would appear,Wulfstan and that the perhaps assumed or audience his circle) would be immediately familiar with the equation of heathen Scandinavian deities deities, it be for therefore, and classical pagan and would not, to them that need spelt out Obinn and prr were being used here as equivalents for Mercurius and Jovis. Another is Napier Mercurius Jovis that the possibility author of 42 simply chose to replace and with deities who might be more immediately relevant to an audience in close proximity to Danish heathenism in the Danelaw. The manuscript layout of this passage in Napier 42, moreover, indicate deities important to that these seems names of non-christian were particularly at

least to the scribe of the extant manuscript,who punctuatedthe passages highlight to so as

Obinn: `Swylc swa waes hi Pe the names of prr and erculus se ent. 7 apollinis mxrne god leton: kor eac 7 ow&en. Pe hxbene men heriab swi&e.'44.While a simple point suffices to break up the clausesin such a way as to aid reading, the scribe exceptionally introduces prr and Obinn highlighting figures. thus these particular with a two-stroke point,

iElfric's De FalsisDeis, but it almost There is no suggestionthat Napier 4Z predates

Wulfstan Evidently followed RElfric but Wulfstan's closely, version thereof. certainly predates by A lfric known in Napier the equations given were well the evidence of 42 suggeststhat Wulfstan's more northerly circle soon after the publication of the Lives of Saints. Given the familiarity by in Scandinavian texts, the and suggested gods this group of uniform use of the Napier 42-,it may well be that 1 lfric was following the lead of Wulfstan's circle in making his equations. The suggestion is supported by Rlfric's own use of this equation in his life of St Martin in Lives of Saints, where he repeats the information which he gives in De Falsis Deis:

Mid Pusendsearo-crxftum wolde se swicoladeofol

' Anglo-Saxon Manuscripts in Microfiche Facsimile, viii: Wultan Texts and Other Homiletic Materials, descriptions by Jonathan Wilcox, Medieval and RenaissanceTexts and Studies, 1i9 (Tempe: Arizona Center for Medieval and RenaissanceStudies, zooo), MS i, slide i, p. 33"

146

halgan beswican one wer on sumewisan . hine and ge-sewen-licneon manegumscin-hiwum Pam halgan eteowde on J ra hxpenra goda hiwe . . hwilon on ioues hiwe Peis ge-haten J6r. . . hwilon on mercuries Je men hatab opon . . hwilon on ueneris Paere fulan gyden . Pemen hatab fricg. and on manegumojrum hiwum45 With a thousandwily arts did the treacherous devil in strive someway to deceivethe holy man,
he himself divers in and showed visible phantasms to the saint, in the appearanceof the gods of the heathen; Jove's in form, who is called Thor, sometimes Mercury's in Odin, is sometimes who called Venus, foul in that of the sometimes goddess, Fricg; into whom men call and many other shapes. That iElfric felt it necessaryto repeat this information suggeststhat he thought that these equations were not widely known. Another possibility is that. Elfric was particularly interested in exhibiting his learning by equating classicaland Scandinavian deities; this would Jovis, for Venus, Minerva his Mercurius but translation of the names account and not have existed whose name no vernacular equivalent seemsto from his source, Sulpicius for

46 Vita The author of Napier 42,,by contrast, seemsto assumethat Severus's Sancti Martini. It familiar likely, be his least immediate these with equations. seems extremely circle would at Scandinavian from deities his lElfric then, that equations of classicalwith northern received England, but this cannot be definitely proved.

Society, O. Text English Early Skeat, W. Walter Saints, by Lives S. 76,8z, 94 and 114, ed. of c's (p. Skeat's (no. lines Trbner, (London: translation 710-17). z65). 31, ii, 19oo), z64 i88i, z vols 46Ae Skeat, by Saints, Lives 11,457-ed. of cs 45Ae

147

Deities Post Danish After iElfric: in iElfrician Literary Culture 4.2.2

O The interest in sinn displayedby iElfric, and by Wulfstan and his circle, did not .
disappear A brief the their simply publication of work. with consideration of the dissemination and treatment of iElfric's De Falsis Deis subsequent to the publication of the Lives of Saints reveals that this tradition continued to attract considerable interest. Pope identifies seven manuscripts as containing material from Mfric's De Falsis Deis, in addition to

47 Wulfstan's the manuscript containing reworking of the text: Pope/Clemoes Si lum

R

Ker Number
41A

Manuscript
Corpus Christi College, Cambridge,
manuscript 1'72

C
L
W

57
18
162

Corpus Christi College, Cambridge,

manuscript 303

University Library, Cambridge, manuscript

Ii-I-33

British Library manuscriptCotton Julius

E. vii

333

Bodleian Library manuscript Hatton 116

G
Xk
T(I14 & 113)

209
66, art. a
331

British Library manuscript Cotton Vespasian D. xiv

Bibliotheque Nationale, Paris, manuscript

Lat. 7585 Bodleian Library manuscriptsHatton 114

and 113 (Wulfstan)

Lives Pope'smanuscript W appears the closestextant version of the of to preserve Saints to that in which LElfric issuedthem, with an appendixcontaining, among other things, 48 W, but in found in The Deis. De Falsis unordered, occur manuscript samepiecesas those Saints like Lives intended, L. These the of primarily probably were manuscripts manuscript 1, Homilies Catholic from lives S itself, for pious reading. Manuscript consistsmainly of saints'

48 Clemoes,

" SeeHomilies of 1EWc, Pope, by 667-68. ii, ed.

p. zw.

148

49 for have been Manuscript G is little interest in and may similar compiled a purpose. of iElfric's description it the transmission examining of of the paganclassical gods,since Daniel from De Falsis Deis, is for homily contains only the story of which usedasan ending a from Catholic Homilies IL" According to Pope, this manuscript constitutes`amiscellanyof theological pieces',and, as such, it may well alsobe intended for devotionalreading, rather 51 than public preaching. The intended use of De FalsisDeis in manuscriptsC and R, however,is lessclear. In
C drawn largely from Lives Saints, there two manuscript are the other, sections, one of liturgical Catholic Homilies from 11.52 I A arranged according to the year, and similar following liturgical is be arrangement of two sections, one miscellaneous, one the year, to found in manuscript R, in which an Old English colophon specifically states that the homilies in the first section may be delivered at any time, but those in the second section 53 These for De Falsis days. tendency the two then, only on specified manuscripts, suggest a Deis to be used as a sermon, or as material towards sermons.

From the point of view of determining how. Elfric impactedon Anglo-Saxon

knowledge of, and interest in, pagan and heathen myth, manuscript Xk is by far the most interesting. This does not even approximate a full text of De Falsis Deis, and is not a straight Isidore's Etymologiae, The is written partly manuscript a copy of quotation of part of the text. in a continental Caroline minuscule of the end of the ninth or beginning of the tenth 54 On Caroline English in the tenth the end of century. minuscule of an century, and partly has hand English leaf blank last the of eleventh century added a the of this manuscript an

49Homilies Pope, by f' 68. t, c, ed. of Homilies of zE c, ed. by Pope, 1, i, s. The Exploitation Source: Swan, VElfric Mary sl Homilies See Pope, by as of t, 2,4. also c, ed. of iE (unpublished doctoral Centuries' Twelfth Late Tenth from JElfric's Catholic Homilies thesis, to the University of Leeds, 1993),pp. 144-47" 52Homilies of Pope, by 1,18-19. c, ed. if s3Homilies ofi Pope, by 1,63. c, ed. 54Homilies ofi'lfric, ed. by Pope, [, 88.

149

brief description of the paganclassical Mfric in is heavily dependent gods,which upon that of ss De FalsisDeis. That JElfric should havebeenusedin this way seemsto suggestthat the equations
he drew which were not common currency in all quarters in late Anglo-Saxon England, for the author of this note is specifically interested in pulling out these equations from Tlfric's work, suggesting that they were in some way noteworthy and unusual. At the same time, the note seems to engagewith the section `De Diis Gentium' in the Etymologiae, and indicates a 56 interest in Elfric's Isidore heathen deities. deities particular equation of classicalpagan with 57 important knowledge England, but in Anglo-Saxon was an source of of classicalmythology here the interest is clearly focussed not simply on classicaldeities Apollo elsewhere in De Falsis Deis for JElfric mentions

but specifically on their Scandinavian equivalents. As

Exeter Old Martyrology, discussed English the with manuscript of the above, which specifically engageswith and rejects here heathen its the classicism of source, so we see

for heathen individual evidence classicism of another eleventh-century engaging with the their sources, in this casethe work of iElfric and Wulfstan's circle. Coming at this late date, Danish heathenism Danelaw long dissipated, with the we might see this as a since of the begin develop heathenism in the to the which precursor of more plainly antiquarian uses of legendary histories of the twelfth century onwards (seesection 5.1,below).

ssHomilies of zE

Pope, by 88. t, ed. c,

56Isidori Hispalensis EtymologiarumLibri, ed. by Lindsay, i, book 8, chapterii. 57SeeHerren, `The Transmission and Reception of Graeco-RomanMythology', pp. 9o-9i.

150

Da VIIII Woden Magico-Medical 4.3 3enamwoden wuldortanas: as Figure

The Second Merseburg Charm 4.3.1

The Second MerseburgCharm German manuscript of provenance -

High Old German it does, in in occurring, as and a

doesnot obviouslybelong in a discussionof the usesof

Woden and Oiinn in Anglo-Saxon England. As an instance of a tenth-century literary Woden, involving however, Old English Nine Herbs Charm, it charm clearly relates to the discussed is in below. Given the relative absenceof evidence of traditions which section 4.3.2,, involving Wodan on the Continent in the tenth century, it seemsuseful to discuss this piece here. The charm is a historiola, comprising a narrative section involving Wodan and a figures, formula: followed by number of other a short magical Phol ende Uuodan holza. vuorun zi

du uuart demo Balderesvolon birenkit sin vuoz biguol Sinthgunt, Sunna thu era suister; en Friia, Volla biguol thu en era suister; Uuodan, he biguol thu so uuola conda: en benrenki, bluotrenki, sose sose lidirenki: sose ben zi bena, bluot zi bluoda, 58 lid zi geliden, sosegelimida sin! MS. Cathedral Library Merseburg, This charm appears folio 85` 136,a ninthof on have been folio found On is to the same another charm, which appears century manuscript. 59 Bostock in into the manuscript together with this charm the tenth century. copied identifies the dialect of the charms as`Middle German, possiblyFulda', on the basisof the

58 for Lesebuch, by Braune, See below Althochdeutsches 86. translation. ed. p.

59J. Knight Bostock, A Handbook on Old High German Literature (Oxford: Clarendon Press, 1955),p. 16.

151

60 forms showing unshifted /d/ and /g/ but /t/ shifted to /z/. The <uo> spellings suggest the influence of Frankish orthography, which is to be expected in texts of this date in Middle German dialects.61To say that the dialect possibly originates from Fulda is, however, dialect from large best; is quite a region, originates misleading at all we can safely say that the The localised Fulda be dialect cannot of which was a major centre of manuscript production. Fulda, higher in is to such although there a chance of any given manuscript containing texts dialect having been in Fulda than anywhere else the region; we should not produced at a forget that here we are dealing with a probability calculated not on the basis of the dialect, but on the basis of the fact that it has been recorded in a manuscript. Bischoff claims, Fuldaer On Typ `nicht dem bekannten entspricht'62. moreover, that the script of the charm bulk hand, Bischoff identifies the other the ninth-century script of the of this manuscript as `eine prezise Sptform, die wir nur aus Fulda kennen', which does present us with the late have in how the course of the travelled the manuscript might question of whither and 63 Fulda. if did it not remain at or near ninth and early tenth centuries,

The charms are, asmentioned above,a later addition to the manuscript, which was includes in a copy of the the ninth century; the ninth-century section originally written 6' 64 Frankish Abrenuntiatio Diaboli, and is, accordingto Bostock, theological in character. The in hardly heathen deities is list what one might expect a charm use of such a of namesof have least but Christian, by down in we one other at a which was all probability written instance of the use of clearly heathen charm elementsby a Christian, in the Nine Herbs Charm. The latter is clearly intended asa performancetext; its placein the Lacnungacan

b0Bostock, p. 16. 61See Denkmler Die Murdoch, O. J. Groseclose Brian Sidney poetischen althochdeutschen and also (Stuttgart: Metzler, 1976), p. So, who identify the dialect of the charms more specifically as `East Frankish'. 62Bernhard Bischoff, `Palographische FragendeutscherDenkmler der Karolingerzeit', in Aufitze zur Schriftkunde und Literaturgeschichte,3 vols (Stuttgart: Mittelalterliche Studien: Ausgewhlte (86). (1981), Studien, Frhmittelalterliche in 73-III Originally Hiersemann, 1966-i98i), iii published pp.
5 (1971), 101-34-

63Bischoff, p. 86.

" Edited in Die kleineren Steinmeyer, Sprachdenkmler, by p. z3. ed. althochdeutschen

152

leavelittle doubt that it wasintended to be recited (by a Christian) aspart of his or her (see The down below). intention the section 4.3.2, medical practice of the personwho wrote Second MerseburgCharm is lessclear. It is at leastconceivable that the charm may havebeen for but in its antiquarian or admonitory value, either caseone might expectthat preserved been While have it. be included some commentary would with such possibilities cannot discounted, likely down it then, seems than not that the charm waswritten wholly so more that it could be rememberedand recited. The exactmeaning of the charm is difficult to determine. The basictranslation
few difficulties: presents

Phol and Woden went to the forest.

Then for Balder's (or `the lord's') horse its foot became sprained (ie. `Then foot Balder's horse became the of sprained').

Then Sinthgunt sang [a charm], Sunnaher sister, [a Frija Volla her then sang charm], sister, do. how Woden he knew j[a then to sang charm], as well Thus for bone-sprain,thus for blood-sprain, thus for limb-sprain: bone to bone, blood to blood,
limb to limb, thus may they be glued together. There is, however, considerable room for doubt as to the number of deities mentioned and, indeed, which deities are referred to in the charm. The name Phol is extremely mysterious interpretations, It has ranging attracted a considerable variety of and occurs nowhere else. from the identification of the name as a version of the christian Saint Paul, the classical Apollo, a form of Baldr, the masculine counterpart of Volla, to explanations which do not 66 indeed, None it there are of these seemsentirely satisfactory, and, regard as a name at all. few basic facts upon which to base any interpretation. `Phol' ought presumably to alliterate is it has been `vuorun', it a thought, would rule out the possibility that which, with often form of Paul or Apollo. This is based, however, on the early development of Latin or Greek /p/ when borrowed into the Germanic languages, and does not take account of the High German Sound Shift, which would eventually produce /pf/ in place of such an initial stop.

65 Bostock,

p. 99.

153

This affricate is frequently written <ph> in Old High German texts, and may havebeen, at least at some stagesin its development,capableof alliterating with `vuorun'. The fact that the `Phol' in line, after the text waswritten, abovethe waswritten might suggestthat the <h> of `Pol' `vuorun', it. that therefore and altered corrector was concerned should alliterate with This could havebeen done without any regardfor meaning, possiblyby a corrector who did `Pol'. It however, is, not understandthe term alsopossiblethat the corrector was correcting a he digraph here if term the or she understood perfectly well; so, the use of provides <ph> for Latin Greek borrowings interpretations involving evidence rather than against or beginning with /p/. One must also assume that such borrowings were early,otherwise the likelihood must be that the corrector was correcting for meter more than for meaning.
The second line also presents a problem, since it is not clear if Baldr is referred to by 67 Old `bealdor'. If English `balderes', or if this is simply the word meaning `lord', cognate with High Old German in latter be the the only example of this word an were the case,this would dialect, but the same would be true of Baldr. In the absenceof any other evidence, the proper `lord' It highly is likely is would than the noun. unlikely that a noun meaning name more

find High German, is but it Old in to not surprising a only appearonce the entire corpusof here Sinthgunt, The and nowhere after all, appears name proper name appearingonly once. else. The third difficulty with the interpretation of the charm lies in determining whether Sinthgunt, Sunna, Friia and Uolla are all to be understoodas chanting charms,or only Sinthgunt and Friia. The charm text (the order of which has been followed in the translation `Sunna half-lines does illustrate in the era whether make clear this not to point) above, order Friia, Sinthgunt `Volla or stand and qualifying suister' and era suister' are adjectivalphrases led This has deities. list in to also the charm-working of entries alone as separate

66SeeBostock, pp. 23-26.

67On this problem, seeD. H. Green, The Carolingian Lord: Semantic Studieson Four Old High German Words (Cambridge: Cambridge University Press, 1965), pp. 14-17.

154

four different goddesses, or two aspects each considerablespeculationas to whether theseare 68 her Friia hypostases. with three of of two, or even In this charm, then, we seeWodan in a situation which we haveno real reasonto
having least formed is as regard as part of a pre-christian mythological narrative; the story at likely to have been created to fulfil the needs of the charm. Both this and the Nine Herbs Charm involve a narrative of which Woden/Wodan seemsto be the principal character, but in neither caseis there any reason to supposethat the narrative is not specifically intended for It based is these the charm. upon mythological narrative, possible that charm narratives were but it seems equally plausible to suggest that narratives were created specifically for magical Wodan his In terms and of purposes which stand outside the ordinary sphere of myth. Wodan; for both the these charms seem to reflect magico-medical attributes attributes, deities but be tailored to the ailment, the choice of narrative may Wodan and particularly of

(whether literary be by the attributes or mythological) of must conditioned

Wodan. There can be little doubt that a conception of Wodan as a healing god is at work behind these charms. The dating of this attribute is, however, more difficult. Both these charms involving is date, late Wodan down and there no way of at quite a were written a narrative about determining when they achieved their current form. Concerning the magic formula of the SecondMerseburg Charm, Bostock points out that `there are parallels in the Indian languages is least doubt is extremely ancient'69 possible that this part at which are so close that no . This, however, is not evidence for the dating of the narrative section of the charm, which (much One later have been earlier might note that the other addition. a could easily O6inn, (on Ribe from inscribed that on the skull-fragment recorded) charm mentioning deal in is is below), character simpler great a and charm, a narrative not 5.6, which see section 70 This in Nielsen, it is (although, the that suggest would verse) to according and structure

69 Bostock,

68For a useful summary see Bostock, p. zz.

p. zo.

70Niels Age Nielsen, Danske Runeindskrifter: Et UdvalgmedKommentarer (Copenhagen: Hernov, 1983),p. S6.

155

formulae later additions to narrative sections of these manuscript charms are probably which likely it least have been in is that the apparent at as may very early composition, although formulae function but is a of their simplicity and universality of such not of their antiquity common aims. It would appear, then, that this charm's use of Wodan reflects an association of Wodan with healing which is not evidenced before the tenth century; it is possible that the Wodan but is for is the this, and we should perhaps place there no evidence association older, healing deity within the framework of a later, literary tradition within which he forms part of healing Northwestern European Mercurius figures, attributes among whom the a complex of fits Hermes become In well quite generally spread. this respect, as we shall see, the charm of late in is Nine Herbs Charm, Woden in the the which also evidenced with the presentation of tenth or early eleventh century.

Also related to the problem of literary or traditional origins for this charm narrativeis
Norse literature Old for Baldr Baldr. The lack outside the possible presence of of evidence from heathenism, Scandinavian late in Baldr development tends to suggest that was a special development in literary around the ninth to tenth centuries onwards, or even a purely Wodan himself, for especially on christian contexts, although given the slender evidence even Baldr. It is, however, lack is be hardly Continent, of certain that this evidence of a the we can Baldr have borrowed A in interesting to note as a this connection that thelweard appearsto Bldg figure Anglo-Saxon for the unrelated replacement (see section 4.2, above). This

for further he in Baldr is if evidence this provides at all, charm that, to would seem suggest the late, literary character of the text.

for difficult, is in be The idea that this charm may essentially character clearly written
Christian, but by down it imply actually a written that not merely was this would tend to difficult While the Christian. by given to especially accept, this may seem a compiled be Germanic better-known too not we should perhaps churchmen, the strictures of many of know those the all of attitudes reflected well ready to assume that those whose writings we This is not were who men there educated to that were to say not write. who were able

156

Christians in Germany and England around the tenth and eleventh centuries,but rather that, in a society where medicine and magic formed a continuum, the potential sin of compiling or involving heathen deities have been disregarded by some on the grounds using a charm might 71 its It of perceivedefficacy. seemsprobable,then, that this charm really reflects the developmentof Wodan asa literary, rather than a religious, figure, although this developmentmay also involve an understandingof his name aspossessing somesort of magical efficacyor connotations.

The Nine Herbs Charm 4.3.2

The SecondMerseburg Charm is interestingly paralleled by an Old English charm, Nine Herbs Charm. The between to the generally referred the two may as similarities visible however, formulae the well, reflect universality of charm rather than a more specific between The Nine Charm Herbs is Anglo-Saxon the two texts. connection an unusual it is Old English in the charm, since one of relatively small number of such charms which are The long, is involves verse. charm also unusually and a number of narrative sections, as well (editorial) directed herbs in the title; we might characterise as commands referred to at the One Woden historiola. describes killing this charm as a complex of the narrative sections a behave from factors harmful disease in This not to a snake and extracting promises manner. is then immediately followed by a description of `witi3 drihten' creating helpful herbs:

Liuzza, Michael `Anglo-Saxon Roy Anglo-Saxon homiletic in manuscripts;see materials some with

" This is not, however, to dismiss all medicine among the Germanic peoples at this time as entirely based on superstition. This view, which used to be the normal view of Anglo-Saxon medicine, for instance, has rightly been discredited in recent years. The work of Cameron, in particular, has done M. L. Cameron, found in dispel the over-emphasis on superstition earlier scholarship; see much to `Anglo-Saxon Medicine and Magic', Anglo-Saxon England, 17(1998), 191-2I5,and also M. L. Cameron, Anglo-Saxon Medicine, Cambridge Studies in Anglo-Saxon England, 7 (Cambridge: Cambridge University Press, 1993).In this connection one might also note the juxtaposition of disapproval by Church) (which in the are also magical character and viewed with prognostic materials

157

/ he toslat +wyrm com snican, nan. VIIII wuldortanas, ba 3enamwoden pa ba kt heo VIIII sloh naeddran tofleah. on Jaer3eaendode xppel ] attor, pat heo nxfre ne wolde hus bu3an. on finule, fela +fille 7 mihti3u twa, pa wyrte 3esceop drihten, / witi3 hali3 on heofonum pa he hon3ode, VII / j sette wende on worulde, '2 bote. earmum 7 eadi3um, eallum to
A snake came crawling, he tore no one. Then Woden took up nine glorytwigs, and struck the snake so that it flew into nine pieces. There apple and poison swore that it would never return into the house. Chervil and fennel, the greatly-powerful pair; the wise lord, holy in the heavens,made those herbs he hanging, he into when was them them the seven established and sent help both blessed. to worlds, the wretched and the everyone, This passageseems to contrast the potentially harmful `xppel j attor' and the beneficial `fille finule'; it be Woden between 7 can also seen as creating an opposition harmful agents or receivestheir promises not to act harmfully destroys who

`witig drihten' and the who

The helpful. `witig drihten' be creates agents which are actively phrase could used to refer to Woden, but given the structure of opposition, and the tradition of opposing Woden the false God (expressed I(B), Maxims below), in it the true see section 4.4, creator seems creator and `witig fact, drihten' is, in the christian god. probable that the

This Old English charm occursin British Library manuscript Harley 585,in a According Grattan Lacnunga. known to and asthe collection of medical recipesand charms ' Singer, the manuscript datesto about the year moo.? There can be little doubt that the from Christian, by it is the abovepassage that the and alsoclear manuscript was written a Lacnunga. before became heathen it in if it part of the was origin, was christianised charm,

Prognostics in Context: A Survey and Handlist of Manuscripts', Anglo-Saxon England, 30 (zoos), i8i2-30.

'2 J. H. G. Grattan and CharlesSinger,Anglo-Saxon from the Magic and Medicine: IllustratedSpecially My Semi-PaganText Zacnunga' (London: Oxford University Press,1957-), translation. PP-I5z-5473Grattan and Singer, pp. zo8-og.

158

MersehurgCharm, however,that it is not necessarily It has been seenwith the Second involving heathen deities be that the namesof reasonableto assume manuscript charms must heathen have, been down. It is survivalsof old, charmswhich unusually, written not
impossible that Christians engaged more closely with these charms than simply in copying The in Anglo-Saxon them. this presence of charm an medical manuscript of around the year it least It Anglo-Saxon iooo makes possible that at one christian actually used the charm. demonstrate It felt be is, that may also to such charms were effective. moreover, clear that this charm as it appearsin the manuscript was not simply copied down by a Christian, but (if in degree by Christian, for includes least it actually some not wholly) compiled at a one Christ: `crist [e]' ('Christ form to reference stod ofer adle aen3ancund stood above any of 74 sickness').

The possibility remainsthat the narrativeitself is an old, heathen element in the

The however, leads into the charm. oath-swearing which completes the narrative, straight (presumably lord fennel, the passageon creating chervil and and the seven worlds, christian) 75 Grattan it Singer found in the which sevenspheres classicalcosmology. and connect with far likely, light in the of this transition, that the narrative passage of this charm seems more has at least undergone some christian recomposition, and indeed it is perfectly probable that it is originally a christian composition. In fact, the use of Woden in this charm may not be here Woden in dissimilar in that the genealogical materials, too to the use of entirely

Woden's power rests not on his status asa heathendeity, but on his positioning as an AngloThe link here is quite specific, Saxon link to antiquity, and, in this case,classical traditions. Woden but himself, his depends on equation not on the reinterpretation of moreover, and England having late Anglo-Saxon in Mercurius Mercurius. as was clearly recognised with English Life Hippocrates Old in Galen been a medical deity; he appears the of and alongside Saint Pantaleon,aswell asfeaturing twice in the Old English version of the Herbarium of

74Grattan and Singer, p. i56. 75Grattan and Singer, p. iss (note z).

159

Pseudo-Apuleius, in one instanceasthe provider of a 76 herb Ulysses. In to the protective Nine Herbs Charm, therefore, it seemslikely that Woden is functioning for asan equivalent Mercurius as a figure of magico-medicalpower and resonance, and particularly asan 77 for Mercurius healing herbs. The appealis, of equivalent asthe provider of protective or course,to the power of the christian god to heal, but a referenceto a healerof authority (albeit only human authority) is still a useful and valid part of the formula. In somesense, then, this would not be so far from the usemadeof Woden in the Anglo-Saxon genealogical in both materials, that they and the Nine Herbs Charm use his name,rather than the deity himself, asa political or magical figurehead.The use of Woden here, however,also rests literary linking him Mercurius/Hermes healing deity his tradition upon a with asa and Mediterranean reflexesand associated figures. This charm is, then, most important asevidence literary development the of of
Woden, but may also bear some relevanceto the peculiarly Anglo-Saxon use of Woden as a literary figure of power, whether political or medical. This may account, at least in part, for the apparent willingness of a christian Anglo-Saxon to use Woden's name as part of his or her medical practice. The euhemeristic view of Woden which takes hold in England allows him to enter the narrative of such a charm merely in the capacity of a distinguished man of looking This at this charm; the old. seems, on the whole, the most reasonableway of

Hdvaml lots Tacitus it described by hardly attempts to connect with or with the castingof 78That this be disputed, but is it in to text not charm wasa practical should seemconvincing.

76This passage in the Life of Saint Pantaleon is available through the Dictionary of Old English Corpus //ets. umdl. umich. edu/cgi/o/oec/oec-idx? type=bigger&byte=79944i o&qi=gallige&qz=&q3=> <http: [accessedz December zooz]; The Old English `Herbarium' and `Medicina de Quadrupedibus',ed. by Hubert Jan de Vriend, Early English Text Society, o.s. z86 (Oxford: Oxford University Press, 1984), PP. 94-95 and 114-15. " We should note that Woden and Mercurius Saint Veronica, fulfilling in such a role; are not alone for one, also seemsto be used in this way (seeMary Swan, `Remembering Veronica in Anglo-Saxon Old Middle English England', in Writing Gender and Genre in Medieval Literature: Approaches to and Texts, ed. by Elaine Treharne (Cambridge: Brewer, zooz), pp. 919-39 (especiallypp. 34-36).

78On the former, seeNorth, p. 87; on the latter, seeGrattan and Singer p. 54.

160

be literary, all probability understood as essentiallypart of a christian tradition relating to Woden, and most probably relating to him asa Mercurius figure.

Woden Woden False Creator 4.4 worbte weos: as

Modern critics have not always been kind to Maxims I, a poem, or group of poems, is Exeter in Book. In the which extant the introduction to their edition of that manuscript, Krapp and Dobbie characterised Maxims I as follows: `The entire text gives the impression of from by a mass of unrelated materials gathered a number of sources, and assembled the less compiler more or mechanically, with no attempt at selection or logical arrangement'79 . This view has, of course, inspired many to try and find order in the poems.8 While there is a internal however, to these their orderliness certain order poems, servesit the and ends which

be fully if can only understood one considers not only the structuring of the poems

themselves, but also their positioning within Anglo-Saxon literary traditions.

At this point, it will be useful to outline what can be determined about the
history In his Muir composition and of reception of this text. new edition, places the writing 81 Book Exeter Crediton Of Exeter. least the of around 965-975, probably at at or equal interest must be the provenance and dating of the composition of Maxims I itself. Krapp and Dobbie find no distinctively Anglian features in the poem, and so accept the view that the from West-Saxon the eighth century poem was probably of origin, composed at any time 82 Old English The localising dating through to the writing of the manuscript. verse and of be fraught dialect linguistic is basis to with of and other characteristics now recognised on the

79The Exeter Book, ed. by GeorgePhilip Krapp and Elliott van Kirk Dobbie, The Anglo-Saxon Poetic Records,3 (London: Routledge; New York: Columbia University Press,1936),pp. xlvi-xlvii. 80For a useful summaryof theseefforts, and yet another, seethe chapteron The Old English Maxims', in Paul Cavill, Maxims in Old EnglishPoetry(Cambridge:Brewer, 1999),pp. 156-83. 81The ExeterAnthologyof Old EnglishPoetry,1,1-38' The Exeter Book, p. xlvii.

161

difficulties, and we would perhapsbe wise not to rely too heavily on the text being of WestSaxon origin, just aswe cannot establisha reliable date for the composition of the poem.83 Maxims I is, as stated above,sometimestreated asone poem, and Krapp and Dobbie
`it divisions is impossible to tell with any certainty whether the three sectional state that indicated in the manuscript were intended by the scribe to be taken as three parts of a single 84. do Muir, divisions however, believes three that the sectional poem, or as separatepoems: 85 indicate three separate poems. On the basis of the manuscript division indicators as he describes them, in these and the immediately surrounding poems, one must agree with this 86The I(B). I, Maxims Maxims but here is, then, not poem which chiefly concerns us view. The A and C poems lie outside the scope of this thesis, but it will be valuable to point out B here. the coherence of the poem

Firstly, one should considerthe manuscript context of this poem. It comesin a

Maxims I(C), Maxims I(B), I(A), Fates Mortals, Maxims The sequenceof wisdom poems: of The Order of the World and The Riming Poem. That The Fates of Mortals and The Order of describe however, One World have is equally well the might, unifying themes clear enough. Maxims dealing lines I(B) first five the with the order of the world: as of Forst sceal freosan, fyr wudu meltan, brycgian, is eorpe growan,

lucan helm wegan, wundrum water inbindan An sceal eorpancipas. forstes fetre felameahtiggod87 Frost must needsfreeze,fire crumble wood, earth burgeon, ice build bridges lock away earth's miraculously and the canopy a support and water frost's God One of powersmanifold, shall unbind the alone, the seeds. fetters."
I

83See the summary of dating methods and their difficulties in Amos, pp. 167-70. 84The Exeter Book, p. xlvi. 85The Exeter Anthology of Old English Poetry, II, Sao.

J. S. A. by Translation, Prose Old Poems English 88 in An Anthology Poetry: Anglo-Saxon trans. of Bradley, Everyman Classics (London: Dent, 1982),p. 348.

86The Exeter Anthologyof Old EnglishPoetry,notes to 1,247-64. g' The Exeter Anthologyof Old EnglishPoetry,i, z54 (lines i-S).

162

describe winter, a time when snow and ice The first three and a half lines of this passage fires In build line keep to warm. S, the repetition of the word cover the earth, and people `frost' points up the fact that God's power to dispel winter affects all the symptoms of winter So far, four has built in lines. brief the the previous mentioned picture of the poem up a (and leads its world natural particularly of winter) which audience up to the creator of this The `An' line four God's in natural order. alliterating unique power over the emphasises have before, be the culmination of the semantic and gone and seemsto phenomena which God In begins lines. with as creator syntactical progress of the preceding essence,the poem and controller of the world.

From here, the poem follows the suggestionof God's power to dispel winter, and from largeHaving the the poem moves through the seasons, moveson to summer. passed God's human society. scaleorder of the natural world to the smaller scaleof the ordered This does less human in in it than nature. section starts with the no society order appears king, and moves down through noblesto the variouslower echelonsof society,the Frisian, from finally, infirm to thosewho are outcasts society,the the merchant, and, past the from human This leads for into the need conduct, the good conduct, and, murderers. The humans is from God's manufacturesof completed. artifice to the artifice of progress humans and their proper usersare enumerated,and the poem returns to the ultimate artificer its finally, has the poem recapitulates opening, explicitly referring set out; of all that the poem God God, the creator: to and specificallyto Scyld scealcempan, sceaftreafere, leornere, bec bryde beag, sceal husi halgum men, hxpnum synne. Woden worhte weos; wuldor alwalda is zt rice god, rume roderas-J sawlanergend, sylf so&yning, forgeaf Jaet lifgaj, we on se us eal Pam ende eallum wealdeb and eft xt $9 Jt is meotud sylfa. monna cynne.

89The ExeterAnthologyof Old EnglishPoetry,i, z56 (lines 59-67).

163

A shield necessarily goeswith a soldier, an arrow with a poacher;a ring bride, books holy necessarily goeswith a with a student, the eucharistwith a heathen Woden fashioned Ruler idols; man, and with a sins. the of all fashionedheavenand the spaciousskies.He is the mighty God, the very King Saviour live of truth, the that of souls,who gaveus all we on and who at the 9o dispose He Lord. himself is end will again the ordaining over all mankind. How might an Anglo-Saxon audiencehaveunderstoodsuch a poem?The answer,of
depends Anglo-Saxon A literate is course, upon exactly which audience one talking about. Anglo-Saxon audience is the easiestsort of audience for us to approach, for we simply have idea ideas England. form Anglo-Saxon in in what no and narratives were available oral only Given that we know that Maxims I(B) was available in an ecclesiastical literate and therefore

late in into it the tenth and early eleventh century, when was copied the context

Exeter Book, and when the Exeter Book was in use, it is not unjustified to consider how the have There is been in poem may understood such a chronological and social context. an interesting parallel for this poem in Beowulf, a poem which is also extant in a manuscript of around the year iooo: 91

Sxgde se Pe cupe frumsceaft fira feorran reccan,

[te], pat iElmihtiga se eorhanworh cw c bebugeb, wlitebeorhtne wang, swawater gesettesigehrepig sunnan and monan leoman to leohte landbuendum, foldan sceatas and gefr ctwade leomum and leafum; lif eacgesceop

90Anglo-Saxon Poetry, trans. by Bradley, p. 349. 91For the date of the writing of the Beowulf-manuscript, seeKer, Catalogue of Manuscripts, p. 281(no. indeed, Kiernan's dating for the composition of the the manuscript - and, m6); see also arguments (New Manuscript `Beowulf `Beowulf Kevin S. Kiernan, in and the poem - to the eleventh century Brunswick: Rutgers University Press, 198!; repr. 1984),and in Kevin S. Kiernan, `The EleventhCentury Origin of Beowulf and the Beowulf Manuscript', in The Dating of `Beowulf, ed. by Chase, pp. by due Kiernan's folio does find Newton to revision not 179`as a palimpsest understanding of 9-22. Kingdom East (Sam Pre-Viking Newton, The Origins `Beowulf be B of convincing and the of scribe to Anglia (Cambridge: Brewer, 1993),pp. 8-9).

164

cynna gehwylcum

Parabe cwice hwyrfap. 92

He who was skilled in recounting the creation of men in time distant declared Almighty look that the the made earth, a plain radiant to upon which water he, delight his in taking encircles; the sun and the achievement,established luminaries, light for living he in those moon, those the as world; embellished branches leaves; life he in the earth's surfaces with too and with created each 93 of those species which go their vital ways. This passage figure providesus with an encapsulated poem which presentsa of how God human how he for beings. authority recounting createdearth and sky, and provided The figure of authority is lacking in Maxims Z(B), but the key themes of the createdworld (particularly represented by earth and sky), and of God's power over, and provision for, human beings, are the same. A passage is Beowulf Old in in which strikingly reminiscent of this section occurs the English translation of Bede'sHistoria Ecclesiastica GentisAnglorum.Here Bede recounts the Cxdmon, from him, in instructs the story of who receivesthe gift of versifying an angelwho Old English translation, `Sing me frumsceaft'94. The word `frumsceaft'for `creation'is usedto Katherine O'Brien O'Keeffe As Beowulf introduce the encapsulated `in in notes, poem aswell. Bede, Hymn Bede's Old English Caedmon's in is the part of the main text; manuscriptsof Hymn'95 Latin text it is found (if at all) asa marginal glossto Bede'sLatin paraphrase of the . in Latin the poem with which Although Bede himself, then, seemsto haveparaphrased Cxdmon respondedto the angelic command,the Old English translation recordsCaedmon's is Old English its in text, and this poem now normally versewithin song about the creation Hymn: West CComon's Saxon version of referred to asthe Nu sculon herigean heofonricesweard,

92Beowulf and the Fight at Finnsburg, ed. by Fr. Klaeber, 3 edn (Boston: Heath, i95o; first published (lines 19"), pp. 4-5 90-98).

93 Anglo-SaxonPoetry,trans. by Bradley, p. 414. 94The Old English Version by People, English Bede's Ecclesiastical Hstory ed. and trans. of of the Thomas Miller, z vols, Early English Text Society,o.s. 95,96 (vol i), no, in (vol ii) (London: Early (book English Text Society, 189o-i898), 1,344 4, chapter2.5).
9sKatherine O'Brien O'Keeffe, Visible Song. Transitional Literacy in Old English Verse,Cambridge Studies in Anglo-Saxon England, 4 (Cambridge: Cambridge University Press, 1990), pp. 2,5-2,6.

165

his meotodesmeahte 7 modgepanc, he weorc wuldorfxder, swa wundra gehwaes, Drihten, or onstealde. ece he rest sceop eorran bearnum heofon to hrofe halig scyppend; pa middangeard moncynnesweard, Drihten, ece defterteode firnen foldan, frea aelmihtig. 96
Now we must laud the heaven-kingdom's Keeper, the Ordainer's might and his mind's intent, the work of the Father of glory: in that he, the Lord everlasting, appointed of each wondrous thing the beginning; he, holy Creator, at the first created heaven for a roof to the children of men; he, Keeper, Lord for Ruler, fashioned mankind's everlasting, almighty afterwards 97 mortals the middle-earth, the world. Clearly, the key features of these two creation-songs, Cadmon's Hymn and the Beowulf be heavens to the passage,seem an emphasis on and the earth as the twin central aspectsof the creation, and a concern with God's relationship to humans; God is presented as creating for benefit Maxims humankind. These features in key the world specifically the of also appear I(B), whose final lines perfectly sum up the main points of the poem, namely that God made ('on live'), `rume 'Pe he lifgab' the skies, the that roderas', and the earth which we and we on ('he Creator `paet is humankind is is the meotud sylfa' of everything: and completely controls himself). the creator

Scholarshave,then, been rather distractedby the context of Maxims I(B). In the lacking Old English large the explicit verse,and miscellanyof manuscript context of a framing of the creation-songsin the Old English Bede and in Beowulf,it has not been AngloThis is Maxims I(B) itself basically is that not to say a creation-song. noticed that

96 The Old English Version Bede'sEcclesiastical (book History, 4, chapter z5). T, 344 of 97 Anglo-Saxon Poetry, trans. by Bradley, p. 4; one must note, however, that Bradley translates the Northumbrian version of the Hymn, found in some Latin manuscripts of Bede, and that this version differs slightly in wording from the West Saxon version given above. In particular, note that `eorban' in line S is not the Northumbrian reading, and we should therefore replace Bradley's `children of men' `he (or `heofon' in line `eorban' 6; `children understand of the earth' as an object parallel with with first created the earth [and] heaven for a roof for his children').

166

Saxonsof the late tenth and early eleventh centurieswould necessarily havethought in terms indeed, in that of a genre of creation-songs;nor, should we necessarily they thought suppose What in C&dmon's Hymn, however, is the terms of genres modern sense. that we can say, Old items English (both for for Anglo-Saxons), late the of one canonical of the poetry us and have for MaximsI(B), latter would provided a reading context and that the poem presents former. For late themes to the very similar tenth- or early eleventh-centuryaudience,then, a Maxims I(B) would be instantly recognisable forming as part of a tradition of verseabout the God's order and wonder of creation. At the sametime, thesetexts celebrating(and perhapsevenreinforcing) the order of
be Liuzza Old English the natural world should seen alongside points out prognostic texts. Reform `the jEthelwold's Benedictine texts that such relate not to orderly world of 98 but `the to movement', private world of monastic preoccupation with times and seasons'. This observation points up the possibility of there being quite various ecclesiasticalmilieux in Woden be found, from the possibly stricter milieux associatedwith the which might Benedictine Reform, to circles in which prognostics and charms (both classesof effective, if be In in dubious, be tolerated. related a sense,this point can also morally magical texts) could in a wider sense to the way which such texts approach the natural world; rather than describing or reinforcing the order of the world, prognostic texts attempt to use and read the structure of the world and of time to their users' own advantage.

In Maxims T(B), therefore, drawing on thesetwin contexts of texts concernedwith features Woden interesting it is that asa extremely ordering or reading the createdworld, direct contrast to God at preciselythe point that the poem returns from describinghuman Woden, Not does God's describing this euhemerise only artifice, the creation. artifice to his God, but it him human by him also points up with contrasting seemmerely making This false deity, heathen curiously ambiguous creator. asa and especially position as a draw England, Woden Anglo-Saxon highly late in is and seemsto unusual presentation of

98Liuzza, `Anglo-Saxon Prognostics',p. z, io.

167

together elements of severalof the various different (and usually separate) usesand reimaginings of Woden which havebeen discussed above. The euhemerisationof Woden, presenting him asa human king, is evidenced as early
(see Bede in The as the eighth century, section 3.5, above). tradition became ever stronger, become had by late Anglo-Saxon period, both in the Anglomoreover, and widespread the Saxon Chronicle and in copies of Bede's Historia Ecclesiastica (seesections 3.5 and 4.2, above). Place-names have been thought to present Woden builder, because his as a supernatural frequently is linked to earthworks, but we should consider the possibility that such name in late Anglo-Saxon period as relating to Woden as a place-names were understood the human king of the distant past who built such monuments (see section 2.2.2, above). The Woden (Woden built idols/temples'), be that statement worhte weos' therefore, can seen as lines, Woden heathen king working along similar presenting as a of the past, rather than a heathen deity, and, more specifically, presenting him as a heathen king associatedwith building and artifice. In the Nine Herbs Charm, in contrast, Woden operates as a figure of illdefined magical authority, perhaps borrowing some of the medical associations of his Roman Mercurius (see supposed counterpart section 4.3.2, above).

We should also remark, however,that Psalms95:5 ('omnesenim dii populorum Dominus fecit'; [pagan] `for but the the sculptilia autem caelos all gods of peoplesare statues, (Woden lord heavens') for line has been identified 62, the made the often asa source worhte 99 (as I(B). Maxims It is is this sameversewhich echoed mentioned weos; wuldor alwalda') of De Antichristo in section 4.2.!, above)in the passage Adso's which attracted the translation of O Napier These Mercurius Jovis by prr inn in two vernacularresponses to and of and 42. I(B) from Maxims differing Napier in in the verse, one considerably and 42, are revealing Scandinavian Where Napier immediately deities, the the to another. 42 turns author of I(B) Woden, English figure functions Maxims the prefers who not simply asa author of deity, but also in the more variouswaysoutlined above.Once again,Maxims I(B) can be seen focussing Woden, on the complexity of as not taking the obvious synecdochicroute of

168

Scandinavian deities [... ] `omnes dii Instead, I(B) Maxims to choosing represent populorum'. Woden, him idols both thus with uses associating and gesturing towards euhemerisationand demonisation; Woden is here a human figure of the past, but is also associated his manufacture of idols in numerous saints' lives circulating in Anglo-Saxon England.10 The presentationof Woden in Maxims I(B), then, seemsto draw in elementsof a
He deity. is at once human, but emphatically heathen, wide variety of re-imaginings of the false God's is and possessedof an apparently true creative power, which contrasted with Both Late be in creative power. magic and artifice seem to this creative power. combined Anglo-Saxon England can be seen as developing various `ideasof Woden', but we should fluctuating that this recognise complex of portrayals operates as part of a and malleable figure. In Maxims Woden I(B), conception of a single occupies an extraordinary middle heathen deity, space,not quite characterised as a nor yet as a royal ancestor or a magicofigure, depicted but in medical sympathetic some way to all of these as relating The draws characterisations. poem on aspectsof various, usually mutually exclusive, figure, Woden late Anglo-Saxon this tend to re-imaginings of conceptions of and while I(B) highly his Maxims deity, the attempts complicate or gloss over status as a pre-christian heathen deity, difficult, Woden task of presenting yet not a emphatically as a unusual, and deity.

through

demons inhabit idols humans the the with who of euhemerised

"" On the inhabitation of idols by demons in Anglo-Saxon hagiography seePhilip A. Shaw, `Miracle V Germania Latin Motifs?, in Group Mythographic Sources Norse Hagiographic for Magic: of a as Miracles and the Miraculous in Medieval Latin and Germanic Literature (forthcoming).

99Biblia Sacra,ed. by Weber and others, i, 891(Psalmi,95:5). My translation.

169

Conclusion 4.5
The uses of Woden in late Anglo-Saxon literary discourse are, then, still more Wodan Frankish in English in than the various uses of circles the seventh and eighth and In in this centuries. part, may reflect a relatively wide variety of socio-literary milieux which Woden was available in Anglo-Saxon England, from the ecclesiasticaland royal elites Equally, have been learned through to ecclesiasticswho may than socially prestigious. more distance English Wodan, from this multiplicity reflects the the cult of allowing of authors different in freedom Woden develop them considerable to elaborate and their portrayals of directions. An interest in classical learning clearly plays a role in these developments as well, although not in as an earlier period Woden to the exclusion of placing within

for I(B). instance, Maxims in traditions vernacular and contexts, as,

This interest in different traditions within which to locate and understandWoden is

deity Obinn. Scandinavian development interest in in the the of an also reflected, moreover, English authors at this period understand Obinn as being a straightforward Scandinavian discussed in for Woden. This is section 2.2.3, not entirely surprising, since, as equivalent between dialects the phonemes readily perform a mental translation speakers of closely-related Danelaw Old in English dialect; dialect speakers the and those of the other of their own Obinn Woden, since the two would as equivalent to would, then, quite naturally recognise have however, As we need not assume seen, we sound as though they were perfect cognates. has been that they were straightforward cognates; our understanding of them as cognates English by these very authors. shaped particularly strongly

Obinn, however,help to developwhat has been termed abovea The English usesof Obinn in be Woden identical Woden-Obinn complex. English authors consider to and but (rightly these the of take existence of advantage nevertheless or wrongly), origin Obinn The figures. Woden asseparate and alternative namesas a meansof presenting

170

beyond distinction, however, development such a simple complex extends allowing the of very different usesof Woden and Obinn within different cultural spheres. These English developmentsdo not simply disappear Anglowith the end of the Saxon period. As mentioned above,post-Conquestmanuscriptsinclude illustrations of the (Woden layer his Anglo-Saxon (see top the and sons) of some of royal genealogies section Woden in literary figure deity both 4.1, above). also appears more contexts, as of a and a English antiquity in Geoffrey of Monmouth's Historia RegumBritannie and the texts based (on it below), in which, see on section 5.!, aswell as a more straightforwardly reprehensible 101 Jocelyn's heathen Vita in Sancti Kentigerni. god guise asa In late Anglo-Saxon England we seethe point of junction at which Woden and
s

Obinn become inextricably linked. At the same time, this is a context which proves fruitful Obinn, for later Scandinavian the skaldic conception of also on which much of the draws. This later Scandinavian furthermore, be mythography can also usefully mythography, Woden in it is simultaneous. considered relation to the post-Conquest uses of with which From the junction of Anglo-Danish England, then, we now proceed to the literary and Obinn in Scandinavian cultures and societies. religious uses of

ii, 101 January, ActaSanctorum:

ed.

by Joannes Bollandus and others (Brussels:Greuse,1863), p. 101,

(chapter 6). column i

171

Drinking in the Past: Snorri, Saxo and the Invention of OSinn

Introduction S.i
It has been argued above that the cult of Wodan was relatively limited in geographical and chronological extent, and that modern scholars' understanding of this cult has been strongly shaped distorted and by the understandings and uses of Wodan evidenced

In fact, the works of eighth-century scholars recreate amongst eighth-century scholars. Wodan as a major deity, who was then re-imagined and re-used in still more various ways, in Anglo-Saxon England. Despite Second Merseburg Charm's the particularly tantalising Continent, England it is have in traditions the that we suggestion of similar on chiefly for Woden (see in literary taking evidence on a variety of roles christian culture chapter 4, These did Norman disappear for traditions the above). not wholly with conquest, moreover, Woden continues to appear in literature and iconography in the twelfth and thirteenth his in portrayal. centuries, albeit with some anti-Anglo-Saxon shifts

Obinn? (section What, then, of As discussed he be direct above 2.2.3), need not a Wodan, has been direct he indeed, if is in as usually assumed; cognate cognateof even a Scandinavia, Continent from his development differs in that considerably as on the origin, Late England Anglo-Saxon become in showsthe earliestclear clear this chapter. will Obinn. Woden As has been between interchange influence shown above, and and evidenceof Obinn, largely, if knowledge do of and were probably not clearlyshow eighth-century authors Outside is late AngloScandinavian Scandinavia, it in deity. ignorant of this not wholly, Obinn Woden Saxon England, and in post-Conquest England, then, that are associated and be In England, Anglo-Saxon in at once a particular, there seemsto with eachother. from distinct identity figures, of the two and an attempt to portray them as recognition of the

172

Obinn Norman After however, does the one another. conquest, the identity of Woden and ideological both figures the not present same problems, and interchangeable now more -

take on a quality of exoticism which borrows from and makesuse of the genresof legendary history and historical romancewhich developedthrough the twelfth and thirteenth centuries. It is in this context, then, that we should begin to examinethe literary portrayal and developmentof Obinn in Scandinavia. Since the modern scholarlyidea of the cult and
mythology of O has been heavily informed by the mythography of the twelfth and sinn in particular Snorri's Edda, Saxo Grammaticus's GestaDanorum and it will be convenient to begin, as with

thirteenth centuries -

Codex Regius Poetic Edda the manuscript of the -

the chapter on the invention of Wodan, not at the beginning, but at the end, with the late This doubly because Woden in mythography. this the seems appropriate at point, uses of England post-Conquest mentioned at the end of the previous chapter provide a strong indication of the literary contexts and influences which guided and informed the work of the (eighthFar from twelfth- and thirteenth-century mythographers. considering the earliest for Wodan for the twelfth- and thirteenthcentury) evidence cult of either as a context Scandinavian Scandinavian in the context of century mythography, or, worse still, it is legendary history mythography, precisely the twelfth- and thirteenth-century genres of for historical historical literary such and romance which provide the proper and context Scandinavian mythography of the same period.

is, as Our mythographic inheritance from twelfth- and thirteenth-century Scandinavia banes has been discussed the of scholarshiprelating to pre- and non-christian above,one of Germanic religion. If we abandonthe fruitless effort to understandthis material as evidence for such religion, and instead considerit asevidencefor how twelfth- and thirteenth-century however, interpreted this material such religion, audiencesthemselvesunderstood and becomesa rich and fascinating source.SaxoGrammaticus'smonumental, and often difficult, GestaDanorum begins, around the turn of the eleventhand twelfth centuries,a Edda, Prose Sturluson's Snorri literature finds in its and culmination which mythographical

173

' Snorri's day in Iceland. Saxo,however,did not the saga-literatureof which echoeson after The Gesta Danorum is first and foremost a national to set out write a mythographicalwork. history of Denmark. The twelfth century sawvigorous efforts to createa new English identity by meansof historical works reachingback to history so ancient that they necessarily legendary Geoffrey Monmouth's Historia Regum Britannie, in contained much material. of became England (and, in further indeed, in particular, extremely popular afield, especially
Flanders, Normandy and Champagne).2 Geoffrey's work was translated and versified into Latin, French and English, and was edited in Latin as well. ' England, then, possessed a very interest in vigorous writing and reading national history. Saxo's GestaDanorum should be in seen this context, as an attempt to provide a national history which creates a Danish identity, as well as supporting the idea of Denmark as a nation; 4 indeed, it is probable that

1 The Icelandic doctoral by the sagas,which are outside scopeof this thesis, are the object of research Annette Lassen, who is researching Obinn's portrayal in this literature in the Arnamagnxan Institute University Copenhagen. She her the at of outlined proposed researchand gavesome preliminary in `Odinn's Faces: Study Description God According A the the to conclusions a paper entitled of of Age and Genre', delivered at the International Medieval Congress, University of Leeds, 9-12. July
zooI.

'Julia C. Crick, The `Historia Regum Britannie' (Cambridge: Brewer, Geoffrey Monmouth 1985of of Later (i99i), ), Ages Dissemination Reception Middle in the and pp. 196-2.17. 199, iv: 3 The Latin verse version of the Historia Regum Britannie, a product of Brittany, is edited and discussed by Neil Wright: The `Historia Regum Britannie' of Geoffreyof Monmouth (Cambridge: (1991). Wright The French Britannie, by Neil Gesta Regum Brewer, 1985-1991), trans. v: verse ed. and been has Wace's de Brut, Roman is published: recently of which a new edition and translation version Wace's `Roman de Brut': A History of the British, ed. and trans. by Judith Weiss, Exeter Medieval English Texts and Studies (Exeter: University of Exeter Press, 1999). La3amon's Brut is a Middle English re-working of the Historia Regum Britannie which is largely dependent on Wace, but does Brut The is Geoffrey's familiarity direct text as well. standard edition of the with show some La3amon: `Brut', ed. by G. L. Brook and R. F. Leslie, Early English Text Society, zso and 2.77 An edited version of the Historia Regum Britannie (London: Oxford University Press, 1963-1978). began to circulate relatively soon after Geoffrey's text was available, and is now referred to as the First Variant Version, also edited and discussedby Neil Wright: The `Historia Regum Britannie' of Geoffrey Critical Edition, by Version: First Variant A (Cambridge: The Brewer, Monmouth ed. Ii: 1985-1991), of Neil Wright (1988).

4 Blatt suggests Regum Britannie Historia be Saxo's in and that work should seen the context of the Arhundredes Renaessance', in (Franz Blatt, `Saxo, Reprxsentant for det Iz. en similar texts Tusculanum, (Copenhagen: Museum Boserup, Opuscula Gra: by Ivan Saxostudier, ed. colatina,z (p. 13)) 1975), pp. H-19 "

174

Saxo, who very probably studied in France,was able to consult the Historia RegumBritannie 5 in one of its French recensions. Snorri appears have been working in a very different framework. The Codex to
Regius and the ProseEdda can both be seen as (more less) or successfulattempts to information by form the in synthesise mythological provided verse, probably available an oral in Iceland at this time. Snorri's Prose Edda also deals issue the with of understanding skaldic by Snorri's day, both Snorri and the compiler of the Codex verse which was, obscure, and Regius provide prose explanations of the mythology 6 The difficult Eddaic of tradition poems. history in Saxo (from Snorri English historical works) of national which was working gave Anglo-Saxon genealogical material, ' but Snorri's concern was primarily with mythology, decoding (often, as we shall see, the incorrect viewing poetics as providing a means to the decoding) of poetic mythological references.

Both Saxoand Snorri, however,sharea historico-mythological framework which

derived ultimately from Vergil's Aeneid, and which was popularised in the Historia Regum Britannie, amongst other works. This framework saw the Near East but not always, - often, Troy (Saxo, for instance, traces the Esir to Constantinople) - as the source of western European civilisation. This idea of a link with the past, variously represented in writings from late antiquity onwards, underwent a renewed growth in importance in England and on the Continent in the high to late middle ages,and it seemsclear that this idea was available in Iceland and Denmark, to Snorri and Saxo, just as it was elsewhere in Western Europe. The importance of such an idea of connection with the past also led to an antiquarian impulse, a desire for information about, and portrayals of, the cultures of the past. This impulse was Geoffrey in in details historical Thus the expressed many ways, particularly of writing. of

5 On Saxo's education and its French influences Karsten Grammaticus Latin Friis-Jensen, Saxo see as Passages Danici, Poet: Studiesin the Verse Instituti `Gesta Danorum, Analecta Romana of the (Rome: Bretschneider, supplementum 14 1987),pp. 13-18.

6 Snorri Sturluson:Edda: Skldskaparmal, Viking (London: Society Anthony Faulkes, by z vols ed. Faulkes. Skaldskaparmal, by for Northern Research,1998),1, S.Hereafter referenced ed. as

175

Monmouth himself makesa point of recording significant snatchesof `Saxon'language,such (Old longas English `ciulle, dicimus' `Nime the naues quas word ceol),and the phrase oure as 8 knives', Hengest by ', `take British to order the murder of the used up your chieftains. saxas! Both these examplesoccur previously,in the Historia Brittonum, but Geoffrey does not
he linguistic them; them, thus their also glosses simply repeat emphasising antiquity and interest! One particular area of interest for eleventh- to thirteenth-century antiquarians was Here Geoffrey that of the pre-christian rituals and religious practices of the past. of Monmouth is particularly interesting, and entirely independent of previous sources, as far as He records we can tell. `Saxon' language again with snatches of drinking an elaborate

Hengest's daughter Renwein: by ritual performed Accedens deinde propius regi flexis genibus dixit: `Lauerd King, Waesseil!' At ille, uisa facie puelle, ammiratus est tantum eius decorem et incaluit. Denique interrogauit interpretem suum quid dixerat puella et quid ei respondere debeat. Cui interpres dixit: `Uocauit to dominum regem et uocabulo Quod "Drincheil"'lo debes, honorauit. est autem respondere salutationis Then, approaching nearer to the king, on bended knees, she said: `Lord king, Wassail! ' But he, having seen the girl's face, marvelled at her beauty his he At last interpreter hot desire. what the asked with exceedingly and grew The interpreter her in he had said response. ought to say to said and what girl What lord honoured king `She him: you with a greeting. and to called you "'. "Drinkhail! is in you ought to say response

for instance, This is, of course,rather reminiscent of the drink-distribution scenes of, Beowulf,and, aswe shall see,the distribution of drink is not simply an interesting archaic Obinn, but and, more specificallya recurrent theme of the mythographyof practice, deities, Geoffrey also mentions quite a number of pre-christian therefore, of this chapter also.

Bern, Brewer, (Cambridge: Monmouth 8 The `HistoriaRegumBritannie' of Geoffrey i: 1985-1991), of (Chapter (Chapter Wright (1985), MS. Neil by 65 Burgerbibliothek, io4) 98) and p. 70 p. 5-68, ed. respectively. 9 SeeNennius,ed. and trans. by Morris, pp. 67,73 (chapters31and 46).

' Anthony Faulkes, `Descent from the Gods', Mediaeval Scandinavia, 116981 [for 1978-791),92-I25 (p. 99)" Faulkes discussesthe availability and use of English genealogical material in Scandinavia in (see Lists' Regnal Genealogies `The detail in section 2.3.2). and more

lo The `Historia Regum Britannie', i: Bern, Burgerbibliothek,MS. 568, p. 67 (Chapter ioo).

176

deities have been his his interest to to some of appear references to such of special and Variant First Version Historia Regum Britannie, of the audience, such as the author of the " For Geoffrey's referencesto pre-christian religion. who augments and clarifies a number of instance, in chapter 98 of the Historia, the First Variant redactor replacesGeoffrey's deities `qui mundum istum gubernant' (`who control this world') with deities `quos coluerunt patres (`whom forefathers deities our worshipped'), thus emphasising the positioning of these nostri' 12 At the than their the same time, this alteration also past, rather within supernatural power. (or deities belonging least these to the redactor's to the narrative voice's), and as presents at forebears; for Latin history, it is indeed, to the audience's, own a unusual and striking in Anglo-Norman composed an behalf. Saxon to milieu, claim ancestry on the audience's

Clearly, we have here a significant engagement with pre-christian religion and its cultural and historical contexts.

This interest in the religions of the past may in part reflect the contemporaryliterary by Geoffreyhimself, aswell asby the interest in the spreadof Islam within Europe, evidenced been have deRoland, which appears dissemination pattern of works such asthe Chanson to 13 A literary in frequently creation of a religious copied the thirteenth century. particularly 14 Muslims in be form in can seen such texts, which prompts authors also to of other the forefathers. At the sametime, twelfth- and question the non-christian religions of their for Scandinavia the examination of nonan extra stimulus possessed thirteenth-century 15 In Denmark Baltic form in and crusades. of the christian religions, past and present, the

13La Chansonde Roland, ed. by Cesare Segre, Documenti di Filologia, i6 (Milan and Naples: Riccardo Ricciardi, 1971),pp. xxxvii-xlvii. 14On such literary creations, seeR. W. Southern, Western Viewsof Islam in the Middle Ages (Cambridge, MA: Harvard University Press, 196z), pp. z8-33. 15For a discussion of Peter von Dusburg's literary response,in his Chronicon Terrae Prussiae,to this Peter Seen by Prussians Frontiers: `Mental Vera I. Matuzova, Baltic as see paganisms, with encounter Murray V. Alan by Frontier Conversion Baltic Crusade in Dusburg', ed. r1So-r5-oo, and on the von (Aldershot: Ashgate, zooi), pp. z53-59"

MS. p68,p. 65 and The `HistoriaRegum 12The `HistoriaRegumBritannie', i: Bern, Burgerbibliothek, My 88. Britannie', ii: The First Variant Version, translations. p.

11Wright notes a number of cases(The `Historia Regum Britannie , it: The First Variant Version,pp. lxxv). 1-Iii and xliv-xlv, xlviii,

177

Iceland, perhaps unsurprisingly, we find a concern with pre-christian religion and mythology Europe. in Snorri brings to that evidenced elsewhere this interest to its very similar Prose Edda, his in heathen deities into to which attempts apotheosis rationalise a draws both and mythological cosmological system which christian and classicalpagan on influences (one might compare Mfric's much earlier effort to harmonise heathen and classical deities, The however, pagan on which see section 4.2, above). this seedsof approach are, Saxo's in Gesta Danorum, Codex Regius, to some extent visible the main while the Edda, Poetic be Snorri in direction the to manuscript of the appears an attempt which more fully explores, providing brief prose explanations and introductions alongside the poems it records.

The sheer extent of the mythographicalworks of Saxo,Snorri and the Codex Regius Obinn in in these precludesa complete examination, the current study, of everyaspectof In A Obinn in thesetexts must remain obscure. works. any case,somemotifs associated with
full-length consideration lies outside the scope of this thesis, moreover, since, as has been Obinn Wodan/Woden. from This distinct is in the preceding chapters, substantially shown binn in important discuss the works the most as portrayed aspectsof chapter will, therefore, O inn, how Snorri, Saxo these may relate to early skaldic conceptions of and consider of and Christianity heathenism. liminal between and position which perhaps occupy an unusual

This will lead back to a considerationof what can be determined about the nature, developmentand extent of the cult of O inn in Scandinavia.

Friggjar Fagnafundr ni6ja: s.z, Verse

Oinn's

Central Role in Skaldic

O inn is figure in importantly, figure in skaldic a crucial skaldic verse,and, a crucial Bragi in Already the poetry of phenomenon. as a socio-literary verse'sself-construction O in a lausavisa, in a characteristicskaldic guise- namelyasthe Boddason, inn appears, deity of skaldsand skaldic verse: patron

178

Skald kalla mik, Vi6urs, skapsmi6 Gauts gjafrgtu6, 6hneppan, grepp Yggs Qlbera, 6s skap-M6 a, hagsmi6 bragar. 16 Hvat er skald nema Pat?
They call me a skald, Viburr's fashioning-smith, Gautr's gift-getter, a prolific Yggr's beer-carrier, fashioning-M6 i, poet, the skilful-smith of poetry's What is if poetry. a skald not that?

This verseis recordedby Snorri Sturluson in his Skaldskaparmal. Unlike most of the Snorri Skaldskaparmal, by in this piececomeswith a small piecesof verseused asexamples Bragi Snorri `petta kvab hinn p gamli er amount of contextualising material. recountsthat han 6k um skog nokkvorn sib um kveld, J stefjabi tr9llkona hann ok spurbi hverr par for' ('Bragi the old said this when he was travelling around someforest late at night, when a troll17 follows There him then with a verseand askedwho wastravelling there'). woman accosted form but by in to that above, purportedly spoken the troll-woman, and a verse very similar beginning `Trpll kalla mik' ('they call me a troll'), and ending `Hvat er troll nema pat? (`what Obinn, Obinn 18 by by but byis a troll if not that?'). This verseidentifies three the name not Ribe fragment from The Yggr. Gautr inscription Viburr, may runic on the skull and names, Obinn, below for discussed by-names for as show the earliest evidence the existenceof (section 5.6). The by-namesfrom Ribe, Ulfuit and High-Tr, are not, however,otherwise from by That by-names in the ninth were used skalds attested pre-eleventh-centurysources.

lb Skaldskaparmdl, ed. by Faulkes, i, 83 (verse3oob); seealso Den Norsk-Islandske Skjaldedigtning,ed. by Finnur Jnsson, 2 vols (IA, IB, IIA, IiB) (Copenhagen and Kristiania: Nordisk Forlag, 1912-1915), IA, (Lund: Gleerup, Kock, Ernst A. Norsk-Islndska by Skaldediktningen, Den 1946z,vols rev. S, and by-name p6rr, Mobi is As Faulkes My translation. rather than another a son of points out, 1949), 1,3" Obinn (Skaldskaparmal, ed. by Faulkes, II, 492) for

17Skaldskaparmal, Faulkes, My by 83. translation. I, ed. 18Skaldskaparmal, (verse Faulkes, My by translation. 1,83 ed. 3ooa).

179

doubted. hardly be A list by-names need useful annotated of century onwards, nevertheless, O inn in skaldic verseis provided by de Vries, and a frequencycount of ninth- and tenthfor 19 list following in by-names yields the given this results: century

Name Ba'leygr
F'plnir Gautr

Frequency I
4 i8
3

Grimnir Han a 'r Harbarbr Hoar/Harr Hroptr

Jlfabr Jalkr

i I 6 7
I 3

J6lnir
N' tr Ofnir

2
4 I

Rp

it

7
I i

Svfnir Svei bir

Svplnir
rPribi
Pror

5
4

z,

rottr

Pundr Vi&rir Viburr Vpfu&r

7 8 7

Yggr
Total:

14

As the frequencycount shows,Vi surr, Gautr and Yggr are simplex by-nameswhich are Vries de by list The that in shows given tenth-century verse. skaldic common ninth- and
het Bijdragen Nederlandsche Inhoud, Mythologischen De 19 Skaldenkenningen op In J. de Vries, met Willink, D. Tjeenk H. (Haarlem: Linguistiek, Philologie Germaansche 1934),PPGebied van 4 en

180

20 later That in they appearrelatively they remain quite common skaldicverseaswell. Obinn; frequently suggeststhat they were among the earliestby-namesfor that they are forms, strengthensthis suggestion,since one might expect simplexes,rather than compound Gautr later. its likely have been forms develop to to and cognatesalsoseem compound for legendary Germanic by figure area or mythological over most of the availableas names a for Geat/Geot Anglo-Saxon in the ninth century, since the cognate appears royal genealogies Gothic Amal head Lindsey and East Anglia, and Gapt appears the of at the of the genealogy 21 Origine Jordanes's De Actibusque Getarum. The contents of this verseare, then, dynasty in later, be given the continuing much plausibly early, although they could perfectlywell
by-names. popularity of these

Of particular interest is Bragi's use of the phrase`YggsQlbera', which we may translate

be beer Yggr'. The beer beer-carrier/beer-server `the as we referred to need not of as drink but it, certainly some sort of alcoholic understand Christine Fell, in her

Norse, Old Old English drinks denoting in that accepts and words alcoholic of examination Old Norse pl and Old English ealu both referred to malt-based drinks, although we should in in be cautious about assuming that they were always used such a strict sense perhaps 22Yggr, as discussed above, is a by-name of Obinn. Bragi is describing the skald as the verse. Obinn; Obinn, for beer is, beer-carrier of since the skald the person who carries that Obinn. for beer transmits or carries poetry, we may understand the skald's poetry as drink be described in Essentially this sets up a metaphor as an alcoholic which poetry can Ofinn. This is of the existence a contemporary suggests to clearly related somehow which

36-So. based Vries's de however, 20De Skaldenkenningen, One on a that claims should note, pp. 36-50. heathen late for he lists kennings the of use of twelfth-century revival a the of statistical analysis (Fidjestol, be been literatures have Norse Old in pp. to unfounded shown convincingly mythology

' Christine E. Fell, `Old English Beor', Leeds Studiesin English, n. s. 8 for 1975(1976), 76-95 (pp. 8489).

For Collection', Anglian The " For the Anglo-Saxon genealogies Dumville, the pp. 31,34. see Antonino Giunta by Francesco Getarum, Actibusque lordanis De Origine and Gothic genealogy ed. see (Chapter Italiano, Istituto Storico (Rome: Storia d'Italia, la Fonti 14)Grillone, 199i), p. 36 117 per

2-70-93)-

181

lines Mead Poetry found in the the mythological narrative along of of narrative earlier Skcildskaparmdl.There are difficulties with this, however. Firstly, this kenning single can tell Obinn little it is us about the narrative; all evidences an association of with poetry as alcoholic drink we cannot take this as evidence that Obinn was thought to have stolen this drink

it Secondly, be and made available to men. we cannot entirely sure that the verse is genuine, it is Skaldskaparmal. in Kock Yggs Yggs that or correctly preserved amends Qlbera' to line displays internal (`skal-'/`kal-', the that Qlbrugga', on grounds the each of verse assonance `smiV/`Vib-', etc), except for this one; the emendation creates `Ygga similar assonanceon '/`brugg-'. 23This is an attractive argument, and if Kock is right, then Snorri's elaborate Mead Poetry for brewers Mead in narrative of the of the episode collapses, this of the dwarves, narrative are not the skalds themselves, who merely transmit poetry. Kock's is Haraldr by har&rabi's his Orottir, in he emendation confirmed, moreover, verse on which `IJrottir kank tta: / Yggs fetk lib at smiba' (`Eight arts I know; / To brew the drink states: 24Here Ygg'). have indeed imagine that themselves as the of we confirmation skalds could brewers of poetry, which suggeststhat Snorri's narrative of supernatural brewers of poetry is fanciful. As perhaps rather we shall see, moreover, there are other, still more compelling for distrusting Snorri's in reasons reading of skaldic verse this matter.

The manuscript variants and meter of this versedo not provide any reasonto suppose that the versehas sufferedin the courseof its oral transmission,assumingthat it has indeed lengthy processof oral transmission;thesemanuscriptvariants,moreover, undergone a 25 brief, This if does not alwayscareful,written transmission. suggestonly the errors of a not did lengthy being indicate that the transmission to verse not survivea oral prior necessarily The down; is is we simply cannot tell. question of whether or not the verse genuine written As discussed is in dates it there the nothing versewhich necessarily more problematic. above,

24Den Norsk-Islndska Skaldediktningen,rev. by Kock, i, 166; translation by R. I. Page, in Chronicles Records, Memorials Myths (London: Press, Vikings: British Museum and 1995),p. 168. of the 25Den Norsk-Islandske Skjaldedigtning,u, S.

23Ernst A. Kock, NotationesNorranv: Anteckningar (Lund: Edda Skaldediktning, till Svols och (paragraph Gleerup and Leipzig: Harrassowitz,192,3-1944), 8 II, part 7, p. iooSC).

182

Bragi's later time; at the same time, there is no evidence that shows irrefutably that it to than We in doubt Snorri's the century. ninth may, of course, was composed explanatory narrative dialogue between Bragi but in itself is two the a troll-woman, verses as this placing and a not for being Acknowledging the verses not a genuine product of the ninth century. evidence difficulty Snorri how Mead that this remains, we may profitably consider the presents of Poetry narrative, and how this may, or may not, relate to the mythological witnesses of skaldic verse. The Mead of Poetry episode as recounted by Snorri in Skaldskaparmlis an elaborate fairly involving large Frank has demonstrated narrative a cast of mythological characters. Snorri convincingly that many of this cast are the result of misunderstanding the skaldic Kvasir he for Snorri's be instance, that poetry about which was writing; she shows, need not deity, but denoting fruit in the crushed or pulses used an otherwise unrecorded simply a word brewing (related to Danish kvas `crushed fruit'). 26Frank suggeststhat many of the skaldic kennings relating to the Mead of Poetry follow a simple pattern in which poetry can be (or `that Obinn dwarves `liquid the poetrytherefore, of or giants)', and, referred to as of kennings of the early skalds had as their basea single concept that of verse as an

intoxicating drinki27. While both suggestions are broadly true, they give the impression that Snorri's mythographic efforts have created the entire Mead of Poetry narrative from nothing liquid They ignore this also any connection of more than an extended group of metaphors. This is Bragi's in in misleading. verse quoted above. the early skaldic verse, as sinn with 28 Egill Skallagrimsson's mid-tenth-century Sonatorrek opens with the following stanza: Mj9k erum tregt O

hroera, tungu at

in Memory Norroenum:NorseStudies 26Roberta Frank, `Snorri and the Mead of Poetry', in Speculum University Press, (Odense: Odense Ursula Dronke by Turville-Petre, Gabriel 1981), and others ed. of (pp. 159-6o). PP-155-70 27Frank, `Snorri and the Mead of Poetry', pp. 169-70;seealsopp. i62-68. (IA, for Sonatorrek date 28 Skjaldedigtning, J6nsson,in Den Norsk-Islandske 40); 96o asa suggests between later, Vigfusson and Powell date Egill's composition of Sonatorrek to 975and 980 somewhat (CorpusPoeticumBoreale:The Poetryof the Old Northern Tongue,ed. and trans. by Gudbrand book Vigfusson and F. York Powell, z vols (Oxford: Clarendon Press,1883), t, 4,2-76).

183

loptvega lj6 pundara. Esa vaenligt Viburs Jai of hogdraegt n6 6r hugar fylgsni. 29

It is very hard for me to stir with my tongue the air-path of the burden of There little for is hope Viburr's it be drawn poetry. plunder, nor can easily from the depths of my mind.
Egill here refers to poetry as Viburs pi' (`Viburr's [i. e. Obinn's] plunder'), and in the

he `fagnafundr Friggjar ('the it / / ar Jptunheimum' borinn / or next stanza calls the ni6ja happy discovery of the kinsman of Frigg [i. e. linn], Obinn 30Both kennings past'). appear to reference Frigg'), and once by the by-name Vi'6urr Jptunheim from in the carried off ('the `Friggjar kinsman once as of nibja'

but in fact the reading `Friggjar' does not,

J6nsson, Sonatorrek: in the the manuscript reading appear any of manuscripts of according to is `}iriggia'.31For metrical reasons, the emendation is necessary,and, in the context, almost however, It Sonatorrek is that the versesof were worth remarking, certainly correct. forms later in in in in the the which they appeared manuscripts presumably understood Obinn been lost have Middle Ages, and that the relevance of this kenning to to some or may even many readers of the time.

indicate If Sonatorrekis genuine, nevertheless, thesestanzas that a narrative clearly Obinn by from known Mead Poetry in involving giants was skaldic circles at of stealing the 32 by This is interpretation half latest sometime in the second supported of the tenth century.

Norsk-Islndska Skaldediktningen,rev. by Kock, i, zi (stanza i). My translation. Norsk-Islndska Skaldediktningen,rev. by Kock, i, zi (stanza z). My translation. Norsk-Islandske Skjaldedigtning,ed. by Jnsson, IA, 40. in Einarsson, `Skldi$ (Bjarni i Reykjaholti', been has Sonatorrek questioned genuineness of Eyvindarbo'k: Festskrift til Eyvind Fjeld Halvorsen 4. mai 1992, ed. by Finn Hodnebo and others (Oslo: Institutt for nordistikk og litteraturvitenskap, 1992)) pp. 34-40). Jon Hnefill AMalsteinsson, in his `Religious Ideas in Sonatorrek' (Saga-Book, z (1999)) 159-78),acceptsthat Egill's authorship cannot half in but be the second the sometime composed poem nevertheless was probably proved, absolutely here This is as not an unreasonable position to take, and servesour purposes of the tenth century. Abalsteinsson. it well as servesthose of

29Den 30Den 31Den 32The

184

Frank's analysisof many `poetry-kennings as taking the form `liquid of Obinn/dwarves/giants' (seeabove); such a group of kennings could not havearisenwithout the existenceof a myth Obinn Frank is kennings have been to right to claim that some and giants poetry. connecting interpreted in an unjustifiably mythological light by Snorri, but this cannot be true of by its relationship to Obinn. In such cases, kennings where something is characterised the kenning could not exist without a myth least or at a mythological attribute by which it

be It for interpreted. instance, to refer to poetry as could simply would not make sense, .11 link `Obinn's if ale' there wasno mythological howeverslight betweenObinn and

`the `the be wolf's road' rather than much the sameas calling the sea poetry; this would Clearly, kennings do require the existenceof some natural or mythological whale's road'. Obinn, but prr, (i. by is but somehowrelated to not context e. whales, not wolves, travel sea; 33 by be if to understood their audiences. poetry) they are The status of such a myth is, however,more difficult to establish.The success of the
kenning depended upon its audience knowing a myth with reference to which they could however, have been This kenning. a myth of central religious myth need not, understand the importance; it could have been simply an aetiology appended to a deity. Nor need it have been part of an early, native mythology; in the caseof `poetry-kennings in skaldic verse we We itself a creation of skaldic poets. might reasonably suspect that the underlying myth was linn in Sonatorrek from stole which some giants owned poetry, and a myth can reconstruct `liquid for `poetry fairly This from it the of straightforward explanation = them. presents a Obinn/giants' kennings. One form in the of some need only assumethat poetry was owned liquid in this myth. sort of

An interesting parallel for this narrativeof theft of liquid from enemiesis presented preatum,/ monegum maegpum by Scyld Scefing in Beowulf,who `sceapena meodosetlaofteah'

33As Bjarne Fidjestol has rightly noted, de Vries makes a similar assumption, in De Skaldenkenningen, in `a brief kennings, for a mythology of aspect that of some mention assuming of mythological context kenning must presupposea more general and widespread knowledge of the underlying myth, without (Fidjestol, be kenning p. 270). comprehensible' would not which the

185

('took away the mead-benches from the hosts of enemies, the many races').34In this interpreted `deprived'. `ofteah' is This impression often simply as that statement, gives the Scyld is thought of as simply having destroyed the mead-benches, but this does not seem likely; `ofteah' has a more specific implication, that Scyld carried off his enemies' meadbenches. Mead-benches are not the same as mead, of course, but the fact that Scyld plunders his enemies' mead-benches is sufficiently striking that it deservesour attention. It is Beowulf-poet in that the extremely strange should choose mead-benches this context; they hardly constitute an obvious target for plunder. It would presumably have been more financially rewarding, and not metrically problematic, if Scyld `monegum mxgpum maxima ('carried from It Beowulflikely, that the many races'). seems then, the ofteah' off treasures furniture; felt had that mead-benches were not simply they some greater significance. poet In the context of Beowulf, this significance would appear to be related to the social Wrenn Bolton for drinking. include this passage stating that and a note rituals surrounding `the mead-bench is a symbol of the independence of the Germanic chieftain: to be free and he distribute from high bench have hall he treasure and mead could whose a must respected but drink-giving, it drinking importance This his of and to statement acceptsthe retainersi3'. This hardly be in important drink need exactly the same ways. and treasure are suggests that lack drink however: treasure the mention of and of any the special emphasis on the case, -in this passagein Beowulf indicates that this 'understanding of the lord-retainer

distribution rituals presented in Beowulf is too simplistic. An analysis of the role of drinking drink-distribution and in the courtly milieux depicted in Beowulf strongly reinforces this

fabric important drinking to the social of are extremely suggestion that activities centering on distribution important than the of treasure. these milieux; perhaps even more

With this in mind, let us turn to the variousvignettes presentedin Beowulfof the A brief heroic drinking analysisof the structure of one court. rituals of the socio-military

34Beowulf,ed. by Klaeber,p. i (lines 4-5). My translation.

Medieval Exeter W. Bolton, F. Wrenn 35Beowulf. C. L. Fragment, by Finnesburg With the and ed. " ), p. English Texts and Studies, 4th (revised) edn (Exeter: University of Exeter Press, 1988;repr. 1992. 98.

186

such vignette -

Beowulf's first feastin Heorot

helps to establishthe casefor the

Beowulf Heorot, drink-distribution. is centrality of arrivesat and granted an audiencewith Hropgar.36The audiencefinishes with Hropgar instructing Beowulf `sitenu to symle' ('sit 37 We feast'). feast The take this the now at the may asmarking opening of the episode. immediate concern of the poet is with seatingarrangements:

Pa wxs Geatmaecgum geadoraetsomne benc beorsele gerymed; on

Jx r swibferhpe Pro um dealle.38 sittan eodon,

Then a bench was madereadyfor the Geatishmen together in the beer-hall; in there the courageous to men went sit, proud their strength.
This is certainly a logical next step, but hardly a necessaryone; it would be obvious enough Geats Again, Beowulf-poet to any audience that the the were provided with seating. appears to find something particularly significant about benches on which one sits to drink; the The feast benches `on beorsele'. is audience reminded specifically that these are episode then 39 Unferh flyting Beowulf. This lengthy, between is into the and a and apparently progresses important, part of the feast episode, but it need not concern us here. With the flyting `Dxr haeleka is hleahtor, description brief wxs of merrymaking provided: general complete, a hlyn swynsode, / word wxron wynsume.' ('there was there the laughter of warriors, the noise 40 Wealhpeow At in joyful'). this point, appears,and a passage ensues grew, words were

41 Beowulf, her drink This drink. distributes to and providing culminateswith which she is not solely his reception of the cup. The passage Beowulf making a boast to accompany Wealhpeow's is devoted it Beowulf, however; to around a third of concernedwith distribution of drink to everyoneelsein the hall, starting with Hropgar -- 'Pafreolic wif ful

36Beowulf,

My (lines Klaeber, 611-i2). by translation. p. 2.3 ed. 41Beowulf, (lines Klaeber, 612-41). by pp. 23-24 ed. 40Beowulf,

37Beowulf, ed. by Klaeber, p. is (line 489). My translation. 38Beowulf, ed. by Klaeber, p. i9 (lines 491-94). My translation. 39Beowulf, ed. by Klaeber, pp. 19-2,3(lines 499-6io).

13-15, (lines Klaeber, by 0-98). 3Z. pp. ed.

187

('then ' / East-Dena first the epelwearde; noble woman gesealde serest gavethe cup to the
land-protector of the East-Danes')42dugA in to the and moving on and geogo6 turn:

Ymbeode pa ides Helminga / dugupe and geogope daelaeghwylcne,/ sincfato sealde' ('then Helmings lady the went round each section of the tried and the younger warriors, gave of the 4' There is (albeit limited inclusivity treasure-cup') to them the a strong emphasis on social . lesser importance) here; Beowulf-poet the male aristocrats of greater or stressesthat Wealhkeow goes around `dxl aeghwylcne'('each part') of the dugO and geogA At the same Wealhpeow follows first king, drink time, to the clearly a system of social status, giving and followed by junior in then to the more senior retainers, the retainers, order of social Clearly, drink distribution be being is to of considerable social precedence. understood as importance in the society depicted in Beowulf. This drink-distributing (p. Hrojgar Wealhjeow passage goes to sit with ends when

framed brief description is then of merrymaking, with another 24,64o-641), and the passage feast, Hropgar's bed, desire into leads immediately to go to which ends the and the which feast episode:

Pa wxs eft swaair

pryyword sprecen, sigefolca sweg, Healfdenes sunu

inne on healle
&eod on saelum, semninga op paet seceanwolde

2efen-raeste44
Then there was afterwards, as before, noble-speech spoken in the hall, the in joys, the the until presently son of the victory-peoples, of noise people Healfdene wanted to go to his evening-rest.

is descriptions between drink-distribution framing This two cursory of merrymaking of the drink, distribution interest in Beowulf-poet's It highlights interesting. the of special the very

My translation. '2Beowulf, ed. by Klaeber, p. 24 (lines 615-16). importanceof the femalerole in 43Beowulf,ed. by Klaeber, pp. z4 (lines 6zo-zz). On the considerable J. Michael Enright, by detailed Celtic Germanic drink-distribution in study cultures,seethe and La 7'ene Warband from European Lordship in Prophecy Ritual, Cup: to the Lady with a Mead the and VikingAge (Blackrock: Four Courts Press,1996). " Beowulf,ed. by Klaeber, p. z4-z5 (lines 64z-46). My translation.

188

far interest interest he have had in that this and suggests outweighed any or she might depicting feasting and merrymaking in general. The framing device also makes it clear that the drink-distribution interrupts the merrymaking; it is not part of the merrymaking, actually

but a separate, privileged activity which takes precedenceover merrymaking. The impression is heightened by the statement which returns the narrative from drink-distribution to

[... ] `pa prybword This merrymaking: wayseft swa aer spreceni45. strongly suggeststhat the drink-distribution is be hiatus in to passage the merrymaking, a time when understood as a its with

is general conversation suspendedwhile the solemn act of drink-distribution, (lines boasts (lines 631-38) is 625-28), accompanying and prayers carried out.

The maintenance of the comitatus in Beowulf depends, in part, on the proper in leader dispenses his to observance of such rituals which the retinue; clearly, the distribution of drink plays a crucial role, and can probably be seen, in Beowulf, as being just as important as treasure, if not more important; treasure is distributed in recognition of Grendel, killing distribution but drink is the particular acts, such as of an essential part of the The in the everyday social maintenance of removal of enemies' mead-benches, seen comitatus. The this context, takes on a new significance. mead-benches are a crucial part of the meaddispensing ritual; seating arrangements matter in the world of Beowulf, as we are reminded (see Beowulf feast him Scyld's Hropgar his invites to the moment with above). plundering of be fabric, seen, then, as undermining the socio-military and the enemies' mead-benches can by depriving them of one networks of social obligation and exchange, of those enemy tribes; Scyld in reduces their ability to maintain of the crucial elements the mead-dispensing ritual, the necessaryrituals of the comitatus.

The samepurposecould, of course,havebeenachievedby plundering one'senemies' Obinn's liquid from help This theft of to explain the origins of giants. mead. may, therefore, linn imagined This liquid was equatedwith poetry by skalds.If was as attempting to cause drink be breakdown would an obvious the of the comitatus-ritualsof the giants, stealing their functioned In do for him the skaldic context, moreover,poetry this. asanother of the to way

4sBeowulf, ed. by Klaeber, p. 24 (lines 642-43)

189

for lord-retainer transactions,although in this casethe dispensingwas items necessary Obinn's This lord. dual from theft suggests, significanceto then, that generally retainer to in have lifestyle, but the this myth may context of a comitatus arisennot only or evencourt in also the context of skaldic circles. It is possiblethat an older mythological narrativeunderlies the form suggested in Sonatorrek.This older narrative might be supposed to predatethe composition of skaldic Obinn in This to verse,and comprise a myth which simply stealsthe giants' mead. would lack, however, the explanatorycharacterof most early myths; it could not explain the origin institution, Nor is tribe, this the only of any particular place-nameor natural phenomenon. The Beowulf debate; is dating is problem with such a suggestion. what clear, of an ongoing however, is that the Beowulf-poet'sknowledgeof (or desireto depict, by modern standards, He fill have been limited. Migration Age to content to or she seems culture was accurately) in many details from later Anglo-Saxon culture, as,for instance,in the caseof Scyld'sseaburial; Owen noted the problem that Scyld's ship would, in all probability, have returned on for Newton has doubting the next tide, while, much more recently, archaeological reasons `mound-inhumation in funereal shipthe authenticity of these arrangements, arguing that funeral ceremonies may have been the norm rather than the exception'46 We should . important drinking development Beowulf-poet's theme of as an of the probably, then, see the known and understood as socially socio-military ritual of the court, as a reflection of practices important in some contexts (although not necessarilyfollowed), in Anglo-Saxon England. There is no reason to suppose that such practices date to the Migration Age, or to an UrGermanic ideal of the comitatus lifestyle. Fell points out that `the strongly emotive heroic loyalties linked is code, a and patterns of the with the terminology of medu very closely That drinking ideal looks rituals past than an actual presenti47. much more to an code which Beowulf in related to medu are reflected heard, least have been read or and which must at

beginning of the eleventh century, when the understood, around the end of tenth or

46See Gale R. Owen, Rites and Religions of theAnglo-Saxons (Newton Abbot: David & Charles, i98i), `Beowuf, The Origins Newton, p. 47. of p. 97;

190

Library, Cotton Vitellius British London, A. xv, was written manuscript,

in and skaldic

have been least in tenth that the century, suggests such rituals may well verse of associated,at literary discourse, with exactly those developments of dedicated court circles with which (on bound date Beowulf-manuscript, intimately the skaldic verse seems up of the see section Such develop in Scandinavia, but Anglo-Scandinavian in court circles 4.4). not only also England, where a mixed court culture with elements of the older, more established English Scandinavian Indeed, Roberta Frank has the courts, as well as courts with their skalds, arises. for Beowulf for tenth-century a shared and tenth-century skaldic argued convincingly context (such legendary Sigemund/Sigmundr, is who verse, pointing out congruences of as personnel Sigurbr later Scandinavian literature) by in and common specialisedterminology eclipsed 48 Scylding/Skjgldung). (lofgeornost/lofgjarn, In particular, she claims that Old English royal developing Anglo-Danish stratum of composition, probably genealogies contain a common in the tenth century: sometime What little Old English and Old Norse evidence there is suggeststhat the Scyldings and their legends may have been an Anglo-Danish innovation. By West-Saxons Scyld, founder the strengthened their named acquiring a Scylding, By Norse Danelaw. king in a skald confirmed the calling a position his patron's English heritage. The Beowulf poet's incentive for composing an do had have Scyldings with the something to may epic about sixth-century fact that, by the 89os at least, Heremod, Scyld, Healfdene, and the rest, were family Anglo-Saxon both be royal and the common ancestors of the taken to 49 Danish immigrants. of the

Frank seesthis as evidencefor dating the composition of Beowulfto the tenth Lapidge, for is Although is her trying to shift one, convincing. quite argument century, and 50 fact Beowulf back Beowulf into dating that the eighth century, the the of the composition of

48Frank, `Skaldic Verse and the Date of Beowulf', pp. 135and 126-29. '9 Frank, `Skaldic Verse and the Date of Beowulf', p. 129. '0 Michael Lapidge, `The Archetype of Beowulf', Anglo-Saxon England, 29 (20oo), 5-41. Lapidge Malmesbury) (perhaps West Saxon date for near and an eighth- to ninth-century earlier argued Wessex', Studi Medievali, Monstrorum Liber `Beowulf, Aldhelm, Beowulf, in for the and provenance his he in but (1982), the place-name evidence adduces support of casecould I5I-92, 3rd series, 23:1 its dissemination indicate the composition. area of of the text rather than an area of equally well

47`Beorin Old English', p. 8o.

191

known desired by does or an eighth-century poet could representsocial patterns not mean known English in to the that the samesocial patterns were not the tenth and eleventh folio Psalter illustration Tiberius indeed, on io" of the mid-eleventh-century an centuries; (British Library MS Cotton Tiberius C. vi) showsChrist being tempted by Satannot simply 51 but by drinking-vessels. with a treasurecharacterised several with treasure, very conspicuous On the whole, the present author is inclined to acceptFrank's view on the dating of Beowulf. Even if one did not accepther evidenceasrelevantto dating, however,her analysisof the Beowulf for sharedcontext and tenth-century skaldicversecertainly confirms the identification of the drinking rituals of Beowulfasoperatingwithin much the samecontext as doesthe Mead of Poetry episodein early skaldicverse. We are brought back once more to the lord-retainer circlesin which skaldic versewas
fostered. These circles provide the perfect situation for the development of the myth of binn's (and from They benefiting from the giants. were theft of the poetry-liquid

This form highly development in) selfthe self-reflexive poetry. of a new of participating form desire have included impulse to explain the origins of this of a must surely reflexive for lords for just divine the whom the poetry was composed origins are posited as poetry, and, (see, for instance, the genealogy of the Haleyg jarls in Haleygjatal), so it would have been 52 These early skaldic divine. desirable to establish that the origins of skaldic poetry were Viking loomed large in image in belong plundering raids to a period of which the circles also Beowulf-poet Anglo-Scandinavian the orbit; perhaps precisely the sort of raids which the imagined Scyld Scefing as undertaking when he `monegum maegjum meodosetla ofteahi53.

Whitelock's claim that Beowulf could not havebeen composedin the ninth or early hear have English because to wanted audienceof this period would not tenth centuries an

(G. Storms, for East Anglian Storms has alsorecently advocated the poem origin a seventh-century, ', EnglishStudies, Get into Beowulf? 8o (1999),46-49). `How Did the Deneand the Geatas sl SeeC. R. Dodwell, The Pictorial Arts of the West8oo-rzoo, PelicanHistory of Art (New Haven: (illustration Yale University Press,1993), io5). p. 117
52For Haleygjatal see Den Norsk-Islandske Skjaldedigtning,ed. by Jnsson, IA, 68-71 and Den NorskIslndska Skaldediktningen, rev. by Kock, 1,37-39" " Beowulf, ed. by Klaeber, p. i, (line S).

192

favourably Danes described the poemswhich -

the very peoplewho were terrorising them

is, then, perhapsnot asstrong as it might be.54The Anglo-Danish interchangeof the

England Danish in (as period of settlement opposedto raiding) providesan ideal context for development idealised Viking have been the of an version of raiding activity which may not Danes England, in but Anglo-Danish It in its is, to the peculiar actually scope. moreover, just such a cultural context which fosteredskaldic versein England, and we should hardly be find therefore, to the central skaldic myth imaginedin a form which shows clear surprised, Danish parallelswith raids.

Hpkon for hei5in External Influences Creation 5.3 me6 goy: and the Heathenism Skaldic Verse in of
The term het'6inn In in by relatively rarely early skaldicverse. versecomposed appears heathen poets this is exactlythe situation which one might expect;`heathen'was a christian heathens doubt had themselves their own no coinageused to refer to non-Christians, while (which in their were not, any case,geographically religious practices waysof referring to different different have in terminology required areas). uniform, and would, therefore, Nevertheless,the term hei6inndoesappearin Hakonarmal, Eyvindr skaldaspillir's (AMalsteinsfostri), Hkon Good dating in for the probably to sometime commemorativeverse 55 heathen for `the `Hkon / final gods': the 96os or early 97os; the stanzaof the poem talks of 56 This is heathen (`Hakon departed heibin to the symptomatic of the gods'). rneb gob' (or heathenate) heathen rather, problematic nature of the poem as a self-consciously Eiriksml, here A king. for although similar problem ariseswith memorial-poem a christian

54Dorothy Whitelock,
15-

(Oxford: Clarendon Press, `Beowulf The Audience 1951; repr. 1967), pp. 2-4of

" On the date of Hakonarrnl,seeDen Norsk-Islandske Jnsson, Skjaldedigtning, by which suggests ed. Vigfusson Corpus Poeticum later date, based Powell Boreale (IA, in date 64); the prefer a and of 961 a (1,262 (book Hkon's death 4)). around 970 on placing 56 Kock, (Stanza My Skaldediktningen, by Den Norsk-Islndska translation. 1,37 rev. 2i).

193

heiNnn. is Vigfiisson and Powell claim there no specific self-characterisationof the poem as ] [... heathen `the dirge baptized king, by is that curious as a pure poem on a composed order This Christian i57. heathen 'pure dirge' is the the real understandingof poem asa of a queen Powell's This Vigfusson later like imitation, its and puzzlement. poem, sourceof Hkonarmal, is not, essentially,a heathen poem, but a heathenate poem, reflecting and heathen-christian by liminal contextualised a courtly poetic milieu. In Hakonarmdl, then, we havea poem which depictsa christian king being received
into a heathen afterlife, thus presenting itself as a heathen poem and yet giving away that it is fact Christian, by heathen both by its it the that commemorates a use poem, not simply a and from hez'6inn, which reveals a poet writing a christian point of view, seeing all of the term lack For deliberate hei inn. Scandinavian all the non-christian religious expression as simply Hakonarml heathen is in not simply a poet's of any christian elements the poem, We heathen-christian but liminal compositional context. reflects a commemorative poem, for Hakonarml, having been EirI'ksmal, in be justified the model as seeing would probably in composed a similar context. From as early as the mid-tenth century, then, skaldic verse appearsto have occupied bridge divide form to the of art which was able an unusual and privileged position as a between christian and heathen contexts. This is perhaps not especially surprising. Skaldic Scandinavian in royal and aristocratic contexts verse was, at this time, popular it

This Scandinavia. literature was, moreover, a period of of constituted, essentially, the courtly in Danelaw, being in Anglo-Danish a period of activity as well as contacts the considerable Danes itself, Scandinavia into Christianity and with the conversion of the the spread of 58 half Norwegians taking place over the course of the second of the tenth century.

57Corpus Poeticum Boreale, i, z6o (book 4). 58The importance of interchange within the Danelaw is discussedin relation to stone sculpture and Collins, (London: England Northern in Sculpture Viking Age Bailey, N. Richard 1980), in mythology Notes', in Methodological Northumbria: Mythology `Norse John Lindow, by and pp. 101-42, and D. John by Conquest, before Period in Relations Norse-English the ed. Anglo-Scandinavian England: the Press University America, (Lanham: Series, Colloquium English Old of Niles and Mark Amodio, 4 Response', A in Northumbria: Mythology `Norse McKinnell, the John and 1989), pp. z5-4o; see also

194

Skaldic verseof this period represents,then, both a response to Christianity and a reObinn heathenism from Naturally deity creation of a christian point of view. the skaldic featuresheavily in the mythological systemwhich results, his importance in skaldic since from circles waswell-established the ninth century onwards,through his pivotal role in the Mead of Poetry narrative (seesection 5.z, above).A well-known exampleof a tension between a court's christianity and the heathenismimplicit in the is craft of a skald the Ottarsson's Ha1le6ar in Hallfrebr (vandrxbaskld) form in account, saga,of complaint, the Olafr 59 lausava, his Tryggvason: to of a patron
Q11hefr nett til hylli O bins skipat lj6 um: algilda mank aldar ibju vrra nibja. Enn traubr Jvit Vi&ris vel hugnabisk vald skaldi legg ek a frumver Friggjar fj n, Pvit Kristi Pjbnum. 60 All our race has made poems in praise of Obinn; I remember the highly I valued practice of people, my ancestors; and, reluctant, turn my hatred on Obinn Frigg first I husband because Christ, for the the rule of of serve pleased the poet well.

On Anglo-Danish Roberta Frank's skaldic milieux and mythology see same volume, pp. 41-52. Date Beowulf' (cited Verse `Skaldic `Did Anglo-Saxon Audiences the of above), and articles, and Have a Skaldic Tooth? ', in Anglo-Scandinavian England, pp. 53-68; in the samevolume seealso John D. Niles, `Skaldic Technique in Brunanburb', pp. 69-78. On the conversion of Scandinavia and Scandinavians see Fletcher, The Barbarian Conversion,pp. 369-416. For a convenient summary of the for Scandinavia, in Birgit inscriptions the progress and processof conversion evidence of runic see Sawyer, The Viking-Age Rune-Stones: Custom and Commemoration in Early Medieval Scandinavia (Oxford: Oxford University Press, Zooo), pp. i47-5Z. Although contacts with England, and English influence, appear to have been important in the conversion of severalareasin Scandinavia,DuBois Tryggvason Olfr Snorri's implausibly the that argues christianisation of narrative of emphasisesthe English role in this christianisation over that of the East (DuBois, pp. 175-79).On the importance of English activity in the christianisation of Scandinavia,seeLesley Abrams, `The Anglo-Saxons and (1995), Anglo-Saxon England, Christianization Scandinavia', the 24 213-49" of 59For the episode in context, see Halfrekr Saga, ed. by Bjarni Einarsson (Reykjavik: Stofnun Arna Magnssonar, 1977),pp" 45-5o. 60Den Norsk-Islndska Skaldediktningen,rev. by Kock, i, 86. Translation from E. O. G. TurvillePetre, Scaldic Poetry (Oxford: Oxford University Press, 1976), p. 72-.

195

This versemay be a later invention, rather than a genuine late-tenth-century composition, but it certainly representsa plausible response to the problems of the usersof a in mythologically-rich poetics social contextswhosemythological underpinningswere We be Eiriksmdl Hakonarmal sure, moreover,that shifting. may reasonably and are genuine half it is the the tenth compositions of second of century, and preciselythis sameproblem of is in tradition which reflected thesepoems. a recontextualisedpoetic This alteration to the context of skaldicverse,however,can alsobeenseenasfitting
influence, within a wider reshaping, under christian of the mythology surrounding skaldic Valh9ll late heathen The be verse. concept of can read as a responseto christian eschatology for Christianity believes in the worthy, as and concepts of the afterlife; an afterlife of pleasure factum heaven fights dragon: `et from in est a great well as an apocalypse which an army dracone draco Michael in et pugnabat et angeli et angeli ejus prceliabantur cum prcelium c tlo fought dragon Michael his ('and fought heaven; in battle and with the and angels ejus' was a 61 This framework is in fought dragon his them'). the eschatological paralleled angels and Valhpll, where the worthy enjoy pleasure before fighting eschatological beastssuch as Fenriswolf and the world-serpent at Ragnarok. This is not a novel position to take; Vigfusson and Powell, in an era eager to believe that Scandinavian mythology was untainted by christian influences, nevertheless accept that `in the Walhall system one can clearly see the have Later i62. ] Christian [... Northman the scholars also often religion of effect upon the Scandinavian Christianity in ValhQll as an example of syncretism with pointed out heathenism: It is clear that in many parts of Scandinavia there was a long period, in some in hundred years, which the old and new religions places as much as two by been have beliefs Pagan contact with affected must rituals and overlapped. Valhalla, first for likely, It is Christianity. example, that the concept of 63 influence. Christian in evidenced the mid-tenth century, was shaped under
My 61Biblia Sacra, ed. by Weber and others, ii, 1893(Apocalypse translation. 12:7). 62Corpus Poeticum Boreale, i, cv.

Reformation, From Conversion 63Birgit and Peter Sawyer, Medieval Scandinavia: to circa Boo-r5oo, Press, Minnesota University (Minneapolis: Series, 104" p" The Nordic 1993) of 17

196

At the sametime, one might equallyview the similarities betweenchristian and from heathenism skaldic eschatologyas resulting an early christian re-creation of as Christianity. Klaus possessing a mythological and cosmologicalschemewhich mirrors that of See, for has `das Sonatorrek instance, that von suggested zeigt verschiedene synkretistische Zge', but he also seesevidencein Sonatorrek for literary borrowing from Old English (1 lexis: `Die Wendung hefja "emporheben (ins Gtterreligious upp godheim), oder Gottesreich)" stammt vielleicht aus angelschsischer He Dichtung (vgl. Crist 61i: christlicher 64The hafen f6emum)'. interpretations is these two wies upp ambiguity of engla possible characteristicof the transitional quality of tenth-century skaldicverse,and the mythological Obinn. figure it of schemewhich erectsaround the

In considering how skaldic verse does, or does not, attempt to create a heathen Christianity, we are poetical and mythological response to the mythology and eschatology of brought back again to the question of Snorri's preservation and use of skaldic verse. In Gylfaginning, Snorri presents an overall mythological narrative which offers a heathen To large history a which appearsto combine christian and classicalelements. eschatological his Snorri's however, to rationalise material according to attempts this may reflect extent, Clearly, is familiar. he there the possibility that the relatively rich was models with which Obinn in in large part a response to the skaldic circles represents mythology associatedwith Snorri's bear in but heathenism, late Christianity mind that uses we must also on pressures of of skaldic verse often attribute more complex and mythologically-rich significances to verses Poetry Mead is Snorri's the treatment of of narrative than they actually originally possessed. one such case.

by Mittelalters, des Literatur von ed. skandinavischen Studien in (originally (P. (Heidelberg: Carl Winter, 1981),pp. 389-4OS 401) zur christlichen published (Gttingen: Palaestra, Lange, Wolfgang by Nordgermanen, Dichtung der zzz iooo-izoo, ed. Vandenhoeck & Ruprecht, i9S8)).

Aufitzezur " Klaus von See,`Christliche Skaldendichtung',in Edda, Saga,Skaldendichtung: 12 See, Skandinavistische Arbeiten, 6 Klaus

197

Another caseis his treatment of the roof of Va1hgll. In Gylfaginning,Snorri claims `at Valh9ll var skjpldum ppk i' ('that Valh911 wasroofed with shields'),on the basisof the following piece of skaldic verse:
baki ktu blikja, barrir vru grjti, Svfnis salnaefrar 65 hyggjandi. seggir

On their backsthey let shine - they were bombardedwith stones Svafnir's [Odin's] hall-shingles [shields], those sensiblemen.
Snorri attributes this verse to pjobolfr inn hvinverski, but scholars generally accept 66 it fact hornklofi's HaraldskvAi. Vries's in belongs porbjprn According list de that to to of 67We by-names, this is the only instance of Svafnir in skaldic verse. are therefore reliant on Snorri's word that Svafnir is indeed Obinn. It is certainly possible that Svafnir is a dwarf or Obinn. The by-name does implausible it be but it that should giant name, a not seem of however, for interest, is compound salncefrar of particular a close examination of this word Cleasby, have Snorri over-interpreted a skaldic source. suggests that may once again Vigfusson and Craigie gloss n&fr as `thebark of the birch, used for roofing'. Many of the bark for birch however, was used as a covering people as well suggest that compounds of n&fr, `a Craigie list in Vigfusson Cleasby, ncf -a-mahr, nickname of an outlaw clad and as roofs. ('bark [a: They list hildar fr] X68. [aefr]' `a the poetic phrase also neer of n n and ncefra-stka sleeve 69 birch bark This logical idea is `armour' battle' that the extension of a translation) my of = . This be idea body. human for be to to thought apply might the a covering might used as

65Snorri Sturluson: Edda: Prologue and `Gylfaginning; ed. by Anthony Faulkes (London: Viking Society for Northern Research, 1988;repr. woo), p. 7. Hereafter referenced as Prologue and Sturluson: Edda,trans. by Anthony Faulkes, Snorri from Translation Faulkes. by ed. 66Prologue and `Gylfaginning', ed. by Faulkes, p. 57; Den Norsk-Isldndska Skaldediktningen,rev. by Jnsson, by Skjaldedigtning, Norsk-Islandske Den (stanza Kock, i, is IA, 14. ed. ii);

EverymanClassics(London: Dent, 1987), p. 78.

'Gy faginning,

69Cleasby, Vigfusson and Craigie, sub near.

67 De Skaldenkenningen, pp. 36-5o. Dictionary, 2nd 68Richard Cleasby,Gudbrand Vigfusson and William A. Craigie,An Icelandic-English Press, University Oxford (Oxford: sub n&fr. 1957), edn

198

We interpret `Svfnis body `Svfnir's then, to might, armour. well as shields as salnxfrar' as hall-armour' or `Svafnir'shall-shields', or, more generally,as `Svfnir'shall-covering'. Whichever interpretation one follows, the overall significanceof the phrasemust remain `shields',or perhaps`armour'. The various possibleinterpretationsdo, however,suggestthat,
Obinn, Svfnir have had is in if Valh9ll the poet need not even mind a picture of roofed with `covering', it If interpret be hall is it is the need not shields. we n&fr as which covered; by hall. It is the the occupants of the covering worn possible that might perhaps make more ValhQll bearing as shields or wearing armour, than to see the senseto see the occupants of hall itself as doing so. This interpretation is supported by the claim in Skaldskaparmalthat `skjaldborgin er k9llub h9ll' ('a shield-wall is called a hall'), which suggeststhat a shield wall 70 HaraldskvX&. Snorri's depiction of Valaskj Jf also appearsto from is implied in the verse he has misinterpreted this verse: confirm that Obinn. rann gerbu Par er enn mikill stabr er Valaskjalf heitir. rann stab i pessum Pat hsxti H1ibskjlfin Par Jpk6u er sal, er skiru silfri, ok gubin ok 7' heim. hann p Ok p Alf9br heitir. ti of allan s& sitr 1Pvi sa: er sv There is also a great place called Valaskjlf. O inn owns that place. The gods Hlibskjalf, hall is in it it the made and roofed with pure silver, and there this linn] [i. in high-seat which is thus named. And when Alf9&r that e. sits he seesover the whole world. seat,

Valh9ll, but (`the be in Valaskjalf is form first The valr slain') as unlikely to element of the (`seat form *Valskjlf that an original of the evident similarity of the two namescould suggest be In Valaskjlf. it developed this case, would really to produce the slain'), a parasite-vowel `hall latter be `seat former Valh9ll; the slain'. the the of slain', of would the the samename as Obinn's for fact, in been have two alternativenames Snorri may, then, one awareof what are, he hall in skaldic verse;unsurprisingly, perhaps, misinterpreted them asreferring to two is Valaskjlf, however, for The suggesting not shields, silver distinct places. material roofing We in did moreover, least tradition. might suspect, skaldic appear these not that at one of

My Faulkes, 70Skaldskaparmal, by translation. 1,67. ed.

71Prologue and `Gylfaginning, ed. by Faulkes, p. zo. My translation.

199

Snorri has actually been influenced in his depiction of Valaskjlfby descriptions that of being from 72 built jewels. If this is the case,then it paradiseas various preciousmetalsand binn's seemsunlikely that either silver or shieldswere thought of asthe roofing material of hall in skaldic tradition.
It does seem possible, however, that the idea Valh9ll of arose originally as a kenning for a shield-wall. If Snorri is right that such a military formation `er k9llu h911', 73 it i could be hall Elaboration surely the called a of slain, or a slaughter-hall. and misunderstanding of this metaphor would then have causedthe shield-wall to become an afterlife destination, Obinn. This Obinn by have do idea presided over might, moreover, nothing to with an of as Obinn's do importance a war-god, and much to with the of poetry special preserve- as a immortalisation; have famous is be death, be means of to to an afterlife after one's and to famous after one's death requires that one dies heroically in battle, and that one has a poem (e.g. Eirzlzsmal, Hkonarma'l) composed in one's praise. Obinn, then, would be the crucial deity in insuring the afterlife of dead heroes, and particularly of those who died in battle; this have been literal metaphorical afterlife may then reimagined or misunderstood as a afterlife (under christian influence), and Obinn's association, in this capacity, with dead warriors could have led to his being presented as a war-god. This view is confirmed, moreover, by the fact that highly-localised (even familyNorse discussed frequently in As in destinations mythography. above, appear specific) afterlife for local both destinations, DuBois centralised and afterlife points out evidence section 1.9, but is not prepared to suggest that one type is more likely than the other to have been the 74 Nordland bolder line, localised Nordic in takes that a arguing religions. norm pre-christian destinations were originally more prevalent in Norway, reflecting a syncretistic religious in by Finnish Scandinavian this area, which was gradually eroded populations and practice of

72Compare Revelations,2i: io-2r. For a re-use of this material in a late medieval literary context see Pearl Manuscript: The Poems `Pearl, fitt in Pearl, in Jerusalem New description the 17, of the of the Ronald Malcolm Andrew Knight, by Green Gawain `Sir `Patience, `Cleanness, and ed. and the

Waldron (London: Arnold, 1978), pp. 100-0373Skldskaparmal, Faulkes, by 1,67. ed.

200

Obinnic 75 Nordland incoming While Valhpll. idea cult, which the seesas centering on the of is full Nordland's is he to evidence not really sufficient support the right weight of argument, to point out that a centralised afterlife destination or destinations need hardly have been the belief in Scandinavia; localised for usual or majority the evidence the more early medieval form of eternal rest is consonant with the model of pre-christian religion advancedin section least have been is indicate Valhpll development that the 2.3.1, and at may well sufficient to of literary a more-or-less process not necessarilyreflecting the mythological views of most heathen Scandinavians.

s.4

Asgard

Oinn Troy: Re-inventing and

Hero Classical as a

Saxo Grammaticus was a Dane of noble stock, apparently one of Bishop Absalon of 76 Roskilde's followers, who had been well-educated. Over a lengthy period spanning the turn history Gesta Danorum, huge his he of a composed of the twelfth and thirteenth centuries, Denmark and the Danes from the legendary past through to the present. Of particular interest here are the first nine books of the GestaDanorum, which deal with Denmark's Obinn in importance books These distant kings legendary skaldic of reflect the past. of the Obinn figure became for just the skaldic the of central mythological as milieux, memorialisation processObinn Where Danish kings. legendary helper deity skaldic the of or the tutelary as presents body functions presumably which was mythology to of achronous a through references verse incorporate instead Danorum Gesta to attempts the the skalds' audiences, available orally to The historical into entirely not always are results narrative. a continuous such mythology Obinn, in the in twice for appears instance, which the of of exile the case satisfactory, as, fit Saxo's below). mythological narratives -attempt to Gesta Danorum (see section 542, for intended court audienceswhich was Saxo's history so

66-99. Helgafell', 75 `Valhall Nordland, pp. and

74DuBois, pp. 74-81.

201

in which were, all probability, originally imaginedasoccurring outside human chronology into a historical narrative framework naturally inconsistencies, since creates problemsand historiography do not function in the sameways. mythology and To some extent Saxo can be seenasfollowing the historiographicaltrends of his
In Europe time. western the large-scale national history, working from the legendary origins of a nation through to the recent past, had become very popular and important (although not for the first time, as Paulus Diaconus's Historia Langobardorum, for instance, shows). Geoffrey of Monmouth's Historia Regum Britannie is perhaps the most famous example, but " To it is not the only one by any means. Indeed, following Saxo is this trend. as some extent discussed above, the Historia Regum Britannie shows considerable interest in the pre-christian deities of Britain, and Saxo can be seen as following this pattern. Saxo certainly does follow a for histories his in pattern similar to other such national creating nation an eastern founded Denmark (just Britain Trojans, its is by so owes nationhood mediterranean origin as Constantinople; different from king to the great attempt to provide the same sort of a Obinn Trojan for is Prose Edda, Prologue is in to the where a tradition origin made the Obinn, however, " Saxo's deities far beyond the use of non-christian goes use of suggested). Obinn. The GestaDanorum, in in the Historia Regum Britannnie, especially in the caseof Obinn, in its first nine books, as a hero. Although each of the fact, really seems to use Obinn,

Obinn books hero-figure Danish kings is a hero, who runs through these providesa unifying Of inn wasa human being, in a way that mortal kings cannot. Saxoclaims,of course,that Obinn implied following the usual euhemeristicreading of him. Yet he provides with an Obinn heathen Saxo's As hundreds lifespan of combinesthe mythology of years. we shall see, binn ideas Saxo and with classical about with a euhemeristicunderstandingof availableto Obinn hero, figure, is, Saxo's heroes. deities the behaviour moreover, composite and the of

Civitatibus, Duabus de Gesta Historia Freising's Otto the Blatt, for instance, suggests and of Saxo's (Blatt, background historiographical to work p. 13). Friderici, as contributing to this 78On the Trojan origin in the ProseEdda seeFaulkes, `Descent from the Gods', Pp. IIO-Z4. Prologue Faulkes, by `Gylfaginning, 4-6. pp. ed. and

76 Friis-Jensen, pp.13-18.

202

the first evidencefor, and perhapsthe sourceof, many of the peculiaritiesof Obinn in later Scandinaviantraditions. One of the most crucial elementsof Saxo'sdepiction O inn is that O inn often of in appears person to humans. The euhemeristicreadingof Wodan had first arisenin the Carolingian in English intellectual circles,and Paulus Diaconus, aswe eighth century and have seen,portrayed the supplicationsof the Vandalsand the Langobardsto Wodan and Frea
in terms of humans talking to other humans (see section 3.2, above). Since Paulus believed Wodan to have been a real human being distant he hardly have of the past, could portrayed Obinn in Likewise, Saxo human had been being this exchange any other way. reads as a who for `Tanto ('Men's ludificabantur ingenia' mistaken a god: quondam errore mortalium intelligence was once made ridiculous by gullibility of this kind'). 79We should, then, perhaps be Saxo Obinn dealing humans in that to, not surprised as appearing and with, portrays Saxo is, person; this moreover, a common classicalmotif, and one which, as we shall see, This Gesta however, from have known in Danorum, does the seems to motif classicaltexts. be in to not seem merely an euhemeristic rationalisation of mythological narratives which Obinn. In fact, Saxo beyond far mere rationalisation, of the sort goes mortals appeal to Obinn for Paulus Diaconus, he battlefields, in by portrays as appearing person on carried out for instance in his killing of Harald Hyldetan.

Obinn Danorum, is statedto havedisguisedhimself as Bruni, In Book 7 of the Gesta Bruni drowned in Hyldetan's, had Harald a river: after a confidant of
Eodem tempore Bruno quidam omnium Haraldi consiliorum unice particeps Huic Ringo, ipse habebatur. quocies secreciore nuncio opus et ac conscius habebant, mandata credere consueuerant. Quem familiaritatis gradum fuerat. Quo fluminis educacionis et crepundiorum consorcione adeptus Othinus, labores inter absumpto, eius assiduosprofeccionum cuiusdam aquis legacionis insidiose habitu cura arctissimam regum subornatus, nomine et fallacia fait, inimicicias labefactauit, tantaque ut amicicia serentis concordiam

(book Trbner, 1886), 79SaxonisGrammatici Gesta Danorum, ed. by Alfred Holder (Strassburg: p. 25 Hilda Ellis Davidson Danes, by History The Grammaticus: Saxo from Translation and ed. of the i). Totowa: Rowman Littlefield, Brewer; (Cambridge: Peter Fisher, by i98o), i, 26. and 2 vols trans. Hereafter referencedas The History of theDanes.

203

bello ac necessitudineuinctis mutuum odii rigorem ingeneraret,quod absque 80 saciarinon posseuideretur.

At that time there was a certain Bruni, Harald made the one close whom his If Ring he confidant of all plans. ever needed to exchange messagesof and a more secret nature, they would commit their instructions to this man. Bruni had gained this degree of intimacy becausethey had been brought up together from the time they had shaken the same rattles. However, amidst the toils of his constant journeyings he was drowned in the waters of a river. Odin, disguise deceitful his the assuming of name and clothing, carried out a he kings' bond embassywhereby close and sowed such strife undermined the through his deep artifices that between those who were joined in friendliness kinship he dislike, be it hardened and generated a which, seems, could not satisfied without warfare.
s

This disguise is portrayed as a longstanding arrangement, for Obinn determines the outcome Harald Ring, hostilities, in he between himself instigated, of the not which and an abstract Harald, Bruni: disguised Obinn, but by killing whilst as manner as physically At Bruno, nihil obsecrantis precibus motus, repente excussum curru regem detorsit, ipsius in in terram, ereptamque cadenti clauam caput arietauit 81 interfecit. proprioque eum Bestamine

But Bruni was completelyunmoved by the suppliant'sprayers;he suddenly jerked the king from the chariot, dashedhim to the ground, snatchedhis mace head, him his his dispatched he fell it with own weapon. as and, whirling at in battle in As in Saxo'sversion of Bjarkaml (on which seebelow), Obinn actually appears feature him is important hero of the narrative: an recognising person, and a Ad quod silente Brunone, subiit regem, Othynum hunc esse,olimque familiare sibi numen impresenciarumdandeuel subtrabendeopis gracia 82 habitu tegi. uersiformi corporis When Bruni stayedsilent, it entered the king's mind that here was Odin, a deity once his friend and now disguisedunder this changeof shapein order to his help. grant or withhold
80GestaDanorum, ed. by Holder, p. ZSS. Translation from The History of the Danes, 1,232', `habitu' in `appearance'. be this translated context, as could more plausibly, 81GestaDanorum, ed. by Holder, p. 263 (book 8). Translation from The History of the Danes, i, 24382GestaDanorum, ed. by Holder, p. 263 (book 8). Translation from The History of the Danes, 1,2.43.

204

This topos of recognisinga he human interferes in god when or she physically affairs Obinn. is common enough in later Scandinavian In literature, particularly with regard to Obinn in Eddas, hero disguise in sagas, as the prose and verse to test a or to often appears herald his death. It doesnot, however,play any part in the skaldic verse,although Obinn is frequently associated battle The in lacking is in the earlier with skaldic verse. motif also for Wodan. This for have look in evidence that we to the genesisof this motif suggests Scandinaviantradition only shortly before Saxo'stime; in fact, it is possible perhapseven likely - that the motif originateswith Saxoor within his immediate circle. This motif may depend in well upon a mediterraneanpaganmotif, which also occurssometimes early Actus Apostolorum, Paul Barnabas (section As in christian contexts. noted above 3.2), and are
for Zeus Hermes `turbae in mistaken and appearing to men physically and person: autem cum fecerat Paulus levaverunt lycaonice dicentes dii facti vidissent quod vocem suam similes hominibus descenderunt ad nos' ('but when the crowds had seen what Paulus had done they Lycaonian, "gods disguised have in down their to visit raised up voice, crying as mortals come 83This Frequently be narrative can seen as a reflex of a motif of considerable antiquity. us"'). in classical literature, even as far back as Homer, the gods appear to men in human form. These appearancesoccur in a variety of contexts. The feature of disguise, which is often an The in Scandinavian is the classicalmotif. also common appearanceof element of the motif, but in deity battle in less is still occurs on a number of occasions classical a common, Likewise, deity in Scandinavian in the tradition as tradition. appearanceof a classical for in disguise is deity the purpose of often appears particularly when the Obinn's Scandinavian literature. in individual, just in to testing an mortals visits as literature -

A particularly interesting exampleof

Obinn's

battlefield the personalappearance on

Biarco Saxo dialogue between Gesta Danorum, in book and rendersa where z of the occurs Ruta in verse,giving the following rationale for the decision to useverse:

83Biblia Sacra,ed. by Weber and others, ii, 1722(Actus My For Apostolorum, translation. 14: 10). a Robin Lane Fox, World Christians Mediterranean Pagans in discussionof this passage the see and CenturyAD to the Conversion Constantine (London: Penguin, from the Second 1988), of pp. 99-101.

205

Hanc maxime exhortacionum seriem idcirco metrica racione compegerim, Danici intellectus cuiusdamcarminis quod earundemsentenciarum

T digestus compendio a compluribus antiquitatis peritis memoriter usurpatur

I have particularly composedthis set of admonitory speeches in metre because Danish the samethoughts, arrangedwithin the compass poem, are of a deeds. from by recited memory many who are conversant with ancient It is not entirely clear that `Danici' here implies that this versewas in the Old Danish 85What however, is language,rather than in another Scandinavian dialect. seemsquite clear, Our has he Saxo for is translated. the versewhich concern that claiming an oral tradition here is particularly with the following lines, near the end of the versepassage:
Et nunc ille ubi sit, qui uulgo dicitur Othin Armipotens, uno semper contentus ocello? Dic michi, Ruta, precor, usquam si conspicis ilium?

[Bianco: ]

Ad hec Ruta:
Adde oculum propius et nostras prospice chelas, Ante sacraturus uictrici lumina signo, Si uis presentem tuto cognoscere Martem.

Tum Biarco: Si potero horrendum Frigge spectare maritum, Quantumcunque albo clypeosit tectus, et altem Flectat equum, Lethra nequaquam abibit; sospes 86 diuum. Fas est belligerum bello prosternere [Biarki: ] But now, where is the one whom the peoplecall Odin, in powerful arms, content with a single eye? Tell me, Ruta, is there anywhereyou can spy him?
Ruta replied:

Bring your gazenearerand look through my arm akimbo. You must first hallow your eyeswith the sign of victory

GestaDanorum, ed. by Holder, p. 67. Translation from The History of theDanes,i, 63. 85See Friis-Jensen, Saxo Grammaticus asLatin Poet, p. 2I. 86GestaDanorum, ed. by Holder, p. 66. Translation from The History of the Danes, [, 63.

206

to recognisethe war-god safelyface-to-face. Then Biarki: If I should set eyeson the fearsomehusbandof Frigg, though he is protected by his white shield, and manoeuvres his tall horse, he shall not go unhurt from Leire; it is right to lay low the warrior god in battle. Obinn This is an extremely odd passage. Saxoequates Mars, in is line with which very much Obinn his belligerent king former Insofar Saxo with generalpresentation of as a times. of as Obinn deity, he indeed, him portrays as a portrays principally asassociated with war; the
Obinn in does Likewise, association with poetry which characterises skaldic verse not appear. Obinn's association with wisdom, which seemsvery important in Eddaic verse, as well as in later Scandinavian evidence, does not appear in the GestaDanorum, except, perhaps, in Obinn's ingenious 87 himself. animation of a statue of

The oddities do not end there, however.Saxoincludes a peculiar magicalrigmarole

involve to which seems viewing Obinn through someone else'scrooked arms. Perhaps the Obinn. The Biarco's is oddest element of this passage obscure vow concerning meaning of this vow is not entirely clear, but it seemslikely on syntactical and contextual grounds (and basis Bjarkamdl below) is the the to the tradition; that the of see on vow a other witnesses Why battle Obinn. be `fas' in is destroy It this should threat to attempt to not at all clear. binn Even destroy deity. Saxo's be human able to a can also seems unlikely that one would hardly have been vulnerable in battle, since he is capable of favouring Harald Hyldetan `ut integritatis eius habitus ferro quassarinon posset' (`[so] that no sword could impair his " safety').

The difficulties with which one is facedin dealing with this passage by are augmented

This Saxo's Latin It is is the nature of the text. text version of an earlier vernaculartext. not Saxo's how but Saxo is, the vernaculartext much older than own statement of the clear his difficult It is to tell, then, source attributes considerable ageto that source. provenanceof

87Book i; on this episodeseeShaw, `Miracle asMagic'. 88GestaDanorum, ed. by Holder, p. z47 (book A). Translation in The History of theDanes,1, zzs.

207

how long it exactlywhen the sourcewas composed,and remainedrelevantand in It is how it text grounded well-known mythology. comprehensibleasa well alsounclear final in its Saxo. We transmitted, transmissionto particularly was preserved and cannot, Saxo What be be is that text the understood perfectly. moreover, sure we can sure of that 89 freedom, and a strong tendencyto classicize. Saxo translated his sourcewith considerable Obinn, Possessing Scandinavian is only one eye a common characteristicof appearing
in early skaldic verse (for instance Porbj9rn hornklofi's late ninth-century Haraldskveei refers Frigg's him `eineygja / Friggjar fabmbyggvi', `the to embrace')9and as one-eyed sharer of Saxo be This Norse literature from thought to confirm that throughout that time on. might had a genuine Norse poetic source, although this detail appearsto be very widespread and Saxo independently known Thus have been it to of the presumed source may very common. for his Bjarkamal. Wherever he gained the detail, however, Saxo stamps it with his classical Obinn, `uno semper contentus ocello' ('always learning, wittily connecting the rascally Martial `uno described by dinner-guest larcenous as satisfied with a single eye'), with the 91 it At ('satisfied lumine' time, the seemspossible that same with a single eye'). contentum Saxo who translates Obinn influenced dialogue Mars in the quoted above -- was also as

by two references to `caecoMarte' ('blind Mars') in Virgil's Aeneid; it is just possible, in fact, that Obinn's least Danorum Gesta in impairment partially, on classical the rests, at visual

92 literature.

Edda fragment Laufdss in lausavisa Ranisch identify Heusler and as a which occurs a O 13 This is inn. Biarco's Bjarkamdl Norse threat stanza against to corresponds which of a

9' Den Norsk-Islndska Skaldediktningen,rev. by Kock, t, iS (stanza iz). My translation. 91M. Val. Martialis Epigrammata, ed. by W. M. Lindsay, Scriptorum Classicorum Bibliotheca My line book Press, 8, Clarendon (Oxford: Oxoniensis, 2nd edn i. no. 59, 192,9;repr. 1959), translation. 92P. Vergili Maronis Opera, ed. by R. A. B. Mynors, Scriptorum Classicorum Bibliotheca Oxoniensis My (book line line (book Press, Clarendon translation. (Oxford: 518). 9, 335),322 Z, 1969), pp. 137 Art aus den Fornaldarsgur und anderenProsawerken,ed. by 93Eddica Minora: Dichtungen eddischer Andreas Heusler and Wilhelm Ranisch (Darmstadt: Wissenschaftliche Buchgesellschaft, 1974

Poet, Latin 89SeeFriis-Jensen,Saxo Grammaticus 83-98pp. as

208

94 Biarco's Hrolfi in kraka, threat which appears very similar to the version of saga suggesting that the identification of this stanza as part of a Bjarkaml is reasonably convincing. Both Obinn. Biarco these versions appear to share a common tradition of threatening to strangle Perhaps Saxo suppressedthis image as undignified; Friis-Jensen's argument that Saxo was 95 Bjarkamal Latin Even if to the this attempting present as a epic might support suggestion. Saxo's source had a different form of the threat from that which is preservedin the later Norse texts, however, it seems likely that Saxo understood and intended a threat.

The Norse versionsof this scenedo not, however,preserve for Ruta's any equivalent
looking her in the technique through role and of crooked arms order to see Obinn safely.

The fragments contained in the prose Edda do not include anything corresponding to this (although, of course, the original poem could have done) while Hrolfi sagakraka has nothing Venus based in Ruta's here is Friis-Jensen role upon the role of comparable. argues that Book 2 of the Aeneid, when she revealsto Aeneas that severalOlympian deities are helping to destroy Troy. 96This argument is plausible, and again suggeststhat Saxo's source may have been considerably closer to the Norse Bjarkamal fragment than might at first appear. What is not clear, however, is how Saxo arrived at Ruta's curious instruction to

lumina Biarco: `Adde oculum propius et nostrasprospicechelas,/ Ante sacraturus uictrici Obinn, Obinn, is it This or making safeto see signo'. apparentlymagicalmethod of seeing Aeneid in Venus's to the the which the of passage gods revealing of method entirely unlike 97No obviousparallelssuggestthemselves. We may suspectthat Saxo's Friis-Jensen refers. fi kraka Hrd by from differed the and tradition saga the represented source considerably Bjarkamdl fragment, but if we acceptFriis-Jensen'sargument that Saxowas deliberately Saxo himself The likely. less Vergilian his that evidence this seems epic, a source as recasting by his detail is inventor the word of rare use reinforced perhaps this narrative the of was

[facsimile reprint of first edition, Dortmund: Franz Wilhelm Ruhfus, 19031),p. 32; on the manuscript for the stanza, see p. xxiv. witnesses

Latin Poet, 94SeeFriis-Jensen, Saxo Grammaticus p. 97. as

95See Friis-Jensen, Saxo Grammaticus asLatin Poet, pp. 7I-IOI. 96See Friis-Jensen, Saxo Grammaticus asLatin Poet, pp. 97-98.

209

for have Various been Elton Powell Vergil's this word suggested. sources and chelae. suggest Scorpion Georgics, book i, line 33`wherethe chelae Zodiac, in the are the clawsof the and are Friis-Jensen suggests `that Saxo,deliberatelyor in the next line called brachiai98.
(Greek) for has the arms at v. z8i, chelae,on chosen the extremely rare word subconsciously, Chelae for Libra, his interest imagery: is in the constellation a name and account of weighing Thesaurus Latinae Linguae `the balance'; instance the only of means according to arms of the Lewis Capella'99. According Martianus `human in is in the meaning to and a poem arms' Short, chelaedenotes not the arms of the balance, but the arms of the constellation Scorpio; 10 Friisfor Libra. Libra, be into the word can used metonymically since these extend Jensen's weighing imagery might still lie behind Saxo's use of chelae,since the word can refer Scorpio, (and just Libra, The the the claws) of not to the scales. recognition that the arms Saxo however, by indicated also used the word suggeststhat chelae, scorpion, are the arms Ruta's in We interpret indicate the use of chelae speech as a may crooked arms. specifically to image her holding how is indicating the the scorpion of arms; she cue, precisely visual her head, inwards holding be is them curved as on either side of pictured suggests that she to heighten flavour The does. that to the the sense also serves word of astrological scorpion a Obinn. Obinn, he Biarco is whilst views or to protect she using a magical method to reveal

That Saxo has createdhis own idiosyncratic-

Latinate highly and -

version of the

The Biarco's `fas' in his by is threat. use of the word traditional scene also reinforced based Biarco's do Bjarkaml threat sense of any as upon portray to not witnesses vernacular Obinn's has forces. Saxo for for but the chosen opposing support threat of revenge a as right,

97See Friis-Jensen, Saxo Grammaticus asLatin Poet, p. 97.

Oliver Elton by 98The First Nine Booksof theDanishHistory of SaxoGrammaticus, trans. and with an (note Nutt, David (London: 8o Powell York i). introduction by Frederick 1894),p.
99Friis-Jensen, Saxo Grammaticus as Latin Poet, p. 213,note 28o. Of course, Martianus Capella's Blatt, (see, for instance, in importance p. renaissance the twelfth-century work was of considerable The Way in Inge Skovgaard-Petersen Saxo, to known out points to as i9), and was clearly Grammaticus, Saxo Denmark', in History Saxo's Books Three First Byzantium: A Study in the of of interest identify be Friis-Jensen (p. to Friis-Jensen, an by nevertheless, right, may I29). 121-33 pp. ed. in weighing imagery in this line.

210

`fas ', a word redolent of classical Roman attitudes about what is fitting and right, to translate, deals in all probability, a passage imperative In Saxo the which with of revenge. this, may his loyalty lord time; to one's taken to the extreme of dying with him is express an attitude of Saxo Bjarkamal, lord the theme of the to that the and adds extreme attempt to avengeone's Saxo by kill deity. does however, imperatives to even these attempting a not, present within framework loyalty between lord he the of an exchange of gifts and and retainers, rather imperatives into duties In doing he these the elevates grand moral of the classicalworld. so, Obinn in Mars, in the the also re-imagines mould of who appears person on classical battlefields, and can even occasionally be wounded by mortal warriors. As noted above, Saxo Obinn Mars, despite the well-evidenced tendencies of previous classicising authors even calls Obinn Wodan (see Mercury to equate sections 3.2,3.4 and 4.2). or with It is, of course, impossible to date the composition of Bjarkamal, since the poem does Norse Saxo's few based is hypothetical it translation, a pieces of a upon a poem not exist; Hrolfs in Edda, in saga and the version of the story preserved version preserved the prose We have seen that the Norse versions of the story seem to belong to a single kraka. 101 Saxo. If focus by from differ that recorded we on attempting tradition, which may somewhat have Bjarkaml Saxo's date predates all the extant versions, we to source, since this version of This Gesta Danorum. dating in version of of the composition of the a terminus ante quem our Bjarkaml cannot have been composed later than the early thirteenth century. Our only (see Saxo's however, is dating, for the of poem's age statement more precise a evidence

Saxo's version of the poem. above),and the contents of Both these forms of evidencehavetheir problems.Even if we acceptSaxo'sindication form in it is that reachedthe which age,we neednot suppose that the poem of considerable Saxo knew it until very recently, if it was orally transmitted. We havealso seenthat there is a Some have, free his Saxo nevertheless, scholars material. with wasvery strong probability that Woolf has Rosemary basis its Saxo's date that argued the content. of to on source attempted
University Press, (Oxford: Oxford Dictionary Latin 100 Short, A Charles T. Lewis See Charlton and 1879;repr. 1933),sub chele.

211

dying lord Bjarkamal in highly is distinctive, the specific motif of men their with and shared 102 Maldon. On English Battle Saxo's Old basis, be this with the of source might supposed to date to before the composition of the Battle of Maldon, in the tenth century. It is not clear from this supposed parallel, however, which poem should be considered the earlier (even if Woolf dying ideal the considers that one parallel sufficiently convincing). the argues of men lord Germanic heroic literature. If their with was not a commonplace of early this is the case, Battle Maldon perhaps we should see the of -a heroic very non-traditional poem as the

in Bjarkamal, innovation is introduced; being the poem which a version of a story which likely have had long have introduced innovation. is to tradition, to seems such an a unlikely Frank's response to Woolfs article reads the Battle of Maldon in just this way, as an innovative poem which `peers,not backwards through the mists to Germania, but just around Europe in which the profession of warrior was a way of the corner, to an eleventh-century achieving religious perfection and a martyr's crown'103

Klaus See for Saxo's date While Woolf hassuggested von source, a relatively early
In `Hskarla his dated be to the twelfth century. article argues that that the source should hvpt. Nochmals zum Alter der Bjarkamal' he advancesarguments for his dating based upon 104 based hvpt. Such Hskarla arguments upon a the contents of the poem and upon the title date later have been are not particularly persuasive. title which may added to the poem at a Von See's arguments based upon the poem's contents are, however, potentially more Obinn's battlefield is known He the not on appearance that personal out points convincing. 105 in other sources before the thirteenth century. The dialogue form of the poem he also 106 On heroic identifies as typical of younger verse, rather than earlier examples of the genre.

Nouvelle Vague?, Anachronism Lord: in Dying 103 Men Ideal The Frank, Roberta or with their of (p. Lund, Wood by Places, People io6). pp. 95-io6 and ed. and Norroenum,ed. by Klaus von See, Huskarla hvpt.Nochmals zum Alter der Bjarkaml',in Speculum debate Frank Roberta the Dronke and others, pp. 421-31. on this point, providesa useful summaryof Lord', With Dying Men Ideal `The in See's their pp. 97-98" of view, von and upholds '05See,`Hskarlahvpt', p. 423.
106 See, `Hskarla hupt', pp. 423-24.

11See Fornaldar Sogur, ed. by Jnsson, 1,98-104 (chapters 5o-5z). 12 Woolf, 'The Ideal of Men Dying with their Lord', 63-81.

212

balance,we should probably seeSaxo'ssourceas a product of a christian court concernedwith Obinn idea historical identity is which uses as a symbol of a past which an of national heathen and yet noble, and, crucially, capableof providing exemplafor the conduct of Saxo These but he his the retainers. aswell, concernsare very much concernsof modifies Obinn in be he ideals, to portrait of classical attempting to with what perceives accordance for Scandinavian Danes based linked in is traditions, the createa past and which at once upon to the Roman empire. Von See'spoint that Obinn is not portrayedin other sourcesbefore the thirteenth lead battlefield in us to a wider conclusion century asappearing person on the can perhaps
Gesta Bjarkamal did than merely that the not significantly predate the composition of the Danorum. If we accept that this is a viable dating criterion be it to so - then and seems Obinn appearing the Bjarkaml is almost certainly the earliest extant example of the motif of Obinn battlefield in in person on battlefields. Since upon a goes on to appear person on a likely Danorum, it Gesta in that we are seeing the seems number of other occasions the Danorum. Gesta Bjarkamal in development the and of this motif earliest stages of the Although the motif seems to have appearedfirst in the Bjarkamal, its reminiscence of in frequently deities literature, in appear person, particularly on which pagan classical battlefields (for example, in the Iliad and the Aeneid), would no doubt have appealedto Saxo, Obinn Saxo's description More Latin learning. his formidable as of specifically, with Mars, described Aeneid in as armipotens, also when armipotens calls to mind an occasion the Mars `Hic battlefield: be animum armipotens the on appearance as read an can makes what ('Here Mars, in / Latinis powerful arms, addidit et stimulos acris sub pectore uertit' uirisque battle-lines beneath Latins force the spurred-on turned around the and to added spirit and 107 his chest'). It is not surprising, then, that the motif should go on to appear, now in a more for in book instance, Danorum, Gesta 8, in form, when as, the elsewhere classical clearly Obinn, disguised Bruni, kills Harald Hyldetan in person (see above). It is possible that as his lifetime in during developing Saxo is simply reflecting a motif which was vernacular oral

213

literature concernedwith the legendarypast. Whether classical have inspired models may literature hard determine, is but it is certainly possiblethat they did. It does to such oral likely, Saxo's however, fondness for fact Bjarkamdl in seem the that given the motif, was innovative in Gesta it Danorum, essentially this and unusual using motif, and that wasthe as Obinn Bjarkamal, idea in much asthe to which served popularisethis of appearing person, became feature later Icelandic The idea in the which so much a of mythography. use of this Vplsunga for GrImnismdl in, instance, Reginsmdl Poetic Edda, in the saga,aswell as and must 108 have important dissemination in its in Iceland. also playedan role The classicisingtreatment of Obinn in the Gesta Danorum is also evidenced in his
formation, he Danish in kings teachesto certain association with the wedge troop a which heroes deities by in manner perhaps not entirely unlike the provision of guidance to classical literature. One might, however, also note that the Langobard ethnogenesis presents Frea as Winnili advising the be involving to adopt a military tactic which could a troop seen as

formation (seesection 3.6.z, above).The wedgeformation motif found in the Gesta Obinn Danorum, however, probably reflectsmore Saxo'spresentationof asa mediterranean Danorum, figure with traits, as a deity, in which he parallelsMars. In book i of the Gesta Hading is hailed, whilst sailing past, by an old man, who giveshim adviceon how to draw up his troops for battle: Noruagiam preteriret, animaduertit Quem excepturusHadingus, dum classe in littore senemcrebro amiculi motu appellendinauigii monitus afferentem. diuerticulum Quem, repugnantibussociis, damnosumque profeccionis habuit, in affirmantibus, naue susceptumcenturiandi exercitusauctorem dyadem ordinanda agminum racione curiosius attenderesolitum, ut prima per phalanx, ac per tetradem secundaconstaret,tercia uero octoadisadieccione

207 My translation. Mynors, p. 32-9(book 9, lines 717-18). 108 Edda: Die Lieder des Codex Regius nebstverwandtenDenkmlern, ed. by Gustav Neckel and Hans Kuhn, Germanische Bibliothek, vierte Reihe, Texte, 3rd edn, 2 vols (Heidelberg: Carl Winter, 1962for See, Die instance, (Reginsmal; (Grimnisml) p. 178). see especially 172-'19 and c, 58-68 1968), (Berlin: Mayer Mller, & Ranisch, Wilhelm Text, by Bugges Vplsungasaga i9o8), pp. znd edn ed. nach z6-ZS.

214

insequens duplicitatis succresceret, priorem semperque augmento '9 transcenderet. While coastingNorway with his fleet in an effort to intercept him, Hading noticed an old man on the shore waving his mantle to and fro to indicate that he wished him to put in to land. Though his fellow-sailorsgrumbled that this deviation from their route would be disastrous,he took him aboardand found in him the man to supervise disposition his He had the troops. this careful of for the arrangementof his columns: in the first row he would put two system four in men, the second,then increasethe third to eight, and step up each by doubling front. in succeedingrank the numbers of the one As Fisher and Davidson note, this incident is similar in a number of details to an incident in 110 Reginsmdl. In Reginsmdl himself Hnikarr stanzas i6-zy, a man who calls asksto come Sigur&r Fafnisbani. Once aboardthe ship, he tells Sigur&r about several aboard the ship of battle omens, and then instructs him asfollows:
J eir sigr hafa, higrleics hvatir, kunno, er si 111 fylkia. hamalt eba

those who know how to look properly get the victory, those who urge swordhow know draw to play and up an army or a wedge-shaped column.

'12 Saga. The samenarrative appears It seemsthat `hamalt fylkia' is in chapter 17of Vplsunga formation it formation, is or not remains open a wedge some sort of troop although whether Obinn Obinn, intended in Gesta Hnikarr is by-name is the to question. and that a of Obinn Hyldetan Harald in book Danorum is confirmed by the fact that the sameadvice gives Hading however, is interesting, Gesta The Danorum. particularly 7 of the episodewith Saga (which Vplsunga becauseof the similarities with Reginsml probably made use of and Reginsml).It is not possibleto saywhich of the two versionsof the narrativehas priority, Saxo's is in Reginsmal date narrative simpler than those with any certainty. since we cannot

19 GestaDanorum, ed. by Holder, p. 32.Translation from The History of theDanes,i, 31. "o The History of theDanes,ii, 36 (note 6i). Kuhn, Translation The Poetic Neckel from 11 Edda: Die Lieder desCodexRegius, by 1,179. and ed. Edda, trans. by Carolyne Larrington (Oxford: Oxford University Press,1996;repr. 1999),p. ISS.
l"' Die Vplsungasaga, Ranisch, by 6-z8. z, pp. ed.

215

Old Norse texts, but this could be either because he givesa simpler, earlier version of the the because felt he The the need to simplify an unnecessarily narrative, or complicatedstory. latter perhapsseemslesslikely; Saxoaddsto his sourcesfreely, and doesnot seemto shrink from complicated narratives.Perhaps,then, we should assignpriority to Saxo.Against this, however, one might note that Saxo'sdescription of Hading's battle with the Biarmians (which emergesabruptly from the instructions on drawing troops), up perhapsechoesthe Reginsml board version of the narrative, in which Hnikarr's presence on causes a storm to In Saxo's battle Biarmians, Hading abate. with the they attack using a rainstorm, which the his The is for but it old man repelswith one of own. similarity not sufficient a strong case, difficulties involved in attempting a history of the motif. the emphasises
Despite these difficulties, we should probably see the motifs as related. There seems no reason to place the origins of the motif of Obinn advising the use of a wedge-shaped

battle formation long before Saxo's time. The motif does not appearoutside Scandinavia, and formation be in do the the they while names of to may appear skaldic verse, not seem binn (beyond battle, the associated particularly with natural association of and therefore battle formations, with Obinn). While it is not possible to determine whence Saxo received this motif, the fact that

Saxo he re-used it in book 7 suggests he important. deals it that considered with a number of Obinn, Starkabr Hading heroeswho are personallysponsoredby and, of course, such as and Harald Hyldetan. Hading and Harald Hyldetan are Danish royalty, and, this being the case, Saxo'sinterest in their military tactics need not surpriseus. As the son of a courtier, with a Danish Saxo doubt for Danish was no concernedthat the crown should concern royal power, "' his in historiography. That be militarily effective,and naturally he reflectsthis concern binn his does importance knowledge not merely reflect as a of tactics should provide the Obinn Saxo follow Saxo's history, in it places within a source; and need not simply character

113 On Saxo's position with regard to the Danish monarchy, seeFriis-Jensen, who sums it up as follows: `The discontinuance of the family tradition, Saxo argues, is not a breach of loyalty towards intellectual his for king i. his fighting by he because King, with weapons, e. writing the continues (p. history' i3). patriotic

216

the context of Byzantium, the capital of the Roman Empire in Saxo's day, and just as he Romanises his poetic into Vergilian (see he Romanises sources miniaturised epics above), so Obinn. Hilda Ellis Davidson Peter Fisher `Saxo's interest in and that the wedge suggest formation may have been by fact forces Aeneas strengthened the that the of adopt this way of fighting in book i2 of the Aeneid X114. For Saxo, then, the wedge formation was a Roman formation; although he may have had Scandinavian Obinn for the narrative of a source teaching the formation to Hading, he certainly would have considered the formation to have been a Roman invention. The importance which Saxo is, to then, a attaches this motif Obinn Roman for formation links battle reflection of a conception of a as a conduit which Danish history into Roman civilisation.

O inn's By-Names in Gesta Danorum 5.4"I the

Obinn The Gesta Danorum refers to by two of his by-names,Hroptr (Rostarus/Rosterus)and Yggr (Uggerus).115 Both of these by-namesare well-attested in early known Saxo. What be harder is to to skaldic verse,and we need not surprisedthat they were he Saxo however, did because is by-names; they were establish, use them why usedthese
he Or for in did he the the them? to already attached episodes which uses source narratives because he by-names himself, wanted to create some particular effect, or choose to use the Hroptr, In is in the there caseof no reason to provide some special resonance, these episodes? in by-name the episodes which the name was specifically connected with suppose that the Saxo however, had It is Danorum. Gesta in that poetic sources of course possible, appears the for these episodes, in which casethe versesmight well use Hroptr simply as a variation for Obinn, by-name in that particular. without any special significance attaching to the use of

15 GestaDanorum, ed. by Holder, pp. 79 (book 3; Rosterus), 304 (book 9; Rostarus) and 158(book S; Uggerus).

The History of theDanes,11,37.

217

In the caseof Yggr, nevertheless, it is possible that the name relates to his function in book S of the GestaDanorum as a prophet. Yggr appearsin the name Yggdrasill, which `steed 116 Yggr' it in Snorri's least, means of although refers, to the world-tree. mythography at In Eddaic mythography in Obinn's Yggdrasill be general, connected with appearsto pursuit of wisdom, which might imply a connection between the by-name Yggr and the pursuit of Obinn's That Viking idea in Age wisdom and prophecy. the the of steed existed already likely; Viking Age picture stones such as that from Tjngvide (Gotland) depict seems quite horses Sleipnir, be be Turvilleeight-legged which can probably understood to although 117 Petre has wisely urged caution in making The interpretation. connection of the such an Obinn's future, knowledge steed with pursuit of wisdom and and the equation or of the association of the steed with the world-tree, cannot be proved to have existed at this date, however. Obinn's Some scholars would argue that the complex of narratives involving steed or the world-tree (or both) represents the shamanic journey, whose purpose may be to achieve Obinn's i 18 It however, difficult journeys in is this, to prove wisdom or prophecy. and very journeys in bear in to search of wisdom any caseoften considerable resemblances similar O inn's Baldrs dead literary Thus in traditions. mediterranean ride to consult a seeress Draumar is very similar to journeys to Hades to consult the dead which occur in classical

116 Prologue and `Gylfaginning', ed. by Faulkes, p. i7. 11'Turville-Petre, Myth and Religion of the North, p. 57. For a photograph of the Tjngvide stone, see Corpus of Anglo-Saxon Stone Sculpture, II: Cumberland, Westmorland and LancashireNorth-of-theSands, ed. by Richard N. Bailey and Rosemary Cramp (Oxford: Oxford University Press, 1988), illustration 693. The present author has argued elsewherethat the motif of the eight-legged horse Obinn/Woden English in folklore from have in early medieval existed contexts quite separate may ('Spiders, Snakes and Shamans: The Eight-Legged Horse in Anglo-Saxon Magic', conference paper Magic and the Natural World in WesternIconographyand Literature, delivered at Saints and Serpents: Centre for Medieval Studies, University of Leeds, z6 June zooz). Obinn have also 118 Other elements of the mythology surrounding See, for instance, Motz, pp. 82--87. been argued to indicate shamanic practices, as in Strm's claim that the behaviour of the berserker is Erzbischof: Bedeutung `Berserker V. Strm, (Ake due to a state of shamanic ecstasy und und Ecstasy: Based Religious Papers Read in Berserkerbegriffes', Entwicklung des altnordischen on at the Abo, August by. Nils Finland, G. Held Ecstasy Religious the Symposiumon z6th-z8th of rg8r, ed. on at Holm (Stockholm: Almqvist & Wiksell, 1982),pp. 178-85).

218

literature from the Odyssey through to the Aeneid; an example which Saxo probably knew is Aeneas's reception of wisdom from Anchises in book 6 the Aeneid. One in this might note of Saxo includes, in book 8 of the GestaDanorum, some journeys to connection that Biarmaland which have been thought to have been Vergil's description in parts on modelled Obinn's 119 Aeneas's journey in Vafprubnir, described Likewise, of to the underworld. visit to Vafpru6nisml, in order to have a wisdom contest, bears some resemblance to the Solomon 120 Saturn Adrian These Ritheus English literature. Old in traditions and and traditions and had mediterranean roots, but they have reflexes in Scandinavia as well as in England. Saxo in stands an excellent position, at the end of the twelfth-century renaissance,to act as a Saxo for Moreover, Scandinavian discourse. literary into conduit mediterranean motifs himself is known to have had a thorough classicaleducation, by the standards of his day (which were high), and he can often be demonstrated to be classicising his material (see 121 above). Obinn Vecha book he in In addition to Yggr and Hroptr, 3, when uses the name disguises himself as a female physician in order to rape Rinda. Vecha is otherwise unknown as O Peter Fisher, Davidson `Vecha is Hilda Ellis however. According by-name inn, to and a of from ON vitki (prophet or wizard)"". This etymology should be considered as tentative. It O inn, however, and this fact - coupled with the lack of fits well with Saxo's conception of Vecha for by-name a any other evidence Saxo have invented the name. may suggeststhat

O Eddaic (but in inn for be The name would perhaps appropriate mYtho ' aPhY as presented

19 See History of the Danes, II, 142. Hading's journey to the edge of the underworld in book one of the Laugesen Teilgaard in discussion Anker by Aeneid been linked GestaDanorum has also a of with the Saxostudier, Placering', in Genremassige Danorums `Gesta pp. 20-27 Inge Skovgaard-Petersen, (discussion pp. 27-29; see p. 28). Ritbeus', by James `Adrian E. Saturn' Solomon `Prose 120 The On these traditions see and and ed. and (Toronto: Toronto Texts, Studies English Old Hill, McMaster and I Cross and Thomas D. Saturn, by Robert J. Menner, Solomon Dialogues Poetical The ed. and University Press, 1982) and of (New York: Series, The Modern Monograph America Association 13 The Modern Language of Press, University in Oxford London: America; 1941);note particular the Language Association of Menner, by Vafru' 58,65. pp. noted nismdl parallels with

12'On the high level of Danish Latinity, seeFriis-Jensen,p. 14. Danes, II, 57. 122 The History of the

219

certainly not ashe is presentedin skaldic verse),and could therefore have existedbefore Saxo it. The Gesta Danorum, however,often constitutes the earliestevidencefor traits and used binn This Eddaic in motifs which are otherwiseonly associated would, mythography. with Eddaic traditions. then, suggestthat Saxowas pivotal in the creation of several

Obinn's Exiles in the GestaDanorum 5.4.2

Obinn is twice described as going into exile in the GestaDanorum. In book one he into his following Frigg's her infidelity goes voluntary exile statue and attempts to recycle into jewellery. 123 In book three he is forced into exile by the other gods becauseof their disapproval of his behaviour in raping Rinda. 124 It is not that they disapprove of the rape, disapprove `quoll diui they teterrimum rather scenicis artibus et muliebris officii suscepcione (`because, duties through adopting actors' tricks and women's nominis opprobrium edidisset' he had brought the foulest of slurs on their reputation'). "' These reasons clearly relate to Obinn's his female disguise This Rinda, in disguise. to use of and specifically use of raping female Scandinavian disguise be identified tradition of the use of can plausibly as sei r, and a Rinda be rape of argued using sezr can Kormkr Qgmundarson's basis of on the reference, have for to a considerable time existed

in his Sigurrardrdpa, to `Seit Yggr til Rindar' 126 Saxo his before wrote version of the story.

Danorum. In both There is a clear similarity betweenthe two exilesin the Gesta
Obinn is Denmark, flees his by return to some part of upon a magician, who cases, replaced In Mithothyn, book is killed by is inhabitants i the magician the of the region. and there and

in book 3 the magician is called Ollerus, apparently the name of the Scandinavian deity Ullr. Hilda Ellis Davidson and Peter Fisher argue that `Saxo'sinsistence that he [Ollerus] was

123 GestaDanorum, ed. by Holder, p. z5. 124 GestaDanorum, ed. by Holder, p. 8i. 125 GestaDanorum, ed. by Holder, p. 8i. Translation from The History of the Danes, 1,78. 126 Den Norsk-Islndska Skaldediktningen,rev. by Kock, p. 42.(stanza3).

220

by Odin's implies he here is Mithothyn"2'. This that to the story of called name referring does not seemlikely, however.Although Saxoclearlyfits a usurper or impostor into both for different different narratives,the causes the exilesclearlysuggestthat two narrativesare intended. The usurper figure is a necessary in in exile narratives; the character many classical Jason Odyssey Odysseus's his home, the suitors usurp wife aswell, while and attempt to usurp in the Argonautika is the prototypical victim of usurpation by a wicked uncle. A direct source Saxo, idea be here; is likely it the that need not necessarily sought of exile of a presentedwith figure who is to some extent a hero in the Gesta Danorum, would immediately havethought in terms of his replacementby a usurper.Moreover, asmentioned above,it seemslikely that it longstanding Rinda depended tradition, and the narrative of the rape of upon a relatively Obinn, linked is likely included to the exile of since this closely that this tradition also seems Saxo have his use of sei6r.Given the existenceof this tradition, it seems would unlikely that book in for impetus different i. the exile, retold the story, with a totally One is left, then, with the question of the origin of the tradition of Frigg's infidelity 128 I have book in the talking that the motif of elsewhere argued one. and statue-breaking from developed heathen late based in is traditions which upon statue this narrative 129 Frigg hagiographic motifs of animation of idols. Why this motif should be associated with in this way is not entirely clear. It is interesting, however,that Frigg is presentedastricking her husband even as early as the seventhcentury, when the Origo GentisLangobardorum Langobards he him her than the the to rather that victory tricking grants so portrays Vandals, ashe intended (seesection 3.6.z, above).

12The History of the Danes, 11,57. 128 GestaDanorum, ed. by Holder, p. z5. '29 Shaw, `Miracle as Magic'.

221
i

Obinn Physician Gesta in Danorum the 5.4.3 as

binn

healer Gesta Danorum. in As described than the appearsmore once as a above,

in the narrative of his rape of Rinda in book three he disguises himself as a female physician 13o Vecha, finally Rinda. Saxo in does however, called and succeeds raping not, explicitly state O inn heals in In book huge in that this anyone episode. nine, contrast, a old man called Rostarus cures Sivard of a wound in return for a promise that he will dedicate the souls of all O 131 he kills battle Hroptr, in Rostarus is probably by-name those to him. sinn under the he in book Rinda disguised in three, the since also appears episode, as a smith called Rosterus. 132 The by-name Hroptr is not particularly common, but does occur several times in skaldic verse of the late tenth or early eleventh century. There is no connection of Hroptr by-name Saxo's however. healing in the skaldic verse, need use of the with metal-working or Obinn's depend not, then, upon pre-existing versions of these stories, although a narrative of (see Rinda half have in of the tenth century above). rape of may existed already the second This does not, however, prove that the by-name Hroptr was associatedwith the rape

Saxo known Clearly, have Saxo. Rinda known in the to must a version of the narrative of Obinn, in but from by-name the the of name occurrences given of some sourceas a name The had it. Saxo be existenceof such come across surprisedthat skaldic verse,we need not Obinn in interfering by-names would no doubt havesupportedSaxo'svision of mortal as him have (see in disguise with some readyprovided section 5.4, above),and would also affairs Obinn to Korm$kr The female disguise disguise. for in which usewhen made names describesassei6r,however,must haveexistedin pre-Saxoversionsof the rape of Rinda. This Obinn but it himself, is disguising Saxo's for be entirely motif of could also seenas a source Obinn Rinda in disguises episodewere additions made uses the which possible that the other by Saxo. It is also the casethat many of the other instancesof Obinn appearingin disguisein

130 Gesta Danorum, ed. by Holder, p. 8o. 131 Gesta Danorum, ed. by Holder, p. 304. 132 GestaDanorum, ed. by Holder, p. 79.

222

Gesta (see Danorum heathen the to conform classical models above),while set'6r represents a tradition whose aim is probably not disguise(although Saxomay haveinterpreted it assuch) but the exerciseof magical power by meansof transvestism. To return to Rostarus'shealing of Sivard, Saxodoesnot provide a specific
for healing, `senex liuorem this mechanism merely stating that repentino attrectate tabis ('the dispulit, intendit' manus auxilio subitamque uulneri cicatricem old man, touching the discoloured point of infection, cleared it away with a quick manipulation of his hand and 133 Rostarus The implication is uses clearly that promptly stretched a scar across the wound'). healing is by healing in but is the there no special magical process which the magic process, Obinn Scandinavia. Traditions healing are not confined to accomplished. with associating Anglo-Saxon England and the continent also appear to have had traditions making this in healing Nine Herbs Charm, Woden in the snake-bite, and association, since appears the SecondMerseburg Charm, healing a lamed horse (seesection 4.3, above). That these are Woden's likely; is but role varies perhaps not possible, reflexes of a single tradition his in use these extra-Scandinavian charms probably reflects the association considerably, and different Saxo in is The healing Mercurius healing. Roman deity of a very with of the Obinn in himself disguising here is person. as a physician, and curing patients actually nature; Obinn healing Sivard lacks any distinctive details which might suggest The description of knew Saxo the story as an oral tradition, probably something about the origin of the story. This is, he did interesting have it included not reshape. anecdote which as an and seems to however, simply speculation.

Snorri Saxo From Skalds to and 5.5

It would be unfair to dismiss SaxoGrammaticusand Snorri Sturluson asfantasists; history learned the pre-christian and their own to men, attempting understand these were

223

forefathers. At the same time, we should not forget that they were mythologies of their Europeans. Their scholarship was rooted in European tradition fed a of scholarship, and was by one of the renewals of interest in, and use of, history literature which and classical European in did Snorri Saxo Middle Ages. periodically well up cultures throughout the and honestly to understand and truthfully to represent their past, but they did so through attempt lens the of twelfth- and thirteenth-century scholarship, and the effects can clearly be discerned in their work. 134 At the same time, however, we can, with care, read past the distortions. Saxo and Snorri drink in the past with more zeal than care, but we have the opportunity to attempt hopefully with more successthan Sinfj9tli Snorri's Saxo to strain out and additions to the

deals drink it. Appropriately past, as we with the problem of enough, the casestudy above drinking in the mythology of Obinn, least drinking in in this act of the at and suggeststhat,

Snorri in indication importance, the pre-christian past, of past, provides us with some of the drinking. This insight would not be possible, nevertheless, if Snorri did not preserve skaldic disjunctions for it is the and verse of the pre-christian period, only through questioning inconsistencies between Snorri's account and his skaldic sources that we can accurately gauge Snorri's distortions. the extent and nature of

binn The usesof in twelfth- and thirteenth-century Scandinavian mythography are, O development inn in interesting of as a themselves, and revealan exuberant of course, for Angloin Just Woden is heathen figure associated political ends re-used as past. with the Saxon England, there may well be political motives behind the Icelandic usesof heathenism, Obinn. differ These Saxo's behind markedly phenomena portrayal of and, assuggestedabove, from the Anglo-Saxon re-creation of Woden, in that their effectiveness relies on the status Obinn Woden Anglo-Saxon heathen deity, present as royal genealogies as a whereasthe of

133 GestaDanorum, ed. by Holder, p. 304. Translation from The History of theDanes,i, 283. 134 Friis-Jensen, in her by before Saxo's been have The classicising tendenciesof not only noted work Models', Classical Saxo, Lay Ingellus its but Karsten 'The in Friis-Jensen, and of also monograph on deny Saxo by Friis-Jensen, At in Saxo Grammaticus, 65-77. that the sametime, we should not ed. pp.

224

O Scandinavian however, human. Having inn, to responses and examinedchristian simply for Scandinavian liminal heathen-christian difficult terrain of the contexts and syncretistic literature, we should now turn our attention to what little non-literary evidencethere is O inn. to which unequivocallyrelates the cult of

U1fuR Auk ujin 5.6

Auk

HutiuR:

Traces of the Cult of OSinn

O is an eighthThe earliest certain and datableevidencefor inn in Scandinavia follows: Denmark, Ribe inscription found in which readsas at century runic

I hL I ibburis I uijR ulfuRAukupinAukxutiuR

IAi uiArkiAuktuir

I kuniG [perforation] buur135

This runic inscription appearson a more or less rectangular fragment from the top of a human skull, measuring approximately 6 by 8.5 centimetres. Moltke identifies the inscription, from belonging it basis the transition older the to period of the the as uses, rune-forms of on 136 On CE between futhark, he 65o and the ninth century. places to the younger which later down he dating to the eighth century, certainly no narrows the archaeological grounds 137 basis fragment Stoklund dates CE. of the more precisely, on the than Boo Ribe layers found in dendrochronological analysis of wooden artefacts the same of the

Sources The Icelandic in demonstrated has Gubnason Bjarni did use native Scandinavian sources,as Friis-Jensen, by Saxo Grammaticus, in Grammaticus', Saxo 79-93. pp. ed. of
135 Moltke, Runes and Their Origin, p. z5i. In this transcription, G, H and m represent the graphs of for figure below, See futhark. in futhark, a i, while A represents the new graph the younger the older Inscription Cranium Ribe `The Stoklund, Marie from drawing of the artefact, taken and the Neighbouring Runes Frisian in Futhark', Younger Reduced Transition Scandinavian and to the Museum, Fries Runes Frisian Symposium International Proceedings First Traditions: at the on of the Beitrge Amsterdamer Quak, Arend Looijenga January by Tineke Leeuwarden, z6-29 and 1994, ed. (p. (Amsterdam: Rodopi, Germanistik, zoi). 1996), pp. 199-2o9 45 zur lteren 136 Moltke, Runes and Their Origin, pp. i5o-51. 137 Moltke, Runes and Their Origin, p. 347-

225

\ \\\\\ HIV \\
Figure z- Ribe skull fragment

226

138 fragment, excavations as the skull to the 720s. Moltke also notes that the fragment was from `[... ] an old skull the writer availed himself just knocked the of - not skull of someone head for To fragment the does not appear, the on purpose'139 this that one the might add . from the photograph given by Moltke, to have been for the carefully or neatly prepared inscription. 140 The serration visible in two of the edgesof the fragment from may result fragment cutting the out of a more complete skull by a series of blows with a more or less instrument; fragment hardly it is round-headed without close examination of the actual possible to be certain. 'The other edgesdo not show any possible signs of working in the photograph, although again this can only be definitely confirmed by examination of the fragment itself. Moltke notes that the (clearly human-made) perforation in the fragment was 141 from inside, it inscription hole, the made that the and appears was made after the since the inscription straddles the hole with no sign of letters having been lost. The inscription is in a false boustrophedon follows fragment. first line, in This species of the the the which, edge of inscription fragment that the strongly suggests was produced after any cutting of the (including boring the hole) was done. It is odd, then, that the primary orientation of the inscription indicates that the horizontal edge of the fragment further from the perforation is fragment. It is in the nature of false boustrophedon that part of any the upper edge of the inscription arranged in this manner must be upside down from the point of view of the reader he but inscription it is when or she starts reading, normally not the part of the at which the It location here is the of the perforation would reader to start reading. seems odd, then, that fragment first down, hung be if is the were upside cause the words the reader to read to up lack fragment Marie Stoklund the the that of signs of wear on using the perforation. argues indicate that it was never carried on a string through the perforation, and this presumably

13s Transition to the Younger Marie Stoklund, The Ribe Cranium Inscription and the Scandinavian International First Traditions: Proceedings ReducedFuthark', in Frisian Runesand Neighbouring of the by Tineke Frisian Runes Leeuwarden, Fries Museum, Symposium on 26-29january r994, ed. at the Looijenga and Arend Quak, Amsterdamer Beitrgezur lteren Germanistik, 45 (Amsterdam: Rodopi, (p. Language, The Early Runic Nielsen dating in p. 146. supportsa similar 1996),pp. 199-209 199).
139 Moltke, Runes and Their Origin, p. 346. 140 Moltke, Runes and Their Origin, p. 346.

227

142 be hanging could seento preclude other meansof using the perforation aswell; one should however, lack indicates lack in that total the note, of wear of use of a not a perforation this but little The is, however, direction manner, only very use of the perforation. of the writing demonstrates be in inscription that the perfectly clear, and should read as the transcription given above. Determining the word-breaking and basiccontent of the first section presentsno real difficulties. It consistsof three nameslinked by the preposition `Auk': `ulfuRAuk ujiin Auk Obinn. 143 determine, The It difficult is is to HutiuR'. middle name clearly that of more however, the significanceof the other names.It is possiblethat they are both proper -names best `HutiuR' is Nielsen however, deities. that probably referring to two other points out, Odin', `nok "`Hojtyr", hoje den til while vrre et tilnavn gud', a namewhich understood as He sees `ulfim' is probably related to the word ulv, `wolf. 1`'4 this animal asa symbol relating to
the cult of Obinn, her igen `har interpretation that et tilnavn vi nok and therefore gives the

by-names Our knowledge This Odini145. is of of the existence of til quite a persuasiveview. Obinn is (some Icelandic Old dependent late of material, since skaldic poetry on not entirely inscription) is deal their of existence evidence than this provides younger a great which not O inn (see section 5.2, above). It is not implausible, then, to suggest that by-names for literary based (insecurely) is Danes upon at this time, nor the suggestion existed among the for difficult This later. the account would also view particular centuries evidence of several former buns', `hiAlb is' `hiAlbbur be divided `hiAlbburis', the or which could sequence latter `to help', form the giving apparently a of the verb showing an anachronistic plural 146 fact latter Were form. imperative three that the all the reading, to accept one singular for help deity the of a singular use to puzzling account to might refer one would names imperative with three subjects. It is also possible, however, that the three names constitute a

141 Moltke, Runes and Their Origin, p. 161(note 3). 142 Stoklund, p. Z04. 143 Moltke, Runes and Their Origin, pp. 151-51'4' Nielsen, Danske Runeindskrifter, pp. 53-54145 Nielsen, Danske Runeindskrifter, p. 54.

228

separate section, and that this sequencecan therefore be interpreted as a noun `hLJb', `help', followed by either dative Moltke) (according Bur' `buns', `for an objective genitive to and or a the third person singular of the verb to be, `buri(i)s', `there is (help) for Bur' (according to Nielsen). 14? The latter suggestion by both be photographs of the would appear to supported artefact published by Moltke and Nielsen, on which there appear to be two consecutive i148 in `buns' (ie. `buriis'). It is not possible to be certain from the photographs, however, runes that the second i-rune is not a mark unrelated to the runic inscription; as Stoklund points out, moreover, the photograph given by Moltke has `retouched runes' and is therefore 14g (based `'). problematic on examination of the artefact itself, she reads The next three words are quite straightforward: `uiJR JAiMAuiArki', which we may `against preliminarily translate as the pain'. At this point, Moltke states that `the rest of the inscription must be counted as unsolved'150. Nielsen, however, points out that `I Ai A' appears "' be form He `uiArki' be in to a plural the singular. therefore reads the gwhile appearsto `kuniG' badly-formed form dative rune of then as a part of another u-rune, which would in /-iu/. The Nielsen's `tuirkuniu' is, in then, singular, ending sequence view, a compound first `dwarf, `sword', he the word, with element which relates to the and the second element form unnr which is found in skaldic verse.152 Stoklund, in contrast, reads this problematic [is] `the but `tuirk dwarf as n, and translates the resultant rune not as g or u, unin' as 15' The final Nielsen `buur', hole, be to the takes the word which comes after conquered'. kendte det Bur: `Bur, der traditionsbaerer, som nye 16var en name of the patient and carver,

146 Moltke, Runes and Their Origin, pp. 152--53148 Moltke, Runes and Their Origin, p. 346; Nielsen, Danske Runeindskrifter, p. SS.

147 Moltke, Runesand Their Origin, p. 153; Nielsen, DanskeRuneindskrifIer, p. 54. pp.
202-03.

149 Stokiund,

IsoMoltke, Runes and Their Origin, p. isz. Isl Nielsen, Danske Runeindskrifter, 54. P. IseNielsen, Danske Runeindskrifter, 54p. 153 Stoklund, pp. zoz and zog.

229

tegnsalfabetog kunne digte vers,har lavet en amulet, som han sely har baret haengende'ls4 One might therefore translate lines: inscription the asa whole along these
Wolf and Odin [things: High-Tyr Bur help for is against] and against these a lss [against] dwarf-sword [Signed] Bur. pain and One need hardly accept this interpretation definitive, but it makes good senseof the as inscription and does not seem linguistically impossible; Stoklund's nevertheless, interpretation, which would least is `the is `dwarf-sword' dwarf replace the conquered' at with equally plausible. In either case,the indication that this text is a charm seems to be confirmed by the nature of the artefact, for it seemsto be fact (whether do it did in so) to or not meant hang on something or someone, but is clearly not a decorative piece. It seemssafe to say that be the text may plausibly interpreted as a charm. This charm bears little resemblance,however, to later manuscript charms which invoke Wodan, such as the SecondMerseburg Charm and the Nine Herbs Charm, both of least One down (see however, later which were at written section 4.3, above). might note, that this charm seems to be intended to be efficacious as a writing, whereas the other two are intended being being in presumably spoken, rather than as memory aids, the text efficacious in being written down. The power inherent in the text on the Ribe skull-fragment is passed itself. The in the on to artefact as a whole, which then exerts this power and of manuscript in is their charms, contrast, are effective only when recited; power realised solely through The both longer involving performance. manuscript charms are and more complex, an Ribe formula. The if believe charm, we are to element of narrative as well as of magical Nielsen, simply states that the divine person or persons are helpful to the patient against pain has `dwarf-sword'; highly it any narrative unlikely that the charm certainly, seems and element.

healing, just The charm providesevidenceof an association of with asthe Obinn here be link Wodan/Woden healing, but to as a appealed may not with other charms

Obinn

'S4Nielsen, Danske Runeindskrifter, p. 58. iss Dwarf-sword is presumably a stitch or some similar shooting pain.

230

specifically healing god. The occurrences of Wodan/Woden in healing charms, are, however, frequent more than those of other Germanic deities, although the value of this evidence is (see development mitigated by its sparseness, does be and this seem to a strictly christian previous chapter). Of potentially greater interest is the use of three divine names which Obinn. It is, of course, possible that High-Tyr is used to refer to Tyr, perhaps all refer to but the existence of a Germanic deity Ulfurr is It is certainly the not otherwise attested. simplest explanation of the etymology of the name to regard it as related to `wolf, whether it is a by-name for O sinn or the In is deity. it name of a separate extremely either case, interesting to find so early a connection between Obinn is this the wolf, since connection and known before the Old Norse literary sources. For once, it may be possible to otherwise not establish clearly a relatively old tradition behind some of the Old Norse evidence, which also Obinn IS6 connects with the wolf.

Conclusion 5.7
Safeevidencefor the cult of O inn is, then, almost asscarce for that the cult of as
Wodan. These two sets of evidence, nevertheless,present rather different pictures. Wodan deity Alamanni Langobards, the appears as a particularly associatedwith the and capable of in battle, ferocity in granting victory and very possibly associatedwith special courage and O but inn, hand, battle, in the not associatedwith poetry or wisdom. on other appears a deity, function, foremost but intimately as the skaldic protective, perhaps medical, associated The with poetry. skaldic use of Obinn, involves moreover, a strong element of engagement

Christianity, both in terms of a mythological responseto the cosmological teachings of with

Christianity, and in the creation of an acceptable rangeof usesof

his sinn, and associated

Woden This in in is, the to usesof similar a sense, mythology, within christian contexts.

156 first in Thorbjprn Hornklofi's late ninth-century Haraldskv&i; one For instance, u 6nar appear have been for inspiration the association of uenar may should point out, of course,that the

231

Anglo-Saxon England, in that these too seekto render Woden Christians. At to acceptable Obinn however, liminal heathen-christian from the sametime, differ the skaldic usesof Anglo-Saxon usesof Woden in seekingto draw upon Obinn as a self-evidently religious figure, and within a rich mythological context at least inherited from heathenism. partially It is hard, then, to seeany reasonto considerWodan and O inn being direct as
Although it is possible that they are, and that they underwent processesof widely cognates. divergent development prior to their appearancein the historical does record, this not seem likely. The lack of clear similarities between Obinn Wodan in their earliest appearances and militates against such a view. At the same time, Wodan's earliest discernible attributes clearly Obinn him from ('madness, fit furor'), *wod connect with an etymology while those of from *woJ (`poetry'). This is consonant with an understanding of readily with an etymology deities There for however, literary is, the two as non-cognates. considerable evidence Ocbinn Wodan/Woden, in equations of the and which the two conditions which created by before, if be being the tenth came, to century not portrayed similarly, and seen as Obinn figure. We furthermore, by-names essentially the same might, speculate that some of formed by (in localityincorporation were a similar process early skaldic contexts) of of or 157 deities group-specific who are now not otherwise evidenced.

Obinn The Scandinavian deity is first evidenced in Denmark in the eighth century. O How long he had existedprior to this is unclear.What is clear, however,is that inn's importance in the twelfth- and thirteenth-century mythographywhich has attracted so much In largely his due is to scholarly attention centrality to the practice of skaldic verse. the ninth Obinn if before, be with the practice of poetry, century, not cameto very closelyassociated fulfil functions within the context of a of memorialisation which versecould and with the Obinn The warrior aristocracy. associationof with riddling, trickery and arcanewisdom had in difficulties which twelfth- and thirteenth-century authors perhapsreflects the

Wodan in his Suebic/Langobardic (see with cult section 3.6.3,above). cynocephali

232

The understanding the allusive and obscureverses the of early skalds. portrayal of

Obinn

itself, in his functions within extant early skaldic verse contrast, concentrates on relating its to specifically skaldic verseand cultural milieu, giving little indication that he wasviewed as particularly concernedwith wisdom or arcanematters in the ninth and tenth centuries. With the historiographical and antiquarianimpulsesof the twelfth to thirteenth 11 Obinn life. His took centuries, on new prominencewithin the skaldic tradition, which late heathen him in in less spanned and christian culture a more or unbroken manner, placed Scandinavian to take a good position on a prominent role within the mythography of these Snorri demoting from England Mercurius, Woden took centuries. with not the equationof him to the level of a messenger-god, but the use of Woden asa royal ancestor,strengthening Snorri's interpretation of O sinn asa royal deity, the king of the gods,who wasin reality, in his euherneristicreading, a human king.
O The attempt to classicise inn in Scandinavian mythography clearly has parallels in between differences but Wodan in the classicising uses of there are also the eighth century, Snorri, have Saxo deities. these two treatments of pre-christian a genuine native and crucially, draw from in the shape of skaldic verse, which and within which they can classical tradition, O Diaconus in Paulus in idea inn; the eighth century, contrast, of and situate their Paulus Langobard Wodan culture. to create a non-existent classical attempted to use focussed not on Wodan himself, so much as on the idea of a former greatnessfor the Langobards, which served a contemporary function. Saxo and Snorri, on the other hand, had from form in which they attempted, not always of skaldic verse, a record of the past, the legendary history of successfully, to reconstruct the past, and particularly the mythology and but in in These some cases the past. authors all constructed the past one way or another, how intentions, in it, they and no matter they reconstructed and some -- whatever their perceived their enterprise it. invented they

157 Turville-Petre explainsJQrmunrand Hroptr aspossiblybeing the namesof now-forgotten separate deities (ScaldicPoetry,p. xliv).

233

6. Conclusion: Reading and Writing Wodan

We began the discussion of previous scholarship on Wodan by his remarking on Perhaps better-placed versatility. now we are to evaluatethe reasonsfor this. It has been Obinn; Wodan be identical Wodan I indeed, have shown that need not with argued that Oiinn have been deities in and may very well separate origin, although they were (not deliberately) by christian writers as early as the eighth century. conflated necessarily The influence of the figure Obinn on the usesand understandings of Wodan in the ninth to eleventh centuries, particularly in England, is another reasonfor the variety of guisesin Woden which can appear. The vagariesof the historical development of Wodan's and Obinn's cults are not,
however, the only factor in understanding how and why Wodan was so widely and so Germania. The in multifariously used and re-imagined reception and use of southern deities in Irish, Frankish Anglo-Saxon classical and circles created a context within which

heathen deities such as Wodan could be re-used by christians as part of a learned classical heritage. Moreover, the availability of Wodan as a culted but declining deity on the Frankish periphery allowed for the development of an idea of a classicalpast specific to Germanic tribes. Paulus Diaconus's Historia Langobardorum perhapsmost clearly but the roots of this tradition are present already this vernacularisedclassicism, expresses in Jonas of Bobbio's Vita Sancti Columbani and in Bede'sHistoria Ecclesiastica. A slightly different form of classicismalso impacted on Wodan, particularly in Anglo-Saxon England. Although he could function as a Germanic classicaldeity, he for Mercurius, be and was therefore able to assumeroles could also used as an equivalent Woden Charm. just Mercurius, At Nine Herbs in the sametime, as could as the proper to be understood as an euhemeriseddeity capableof functioning as a figure of a Germanic diety, he be the antithesis par also continued to seen as a non-christian classicalperiod, so I(B). Maxims in the christian god, as of excellence Wodan appearsto have arisen as a deity of the Alamanni and Langobards in the Danube, further the western or perhaps north around the northern stretches of region of date hard It is development. Elbe. be The to the the the more northerly stagewould

234

earlier of the two, as these tribes moved south through the Migration Age; somewhere in the third to sixth centuries CE is the only safe dating that can be given. There is no reason to suppose that the cult spread significantly from the Alamanni and Langobards to before the seventh century, and, with the conversion of these tribes to other groupings Christianity, the cult died out. This cult, like the heathen vast majority of cults, was highly localised in ethnic and geographical terms. Through Columbanus's chance Wodan's in honour (or at least through the narrative of this encounter with a rite encounter, whether the encounter itself occurred or not), and through the strange story how Langobards Wodan became far Christians important the of their got to name, more throughout southern Germania than he ever had been to the majority of their heathen ancestors. Obinn, in contrast, appears to have followed a very different trajectory. Our first Obinn for dates dating have for his to the eighth century, and we can evidence no certain Obinn itself is implausible, if in the origin, although eighth century perhaps not was deity All he first be is is origin the tutelary of skaldic verse. we can certain of that in densities higher higher the evidenced eighth century, and then accrues and of evidence through the ninth and tenth centuries. Much of this evidence is skaldic verse, and it deity deity is seems to show a who specifically the tutelary of skalds and their verse. Perhaps Obinn had existed for a long time prior to this, and was simply co-opted by the for for but is their own purposes; this possible, early skaldic skalds perhaps not plausible, Obinn, functions for little indication and the coincidence of verse gives of extra-poetical for Obinn least is development the at suggestive of of skaldic verse with the early evidence both development interdependent of skaldic verse and a contemporaneous and Obinn

himself. At any rate, Obinn was, as far as our evidence allows, the patron deity of skaldic No doubt in CE. his in ninth early verse the name was understood and tenth centuries have His does undergone the same not seem to skaldic circles as referring to poetry. cult destruction Wodan. On the contrary, process of and re-imagination which affected that of Obinn have largely the cult of seems to although within skaldic circles which existed were, naturally, aristocratic or royal, and therefore most exposed to missionary efforts, the Obinn in Christianity. have functioned appears many ways to cult of as a response to

235

Ultimately, the cult of Obinn gaveway to that of Christ, but the skaldic cult seemsfor a have Christianity. while to offered a plausible alternative to The continuing use of skaldic versein christian contexts entailed a continuing use In Germanic Obinn in such contexts. some ways this mirrors the classicismof the of Wodan; Sturluson's Snorri Skldskaparmal certainly, presents eighth-century usesof heritage, skaldic verse and skaldic mythology very much as a speciesof vernacular classical just as Saxo'selaboration of vernacularversein translation placesthis versewithin Roman The classicaltraditions. continuing composition of skaldic verse at christian courts, however, possibly ought not to be seenas a scholarly exerciseor a form of classicism;here In discourse. being living this tradition of courtly we seeskaldic verse used as part of a by lost heathen and replaced christian mythology of the verseswas gradually context, the
did The Obinn in importance not vanish overnight, of skaldic cosmography elements. both Eiriksmal Hkonarml, in composed as postmortem and nevertheless, as one sees kings journeying kings, both for to an as portraying those and christian praise-poetry Valhpll. in afterlife

In Anglo-Saxon England, aboveall, the rich conjunction of English and Danish

i

cultures -

Obinn Woden and and particularly the conjunction of -

in createda climate

be Mercurius) Roman deity (and, deities indeed, both could and equated could the which influence each other. This has tended to give modern scholarsthe impression that Ocbinn They did, done have Woden and so. shared a common origin, when they need not however, influence each other heavily in christian thought and literature, and they At England. Anglo-Saxon in the the other the with one certainly underwent conflation Angloin or equated conflated always they not that were time, must recognise one same Danish spheres;in some circles there appearto have been attempts to depict the two figures as distinct from each other, probably for political reasons. Obinn) (and in Wodan developments The picture that emergesof the and usesof Although Ages is, Middle the cult of of understanding our then, a complex one. the Wodan, or of any other heathen deity, is necessarilyskewed by the limitations of our historical the cult. of outlines the towards can go we some way reconstructing evidence, CE in Wodan itself the This reconstruction was already the eighth century suggeststhat Obinn, importance The historical based beneficiary of of misprision on skewed evidence.

236

however, seems more grounded in his considerable status within skaldic discourse. The figures by in are adopted, re-used and re-imagined ways which these two cult christian is, better-evidenced but I than their cults, ultimately, not only writers and artists also, interest. would suggest, of greater

237

Bibliography

Arrangement of the Bibliography

This bibliography is not arrangedaccording to the customary division of primary

Since in include the sources used texts preserved manuscripts and and secondary materials. (both felt desirable been it has inscriptions), to with and without textual also artefacts ('Artefacts'). latter for former ('Manuscript Sources') the and the provide separate sections The section entitled `Artefacts' is intended to list only materials which function primarily do fall in within this category not question; materials which as editions of the artefacts `Studies', in with the other secondary materials. appear

Manuscript Sources

Abrenuntiatio Diaboli (Frankish)

(Berlin: Steinmeyer by Elias Sprachdenkmler, Die kleineren althochdeutschen von ed. Weidmann, i9i6)

Abrenuntiatio Diaboli (Old Saxon)

E. Brill, J. (Leiden: Gallee H. Old Saxon Texts, ed. by 1894) J. Mit Anmerkungenund Glossar,ed. by Elis Sprachdenkmler: Kleinere altschsische Niederdeutsche Denkmler, 6 (Norden: Diedr. Soltan, 1899) 'X%adstein,

Adam of Bremen
der hamburgischenKirche und desReiches, Quellen des g. und ii. Jahrhunderts zur Geschichte G. Waitz, by B. Schmeidler, Werner by G. Pertz, H. Bresslau trans. ed. and Trillmich and K. Nobbe, Ausgewhlte Quellen zur deutschen Geschichte des Mittelalters, ii (Berlin: Rsten & Loening, 1961)

238

Adso of Montier-en-Der
Adso Dervensis: De Ortu et TemporeAntichristi, ed. by D. Verhelst, Corpus Christianorum, Continuatio Mediaeualis, 45 (Turnhout: Brepols, 1976)

AElfric
Aelfric's Lives of Saints, ed. by Walter W. Skeat, Early English Text Society, o.s. 76,8z, (London: Trbner, 94 and 114,z vols 188i,1900) Homilies ofi Eric: A Supplementary Collection, ed. by John C. Pope, Early English Text Society, o.s. 259 and z6o, 2 vols (London: Oxford University Press, 1967)

Athelweard
The Chronicle of., thelweard,ed. by A. Campbell (London: Thomas Nelson, 1962)

Anglo-Saxon Chronicle
Two of the Saxon ChroniclesParallel, ed. by John Earle and Charles Plummer, z vols (Oxford: Clarendon Press, 1892-1899)

Bede
Venerabilis Baedae:Historiam Ecclesiasticam GentisAnglorum: Historiam Abbatum: Epistolam ad Ecgberctumuna cum Historia Abbatum AuctoreAnonymo, ed. by Charles Plummer, z voll (Oxford: Clarendon Press, 1896) De Temporum Ratione, ed. by Ch. W. Jones, in BedaeVenerabilisOpera, 8 vols, Corpus Christianorum, SeriesLatina, 123B (Turnholt: Brepols, 1977),vi. 2.

Beowu f
Beowulf and the Fight at Finnsburg, ed. by Fr. Klaeber, 3 edn (Boston: Heath, 195o;first published 1922) Beowul " With the FinnesburgFragment, ed. by C. L. Wrenn and W. F. Bolton, Exeter .. Medieval English Texts and Studies, 4th (revised) edn (Exeter: University of Exeter Press, 1988;repr. 1992)

239

Biblia Sacra
Biblia Sacra Iuxta Vulgatam Versionem, by Robertus Weber ed. and others, z vols (Stuttgart: Wrttembergische Bibelanstalt, 1969)

Capitularies Capitularia RegumFrancorum,ed. by Alfredus Boretius,Monumenta Germaniae

Historica, Legum Sectio II, Capitularia Regum Francorum, I (Hannover: Hahn, 1883)

Cassius Dio
CassiiDionis CocceianiHistoriarum Romanarum Quae Supersunt, Philippus Ursulus by ed. Boissevain,4 vols (Berlin: Weidmann, 1955)

Cicero
M. Tulli CiceronisDe Natura Deorum Liber Primus, ed. by Arthur Stanley Pease (Cambridge, MA: Harvard University Press,1955) De Natura Deorum, ed. by O. Plasbergand W. Ax, M. Tulli Ciceronis Scripta quae Manserunt Omnia, 4S,2nd edn, (Stuttgart: Teubner, i98o)

Cleopatra Glossary
Dictionary of Old English Corpus,<http: //ets. umdl. umich. edu/cgi/o/oecidx? type=bigger&byte=36297700&qi=logeT&qz=&q3=>
z002]

[accessedz6 November

Corpus Glossary
(Cambridge: Hessels by J. H. An Eighth Century Latin Anglo-Saxon Glossary, ed. Cambridge University Press, i89o)

Eddaic Verse
CorpusPoeticumBoreale: The Poetry of the Old Northern Tongue, ed. and trans. by Gudbrand Vigfusson and F. York Powell, z vols (Oxford: Clarendon Press, 1883) Denkmlern, ed. by Gustav Neckel Edda: Die Lieder desCodexRegiusnebstverwandten Germanische Kuhn, Hans Texte, Bibliothek, Reihe, 3rd edn, i vols vierte and
(Heidelberg: Carl Winter, i96z-1968)

240

Art ausden Fornaldarsgurund anderenProsawerken, Eddica Minora: Dichtungen eddischer Heusler Wilhelm Ranisch (Darmstadt: by Andreas Wissenschaftliche and ed. Buchgesellschaft, 1974[facsimile reprint of first edition, Dortmund: Franz Wilhelm Ruhfus, 1903])
The Poetic Edda, trans. by Carolyne Larrington repr. 1999)
11

(Oxford: Oxford University Press, 1996;

Epinal-Erfurt

Glossary

Old English Glosses in the Epinal Erfurt Glossary,ed. by J. D. Pheifer (Oxford: Oxford University Press, 1974)

Eutropius Eutropii Breviarumab Urbe Condita,ed. by CarolusSantini, Bibliotheca Scriptorum

Graecorum et Romanorum Teubneriana (Leipzig: Teubner, 1979)

Exeter Book
The Exeter Book, ed. by George Philip Krapp and Elliott van Kirk Dobbie, The AngloSaxon Poetic Records, 3 (London: Routledge; New York: Columbia University Press, 1936)

The Exeter Anthologyof Old EnglishPoetry: An Edition of Exeter Dean and ChapterMS Press, Exeter (Exeter: University by Bernard J. Muir, 1994) of z vols 3for, ed.

Fornaldar Spgur
Fornaldar SgurNor6urlanda, ed. by Gu6ni Jonsson, 4 vols (Reykjavik: Edda, 1950)

Fredegar Monumenta Krusch, Bruno by FredegariietAliorum Chronica.VitaeSanctorum, ed.

Germaniae Historica, Scriptorum Rerum Merovingicarum, I888)

Hahn, (Hannover: 2

B/ Vatican Genesis Genesis

Vatican Saxon Old Genesis: An Edition B' West `Genesis Saxon Saxon The and the of the 'Genesis', ed. by A. N. Doane, (Madison: University of Wisconsin Press, 1991)

241

Geoffrey of Monmouth
The `Historia RegumBritannie' of Geoffreyof Monmouth (Cambridge: Brewer, I94-I99I), Burgerbibliothek, MS. Neil Wright Bern, by (1985) i: y68, ed. Monmouth (Cambridge: The `Historia RegumBritannie' of Geoffrey Brewer, of 1985-1991), Version: The Variant (1988) First A Critical Edition, by Neil Wright II: ed.

GestaRegum Britannie
(Cambridge: The `Historia RegumBritannie' of Geoffrey Brewer, Monmouth 1985-1991), of (1991) Gesta Regum Britannie, Neil Wright by v: ed. and trans.

Hallfre6ar Saga
rna Magnussonar, 1977) Hallfre6ar Saga,ed. by Bjarni Einarsson (Reykjavik: Stofnun

Harley Glossary
Dictionary of Old English Corpus,<http: //ets. umdl. umich. edu/cgi/o/oecNovember idx?type=bigger&byte=36678133&qi=logeT&qz=&q3=> [accessed z7
2002

Helfand
Helland, ed. by Eduard Sievers, Germanistische Handbibliothek, 1878) (Halle: Waisenhaus, 4

Herbarium (Old English)

Jan de Hubert by de Quadrupedibus', `Medicina `Herbarium' Old English The ed. and Vriend, Early English Text Society, o.s. z86 (Oxford: Oxford University Press, 1984)

Historia Brittonum
History Morris, John by Annals', Welsh History' `The `British Nennius: ed. and trans. and from the Sources(London: Phillimore; Totowa: Rowman & Littlefield, i98o)

242

Historia Ecclesiastica GentisAnglorum (Old English)

The Old English Versionof Bedes Ecclesiastical History of theEnglishPeople,ed. and trans. by Thomas Miller, z vols, Early English Text Society, (vol i), no, iii o.s. 95,96 (vol ii) (London: Early English Text Society, 1890-1898)

Isidore of Seville
Isidori Hispalensis EpiscopiEtymologiarum sineOriginum Libri XX, ed. by W. M. Lindsay, Scriptorum Classicorum Bibliotheca Oxoniensis, z,vols (London and New York: Clarendon Press, i9i1)

Jonas of Bobbio
PassionesVitaeque Sanctorum Aevi Merovingici et Antiquorum Aliquot, ed. by Bruno Krusch, Monumenta Germaniae Historica, Scriptorum Rerum Merovingicarum, (Hannover: Hahn, 3 1896)

Passinnes VitaequeSanctorumMerovingicarum, ed. by Bruno Krusch, Monumenta Germaniae Historica, Scriptorum Rerum Merovingicarum, 4 (Hannover: Hahn,
1902

Jonas de Bobbio, Vie de Saint Columban et de Ses Disciples,trans. by Adalbert de Voge and Pierre Sangiani, Vie Monastique, i9 (Begrolles-en-Mauges: Abbaye de Bellefontaine, 1988)

Jordanes
lordanis De Origine ActibusqueGetarum, ed. by FrancescoGiunta and Antonino Grillone, (Rome: Istituto Storico Italiano, i99i) Fonti per la Scoria d'Italia, 117

Lacnunga Magic and Medicine:Illustrated J. H. G. Grattan and CharlesSinger,Anglo-Saxon

1952

from the Semi-Pagan Text `Lacnunga'(London: Oxford University Press, Specially

La3amon Layamon: `Brut', ed. by G. L. Brook and R. F. Leslie, Early English Text Society,2,5o
(London: Oxford University Press, and 277 1963-1978)

243

Life of Saint Pantaleon (Old English)

Life of Saint Pantaleon in Dictionary of Old English Corpus //ets. umdl. umich. edu/cgi/o/oec/oec<http: idx? type=bigger&byte=799441o&qi=gallige&qz=&q3=> zooz]

[accessedz December

Martial
M. Val. Martialis Epigrammata, ed. by W. M. Lindsay, Scriptorum Classicorum Bibliotheca Oxoniensis, znd edn (Oxford: Clarendon Press, i9z9; repr. 1959)

Martin of Braga
De CorrectioneRusticorum, in Martini EpiscopiBracarensis Opera Omnia, ed. by Claude W. Barlow, Papersand Monographs of the American Academy in Rome, iz (New Haven: Yale University Press; London: Oxford University Press, 19So)

Martyrology (Old English)

Martyrologium, ed. by Gnter Kotzor, Abhandlungen der bayerischen Das altenglische Akademie der Wissenschaften, Philosophisch-Historische Klasse,n.s. 88, nos i Wissenschaften, (Mnchen: der bayerische Akademie 1981) and z, z vols

Maxims I
SeeExeter Book, above

Maxims II
The Anglo-Saxon Minor Poems,ed. by Elliott van Kirk Dobbie, The Anglo-Saxon Poetic Records, 6 (London: Routledge, 1942)

Merseburg Charm, Second

Karl Helm by Lesebuch, Wilhelm Braune, by Althochdeutsches 13th edn, rev. ed. (Tbingen: Niemeyer, 1958)

Miscellaneous Old English

(London: Michael The Library Prose, by Swanton, Everyman Anglo-Saxon ed. and trans. Dent; Vermont: Tuttle, 1975;repr. 1993)

244

Anglo-Saxon Poetry: An Anthologyof Old EnglishPoemsin Prose Translation, trans. by S. ) A. J. Bradley, Everyman Classics(London: Dent, 1982, Dictionary of Old English Corpus, <http: //ets. umdl. umich. edu/o/oec/>

Napier 42
Homilien nebstUntersuchungen Wulfstan: Sammlung der ihm zugeschriebenen ber ihre Echtheit, ed. by Arthur Napier, Sammlung englischer Denkmler in kritischen Ausgaben, 4 (Berlin: Weidmann, 1883) Anglo-Saxon Manuscripts in Microfiche Facsimile, VIII: Wulfstan Texts and Other Homiletic Materials, descriptions by Jonathan Wilcox, Medieval and Renaissance Texts and Studies, zig (Tempe: Arizona Center for Medieval and Renaissance Studies, zooo)

Olaus Magnus
introduction by John Romae iyy-f-, OlausMagnus: Historia de GentibusSeptentrionalibus: Granlund (Copenhagen: Rosenkilde and Bagger, 197z)

Origo Gentis Langobardorum

Rerum Langobardicarumet Italicarum, Saec.VI-IX, ed. by G. Waitz, Scriptores Monumenta Germaniae Historica (Hannover: Hahn, 1878)

Patrologia Latina
Patrologia Latina, <http: //pld. chadwyck. co.uk/pld/search>

Paulus Diaconus
Waitz, G. by Saec. VI-IX, Italicarum, Langobardicarum Rerum Scriptores ed. et Monumenta Germaniae Historica (Hannover: Hahn, 1878) Historica, Germaniae Monumenta Duemmler, Ernest by Carolini, LatiniAevi Poetae ed. Poetarum Latinorum Medii Aevi, i (Berlin: Weidmann, 1881) Neff, Karl by Ausgabe, Erklrende Paulus Kritische des Diaconus: Gedichte Die ed. und Quellen und Untersuchungen zur lateinischen Philologie des Mittelalters, 3.4 (Munich: C. H. Beck, igo8)

245

Pearl-Poet
The Poems of the Pearl Manuscript: `Pearl, `Cleanness, `Patience', `Sir Gawain and the Green Knight', ed. by Malcolm Andrew and Ronald Waldron (London: Arnold, 1978)

Ptolemy
Claudii Ptolemiei Geographia, by Carolus Mllerus, (Paris: Didot, ed. z vols 1883)

Ruin, The
SeeExeter Book, above

Saxo Grammaticus
SaxonisGrammatici GestaDanorum, ed. by Alfred Holder (Strassburg:Trbner, 1886)
The First Nine Books of the Danish History of Saxo Grammaticus, trans. by Oliver Elton by Frederick (London: Nutt, York Powell David introduction 1894) and with an

Saxo Grammaticus: The History of the Danes,ed. by Hilda Ellis Davidson and trans. by Peter Fisher, z vols (Cambridge: Brewer; Totowa: Rowman and Littlefield, i98o)

Skaldic Verse
CorpusPoeticum Boreale: The Poetry of the Old Northern Tongue, ed. and trans. by Gudbrand Vigfusson and F. York Powell, z vols (Oxford: Clarendon Press, 1883) Den Norsk-IslandskeSkjaldedigtning,ed. by Finnur Jonsson, 2 vols (Copenhagen and Kristiania: Nordisk Forlag, 1912-1915)
(IA,
IIA, IB,

IIB)

Den Norsk-Islndska Skaldediktningen, rev. by Ernst A. Kock, z vols (Lund: Gleerup, 1946-1949)

Snorri Sturluson
Snorri Sturluson: Edda, trans. by Anthony Faulkes, Everyman Classics (London: Dent, 1987)

Snorri Sturluson: Edda: Prologueand 'Gyfaginning', ed. by Anthony Faulkes (London: Viking Society for Northern Research,1988;repr. zooo)

246

Snorri Sturluson: Edda: Skldskaparmdl, by Anthony Faulkes, (London: Viking ed. z vols Society for Northern Research,1998)

Solomon and Saturn Dialogues (Old English)

The Poetical Dialogues of Solomon and Saturn, ed. by Robert J. Menner, The Modern Language Association of America Monograph Series, (New York: The 13 Modern Language Association of America; London: Oxford University Press, 1941

Solomonand Saturn' and.Adrian and Ritheus', The `Prose by James E. Cross ed. and Thomas D. Hill, McMaster Old English Studies Texts, (Toronto: and i Toronto University Press, 1982)

Song of Roland
La Chanson de Roland, ed. by Cesare Segre, Documenti di Filologia, (Milan i6 and Naples: Riccardo Ricciardi, 1971)

Statius
P. Papini Stati Silvae,ed. Courtney (Oxford: Oxford University Press,i99o) E. by

Strabo
Strabo, Horace The Geography by Leonard Jones, The Loeb Classical trans. of ed. and Library, 8 vols (London: Heinemann; New York: Putnam, 1917-1932, )

Tacitus
C'ornel Taciti Opera Minora, ed. by Henricus Furneaux and J. G. C. Anderson (Oxford: Clarendon Press, 19oo; repr. 1958) Divi Augusti Libri, ed. by C. D. Fisher (Oxford: Cornelii Taciti Annalium ab Excessu Oxford University Press, i9o6; repr. 1956) Cornelii Taciti Historiarum Libri, ed. by C. D. Fisher (Oxford: Clarendon Press, i9ii; repr. 1956)

Virgil Bibliotheca P. Vergili Maronis Opera,ed. by R. A. B. Mynors, Scriptorum Classicorurn . Clarendon Press,
Oxoniensis (Oxford: 1969)

247

Vita Sancti Kentigerni

Acta Sanctorum:January, il, ed. by JoannesBollandus and others (Brussels:Greuse, 1863)

VQlsunga Saga
Die Vplsungasaga Bugges Text, by Wilhelm Ranisch, (Berlin: Mayer & nach ed. 2nd edn Mller, i9o8)

Wace
Wace's`Romande Brut': A History of theBritish, ed. by Judith Weiss, Exeter trans. and Medieval English Texts and Studies (Exeter: University Exeter Press, of 1999)

Wulfstan
The Homilies of Wulfitan, ed. by Dorothy Bethurum, (Oxford: Clarendon Press, 1957)

Artefacts

Bracteates
Hauck, Karl, and others, Die Goldbrakteaten der Vlkerwanderungszeit:Ikonographischer Katalog, Mnstersche Mittelalter-Schriften, 24, ed. Belting and others, 7 byH. Wilhelm (Munich: Fink, 1985-i989) vols

Latin Inscriptions
Gtternamender antiken Inschriften, Gutenbrunner, Siegfried, Die germanischen Rheinische Beitrge und Hlfsbcher zur germanischenPhilologie und Volkskunde, 24 (Halle: Niemeyer, 1936)

Metal Sculpture
Jitta, A. N. Zadoks-Josephus,W. J. T. Peters and W. A. van Es, Roman Bronze from the Netherlands,Scripta Archaeologica Groningana, i-2,2 vols Statuettes (Groningen: Wolters, 1967)

248

Runic Inscriptions
im lterenFuthark, Abhandlungen der Krause, Wolfgang, ed., Die Runeninschriften Akademie der Wissenschaften in Gttingen, Philologisch-Historische Klasse, (Gttingen: Vandenhoek Ruprecht, & 1966) series3,65, z vols
Age, Danske Runeindskrifter: Et Udvalgmed Kommentarer (Copenhagen: Niels Nielsen, Hernov, 1983)

SamnordiskRuntextdatabas[Windows version i. o; electronic resourcedownloadablefrom h <www.nordiska.uu. se/forskn/samnord. tm>]

Stone Sculpture
Bailey, Richard N., and Rosemary Cramp, eds, Corpus ofAnglo-Saxon Stone Sculpture, II: Cumberland, Westmorland and Lancashire North-of-the-Sands (Oxford: Oxford University Press, 1988)

Studies
Abrams, Lesley, `The Anglo-Saxons and the Christianization of Scandinavia', AngloSaxon England, z4 (1995), 213-49

(1999), Saga-Book, Sonatorrek', Ideas in `Religious Hnefill, Jon Abalsteinsson, 159-78 z

Based Place-Names: Cultic Religions Finnish Norse Old Tore, Ahlbck, on and and ed., Old Nordic Religions in Between Encounters Symposium Read Papers on at the Abo, August Finland, Held Times and on Cultic Place-Names on the i9tb-2, rst of at Wiksell, (Stockholm: Almqvist & Aboensis, Donneriani Instituti Scripta 13 1987, 1990)

Literary Old English Dates Determining Means Linguistic the Amos, Ashley Crandell, of of Academy Medieval The MA: (Cambridge, Books, Academy Medieval Texts, 9o America, i98o) of Containing All Forefathers, Our The Religion Or, the Mythology: Norse B., Anderson, R. of Griggs; (Chicago: Interpreted, Systematized Eddas, Myths of the nd edn, z, and London: Trbner, 1876)
Series Inscriptions, Runic Older Grammar Antonsen, Elmer H., A Concise of the (Tbingen: Sprachstrukturen, Historische A, Reihe 3 Sprachstrukturen, Niemeyer, 1975)

249

Ausenda, Giorgio, `Current Issuesand Future Directions in the Study of the Merovingian Period', in Franks and Alamanni in theMerovingian Period: An by Ian Wood, Studies EthnographicPerspective, in Historical ed. Archaeoethnology, 3 (Woodbridge: Boydell, 1998),PP-371-453 Axboe, Morten and Anne Kromann, DN ODINNPFAUC?: Germanic "Imperial Portraits" on ScandinavianGold Bracteates',in Ancient Portraiture: Image and Tobias Fischer-Hansen Message, by Acta Hyperborea, ed. and others, 4 (Copenhagen: Museum Tusculanum Press, 1992),PP. 271-3o5 Bailey, Richard N., Viking Age Sculpturein Northern England (London: Collins, i98o)
Behr, Charlotte, `The Origins of Kingship in Early Medieval Kent', Early Medieval Europe, 9 (woo), 25-52

Beresford, M. W., and G. R. J. Jones, eds, Leedsand its Region (Leeds: British Association for the Advancement of Science,1967)
Bickermanm, E. J., Chronology of theAncient World (London: Thames and Hudson, 1968) Birnbaum, Henrik, and Jaan Puhvel, eds, Ancient Indo-European Dialects: Proceedings of California, Conference Indo-European Linguistics Held University the of on at the Los Angeles,April zy 27, r963-(Berkeley: University of California Press, 1966) Bischoff, Bernhard, `Palographische Fragen deutscher Denkmler der Karolingerzeit', in Aufstze zur Schriftkunde und Mittelalterliche Studien: Ausgewhlte Literaturgeschichte, 3 vols (Stuttgart: Hiersemann, 1966-i98i), III (1981),pp. 73-III [originally published in Frhmittelalterliche Studien, i (1971),1o1-1341

Arhundredes Saxostudier, for Renaessance', det in Reprxsentant `Saxo, Blatt, Franz, iz. en (Copenhagen: Museum Opuscula Grxcolatina, Boserup, Ivan by z ed. Tusculanum, 1975),pp. 11-19
of in Thirteenth Christi: Famulus Gerald, Bonner, the ed., Bede(London: SPCK, 1976) Centenaryof theBirth of the Venerable Essays Commemoration

Boserup, Ivan, ed., Saxostudier,Opuscula Grxcolatina, z (Copenhagen:Museum Tusculanum, 1975) Saxo Gesta Danorum', Archetype in Fragment Angers `The Ivan, Boserup, of and the Norse and Latin Culture, ed. by Karsten Grammaticus:A Medieval Author between Friis-Jensen (Copenhagen: Museum Tusculanum Press, i98i), pp. 9-26
(Oxford: Clarendon Literature Old High German Handbook A Knight, J. Bostock, on Press, 1955)

250

Religion in der amtlichenkirchlichenLiteratur des Boudriot, Wilhelm, Die altgermanische (Bonn: Ludwig Jahrhundert Rhrscheid, bis Abendlandes 1964) ii. vom f
Bridges, George, `The Almost Irresistible Appeal of Fascism, Or: Is It Okay to Like Richard Wagner?, The Germanic Review, 64 (1989), 42-48

Hugh Browne, Arthur, and Peter Foote, eds, Early English and Norse Studies:Presented to Smith in Honour of his Sixtieth Birthday (London: Methuen, 1963) Butler, L. A. S., `Dark Age Archaeology', in Leedsand its Region, ed. by M. W. Beresford (Leeds: Science, Jones British Association for J. Advancement G. R. the of and
1967)) pp- 97-100

Calder, Daniel G., and T. Craig Christy, eds, Germania: ComparativeStudiesin the Old (Cambridge: Literatures Brewer, Germanic Languages 1988) and Cameron, Kenneth, EnglishPlace-Names,3rd edn (London: Batsford, 1977)
Cameron, M. L., Anglo-Saxon Medicine, Cambridge Studies in Anglo-Saxon England, 7 (Cambridge: Cambridge University Press, 1993)

Cameron, M. L., `Anglo-Saxon Medicine and Magic', Anglo-Saxon England, 17(1998),

191-2I5

Campbell, A., Old English Grammar (Oxford: Clarendon Press, 1959,repr. 1997) Cavill, Paul, Maxims in Old EnglishPoetry (Cambridge: Brewer, 1999)
North Religion Ancient Essay Othin: An in Cult The M., H. Chadwick, the of the of (London: Clay, 1899)

Discussion Poem Study An Introduction W., Beowulf. the R. Chambers, the of a to the with of " Press, University Cambridge (Cambridge: Finn, Offa Storiesof 1963) 3rd edn and Press, Toronto University (Toronto: `Beowuf Dating The i98i) Chase, Colin, ed., of of An Icelandic-English Craigie, A. William Vigfusson Gudbrand Cleasby, Richard, and Dictionary, znd edn (Oxford: Oxford University Press, 1957) Studies Anglo-Saxons: in The Works' in AElfric's Chronology M., `The Clemoes, P. A. of by Peter Dickins, Bruce Culture Presented History Aspects to ed. their and of some Clemoes (London: Bowes & Bowes, 1959),PP-212-47 Culture History Aspects Studies in Anglo-Saxons: The and Clemoes, Peter, ed., of their some Bowes, Bowes (London: & Dickins Bruce Presented 1959) to

251

Collier, Wendy, `The Tremulous Worcester Hand and Gregory's Pastoral Care', in Rewriting Old English in the Twelfth Century, ed. by Mary Swan and Elaine M. Treharne, Cambridge Studies in Anglo-Saxon England, 30 (Cambridge: Cambridge University Press, Zooo), pp. 195-zo8 Conrad-O'Briain, Helen, Anne Marie D'Arcy and John Scattergood, ed., Text Gloss: and Studiesin Insular Learning and Literature Presented Donovan Phezfer toJoseph (Dublin: Four Courts Press, 1999) Crick, Julia C., The `Historia RegumBritannie' of Geoffrey Monmouth (Cambridge: of Brewer, 1985-1991), Dissemination Reception Later in Middle (i99i) Ages iv: the and
Cusack, Carole M., The Rise of Christianity in Northern Europe, (London: Soo-1000 Cassell, 1998)

Davidson, H. R. Ellis, Pagan Scandinavia,Ancient Peoplesand Places,58 (London: Thames and Hudson, 1967) Davidson, Hilda Ellis, The Lost Beliefsof Northern Europe (London: Routledge, 1993) Davis, Craig R., `Cultural Assimilation in the Anglo-Saxon Royal Genealogies', AngloSaxon England, 21 (199z), 23-36 Dickinson, Tania and David Griffiths, eds,Anglo-SaxonStudiesin Archaeology and History, ro: Papers from the 47h Sachsensymposium, York, September r996(Oxford: Oxford University Committee for Archaeology, 1999) Dierkens, Alain, `Superstitions, christianisme et paganismea la fin de l'epoque A de 1'Indiculus Magie, in paganiarum', merovingienne: propos superstition um et Herve Hasquin, Serie Sorcellerie, Parapsychologie, by Laicite `Recherches', ed. 5 (Brussels: Editions de l'Universite de Bruxelles, [no date]), pp. 9-z6
Dirks, J., `Les Anglo-Saxons et Leurs Petits Deniers Dits Sceattas: Essai Historique et Numismatique', Revue de la Numismatique Belge, 5th series, z (1870), 81-iz8; z6932-0; 387-409; 521-41

Dodwell, C. R., The Pictorial Arts of the West Soo-r2oo, Pelican History of Art (New Haven: Yale University Press, 1993) Norroenum: Norse Studiesin Memory of Dronke, Ursula, and others, eds, Speculum Gabriel Turville-Petre (Odense: Odense University Press, 1981)
DuBois, Thomas A., Nordic Religions in the Viking Age (Philadelphia: University of Pennsylvania Press, 1999), PP. 74-81

la Religion Formation de la Dumezil, Georges,Les Dieux desGermains:Essar sur Universitaires de France) 1959) Mythes et Religions (Paris: Presses Scandinave,

252

Dumville, David N., `The Anglian Collection of Royal Genealogiesand Regnal Lists', Anglo-Saxon England, 5 (1976), 23-50 Dwel, Klaus, Runenkunde,ist edn Sammlung Metzler (Stuttgart: Metzler, [19681) Dwel, Klaus, `Runen und Interpretatio Christiana: Zur religionsgeschichtlichen Stellung Kraft, ed. by N. der Bgelfibel von Nordendorf I', in Tradition als historische Kamp and J. Wollasch (Berlin: de Gruyter, 1982),pp. 78-86 Dwel, Klaus, Runenkunde,2nd edn, Sammlung Metzler (Stuttgart: Metzler, 1983)
Dwel, Klaus, `Runeninschriften als Quellen der germanischen Religionsgeschichte', Altertumskunde, S (199z), 336Ergnzungsbnde zum Reallexikon der germanischen 64 Dwel, Klaus, `Frhe Schriftkultur bei den Barbaren: Germanische Runen, lateinische Inschriften', in Die Alamannen, ed. by Karlheinz Fuchs and others (Stuttgart: Theiss, Zooi), pp. 491-98 Einarsson, Bjarni, `Skaldib i Reykjaholti', in Eyvindarbok: Festskrit tZl Eyvznd Fjeld Halvorsen 4. mai 1992, ed. by Finn Hodnebo and others (Oslo: Institutt for litteraturvitenskap, 1992), pp. 34-40 nordistikk og Enright, Michael J., Lady with a Mead Cup: Ritual, Prophecy and Lordship in the European Warhand from La one to the Viking Age (Blackrock: Four Courts Press, 1996) (London: Athlone, Thames South Invasions Fifth-Century The I., Vera Evison, of the 1965)

Fabech, Charlotte, `Society and Landscape:From Collective Manifestations to Ceremonies of a New Ruling Class', in IconologiaSacra:Mythos, Bildkunst und Karl Hauck fr Festschrift, Alteuropas Sozialgeschichte Religionsder in Dichtung : und Arbeiten Staubach, Nikolaus Keller Hagen by Geburtstag, zur and ed. 7f zum Gruyter, de Walter (Berlin: Frhmittelalterforschung, 2-3 1994), PP.132-43 Resen's in Manuscript Lists Regnal in Genealogies `The Anthony, Faulkes, a and Einar by Benediktssyni, fakobi helga SjtIu Library, in ed. Zo.Judii, ar ritger6ir a Magnssonar Arna Stofnun (Reykjavik: Kristjnsson Jonas G. Petursson and Islandi, 1977),pp. 177-90 [for (1982, Scandinavia, Gods', Mediaeval from `Descent 1978Faulkes, Anthony, ii the 791)) 92-125 for English, Leeds Studies in Beor', 8 English `Old E., 1975 1976),76-95 n. s. Fell, Christine

253

Fidjestol, Bjarne, The Dating of Eddic Poetry: A Historical Surveyand Methodological Investigation,ed. by Odd Einar Haugen, Bibliotheca Arnamagnaeana, 41 (Copenhagen: Reitzel, 1999) Fischer-Hansen, Tobias and others, eds,Ancient Portraiture: Image and Message, Acta Hyperborea, 4 (Copenhagen: Museum Tusculanum Press, 1992) Fletcher, Richard, The Barbarian Conversion: From Paganismto Christianity (New York: Holt, 1997) [originally published in the United Kingdom as The conversion of Europe: From Paganismto Christianity, 37i-i386AD (London: Harper Collins, 1997)] Fox, Robin Lane, Pagansand Christiansin theMediterranean World from the Second Century AD to the Conversionof Constantine(London: Penguin, 1988) Fragen und Forschungen fr im Bereichund Umkreisder germanischen Philologie:Festgabe Theodor Frings zum 7o. Geburtstag, Verffentlichungen des Instituts fr deutsche Spracheund Literatur, 8 (Berlin: Akademie Verlag, 1956) f, in The Dating of `Beowuif, ed. Frank, Roberta, `Skaldic Verse and the Date of Beowu 123-39 by Colin Chase (Toronto: University of Toronto Press, 1981), pp. Norroenum: Norse Studies Frank, Roberta, `Snorri and the Mead of Poetry', in Speculum in Memory of Gabriel Turville-Petre, ed. by Ursula Dronke and others (Odense: Odense University Press, 1981), pp. I55-70 Frank, Roberta, `Did Anglo-Saxon Audiences Have a Skaldic Tooth?, in AngloConquest, ScandinavianEngland: Norse-EnglishRelationsin thePeriod before the ed. by John D. Niles and Mark Amodio, Old English Colloquium Series,4 (Lanham: University Pressof America, 1989),PP-53-68 Frank, Roberta, `The Ideal of Men Dying with their Lord: Anachronism or Nouvelle in Honour of in Northern Europefoo-. r5oo: Essays Vague?,in Peopleand Places Peter HayesSawyer,ed. by Ian Wood and Niels Lund (Woodbridge: Boydell, 1991, pp. 95-iO6 (Cambridge, Toughy Art Medieval Monstrous Races in The Block, John Friedman, and MA: Harvard University Press,1981)
Saxo Models', Classical in Ingellus its Lay `The Karsten, Friis-Jensen, and of Grammaticus: A Medieval Author betweenNorse and Latin Culture, ed. by Karsten Friis-Jensen (Copenhagen: Museum Tusculanum Press, i98i), pp. 65-77

Latin Norse between Author A Medieval Grammaticus: Saxo Karsten, and Friis-Jensen, ed., Press, Tusculanum Museum (Copenhagen: Culture i98i)

254

Passages Friis-Jensen, Karsten, Saxo GrammaticusasLatin Poet: Studiesin the Verse of the `Gesta Danorum Analecta Romana Instituti Danici, supplementum 14 (Rome: , Bretschneider, 1987) Fuchs, Karlheinz, and others, eds, Die Alamannen (Stuttgart: Theiss, zooi)
Gaimster, Mrit, Vendel Period Bracteateson Gotland: On the Significance of Germanic Art, Acta Archaeologica Lundensia, series in octavo 2-7(Lund: Almqvist & Wiksell, 1998)

Gardner, Gerald B., Witchcraft Today (London: Rider, 1954) Gardner, G. B., The Meaning of Witchcraft (London: Aquarian Press,1959)

Gatch, Milton McC., Preachingand Theologyin Anglo-SaxonEngland: fEc and Wulfstan (Toronto and Buffalo: University of Toronto Press,1977) Gelling, Margaret, `Further Thoughts on PaganPlace-Names',in Otium et Negotium: by Olof Feilitzen, Science Library Presented Studiesin Onomatology to ed. von and Folke Sandgren,Acta Bibliothecae Regiae Stockholmiensis, 16 (Stockholm: Norstedt, 1973),pp. 109-28 England, History Past: Place-Names Gelling, Margaret, Signposts to the 2nd edn of and the (Chichester: Phillimore, 1988) dei Zulpich" Geuenich, Dieter, ed., Die Franken und die Alemannenbiszur "Schlacht (4,96/97)(Berlin: de Gruyter, 1998) Goffart, Walter, The Narrators of Barbarian History: AD yfo-Soo: Jordanes,Gregoryof Press, University (Princeton: Princeton Deacon Paul Bede, Tours, 1988) the and Words German High Old Studies Four Semantic Lord: Carolingian The H., Green, D. on (Cambridge: Cambridge University Press,1965) Research Criticism Essays in Elegies: New English Old The Green, Martin, ed., and (Cranbury, NJ: Associated University Presses,1983) by Ellen Hofineyer, by Germanen, der Religion Kultur trans. Grnbech, Wilhelm, ed. und Verlagsanstalt,1937) Otto Hfler (Hamburg: hanseatische Denkmler Die O. Murdoch, Brian poetischen Groseclose,J. Sidney, and althochdeutschen (Stuttgart: Metzler, 1976) Grammaticus: A Saxo in Grammaticus', Saxo Sources Icelandic `The Gubnason, Bjarni, of Friis-Jensen Karsten by Culture, Latin Norse between Author Medieval ed. and (Copenhagen: Museum Tusculanum Press, 1981), PP.79-93.

255

Haderlein, Konrad, `Celtic Roots: Vernacular Terminology and PaganRitual in Carl'omann'sDraft Capitulary of A. D. 743, Codex Vat. Pal. Lat. y 9, The Canadian journal of Irish Studies,18(i99z), 1-30 Hadley, Dawn M., and Julian D. Richards, eds, Cultures in Contact: Scandinavian Settlementin England in theNinth and Tenth Centuries,Studies in the Early Middle Ages, z (Turnhout: Brepols, Zooo) Hald, Kr., `The Cult of Odin in Danish Place-Names',in Early English and Norse Hugh Smith Studies:Presented in Honour of his Sixtieth Birthday, ed. by Arthur to Browne and Peter Foote (London: Methuen, 1963),pp. 99-109 Hantl, Otto, Der UrglaubeAlteuropas:Die Edda als Schlssel Steinzeit, zur Verffentlichungen aus Hochschule, Wissenschaft und Forschung, 9 (Tbingen: Grabert, 1983)
Harrison, Kenneth, `Woden', in Famulus Christi: Essays in Commemoration of the Thirteenth Centenary of the Birth of the VenerableBede, ed. by Gerald Bonner (London: SPCK, 1976), pp. 351-S6 Hartman, Sven S., ed., Syncretism: Based on PapersRead at the Symposium on Cultural Contact, Meeting of Religions, Syncretism Held at Abo on the 8th-loth of September, Wiksell, (Stockholm: Almqvist Donneriani Aboensis, & Instituti Scripta 3 1966, 1969)

Laicite Serie "Recherches", Parapsychologie, Hasquin, Herve, ed., Magie, Sorcellerie, (Brussels: Editions de l'Universite de Bruxelles, [no date])
Hauck, Karl, `The Literature of House and Kindred Associated with Medieval Noble Families, Illustrated from Eleventh- and Twelfth-Century Satires on the France Nobility, in The Medieval Nobility: Studies on the Ruling Classes and of Germany from the Sixth to the Twelfth Century, ed. and trans. by Timothy Reuter, Europe in the Middle Ages: Selected Studies, 14 (Amsterdam: North-Holland Publishing Company, 1978),pp. 6i-85

der Bildinhalten den Anteil besonderer `Fnens Karl, Hauck, an (Zur Ikonologie der Goldbrakteaten, Brakteaten vlkerwanderungszeitlichen XLIX)', Frhmittelalterliche Studien, z6 (1992), io6-48 Samleren, (Haslev: Myter Oldnordiske Virkelighed: Anden Skygger Hedeager, Lotte, of en 1997)
Scandinavia during Authority in Political A Passport Art: `Myth to Hedeager, Lotte, and History, Studies Archaeology in Anglo-Saxon in Period', Migration and ro: the Tania by September York, Sachsensymposium, 1996, ed. Papersfrom the 47h Committee for Oxford University (Oxford: Griffiths David Dickinson and Archaeology, 1999), PP. 151-56

256

Religionsgeschichte, GermanischeBibliothek, i. S. Helm, Karl, Altgermanische z; also ReligionswissenschaftlicheBibliothek, S, z vols [vol z in two parts] (Heidelberg: Carl Winter, 1913-1953) Helm, Karl, Wodan: Ausbreitungund Wanderungseines Kultes, GiessenerBeitrge zur deutschen Philologie, 85 (Giessen:W. Schmitz, 1946) Herren, Michael W., `The Transmission and Reception of Graeco-Roman Mythology in Anglo-Saxon England 670-800', Anglo-SaxonEngland, 27 (1998),87-103 Herren, Michael W., `Literary and GlossarialEvidence for the Study of Classical Mythology in Ireland A. D. 6oo-8oo', in Text and Gloss:Studiesin Insular ConradHelen Joseph Donovan by Learning and Literature Presented Pheifer, to ed.
O'Briain, Anne Marie D'Arcy and John Scattergood (Dublin: Four Courts Press, 1999), PP. 49-67 Heyerdahl, Thor and Per Lilliestrm, takten p Odin: Pd Sporet av Or Fortid (Oslo: Stenersen, zoos)

Hill, Thomas D., `Woden as"Ninth Father": Numerical Patterning in Some Old English Royal Genealogies',in Germania: ComparativeStudiesin the Old Craig T. G. Calder by Daniel Literatures, Germanic Languages and ed. and Christy (Cambridge: Brewer, 1988), pp. 161-74
Halvorsen Fjeld Eyvind f Festskr Eyvindarbk: Finn, Hodnebo, 4. mat til and others, eds, litteraturvitenskap, for Institutt (Oslo: 1992) nordistikk og 1992 Diesterweg, M.: Moritz (Frankfurt Germanen der Geheimbnde Kultische Hfler, Otto, a. 1934) Hollis, Stephanie, `Napier XLII (1995)) 4z3-z6 42, Queries, V, Notes Homily Wulfstan's 24o, n. s. and and

Religious Symposium Read Papers Based Ecstasy: Holm, Nils G., ed., Religious on at the on Almqvist (Stockholm: Abo, August Finland, on the z6th-28th of Held at Ecstasy 1981 & Wiksell, i98z) Altertumskunde, der Reallexikon vols i- , 2nd edn Hoops, Johannes, ed., germanischen (Berlin: Walter de Gruyter, 1973-)
History, Archaeology Studies in Anglo-Saxon S and John, Eric, `The Point of Woden', (1992), 127-34

Power Topographies in Rhijn, Carine Theuws Frans of eds, van with Jong, Mayke de, and World, (Leiden: Roman 6 Transformation The Ages, Middle the Early of the Brill, zoos)

257

Jung, C. G., `Wotan', trans. by Barbara Hannah, in C. G. Jung, Essays Contemporary on Events, trans. by Elizabeth Welsh, BarbaraHannah and Mary Briner (London: Kegan Paul, 1947),pp. 1-i6 Contemporary Events, Jung, C. G., Essays by Elizabeth Welsh, BarbaraHannah trans. on (London: Briner Kegan Mary Paul, and 1947) Kamp, N., and J. Wollasch, eds, Tradition als historische Kraft(Berlin: de Gruyter, 1982)

Keller, Hagen, and Nikolaus Staubach, eds, Iconologia Sacra: Mythos, Bildkunst und Dichtung in der Religions- und Sozialgeschichte Alteuropas : Festschriftfr Karl Hauck Geburtstag, Arbeiten Frhmittelalterforschung, (Berlin: Walter de zum 7f zur 13 Gruyter, 1994) Ker, N. R., Catalogue of Manuscripts Containing Anglo-Saxon (Oxford: Clarendon Press, 1957)

Kiernan, Kevin S., `Beowulf and the `Beowui Manuscript (New Brunswick: Rutgers University Press, 1981; repr. 1984) f Kiernan, Kevin S., `The Eleventh-Century Origin of Beowulfand the Beowu f, ed. by Colin Chase (Toronto: University Manuscript', in The Dating of `Beowu Press, Toronto 1981), pp. 9-22 of
Anteckningar till Edda och Skaldediktning, S vols Kock, Ernst A., Notationes Norrcenee: (Lund: Gleerup and Leipzig: Harrassowitz, 1923-1944)

Kramer, Evert, Ingrid Stoumann and Andrew Greg, eds, Kings of the North Sea,AD ZsoSyo(Newcastle: Tyne and Wear Museums, zooo)
Krogh, Steffen, `Zur Sprache des "Schsischen Taufgelbnisses"', Zeitschrift fair deutsches Altertum, 124 1995), 143-ISO

Anhngern `Neues Brakteaten in Axboe, Morten Peder, Jan Lamm, und zu and Schweden',FrhmittelalterlicheStudien, 23 (1989) 453-77
Wessex', Studi Monstrorum Liber f Aldhelm, `Beowu Lapidge, Michael, the and Medievali, 3rd series, z3:1 (198z), I5I-9z (woo), Anglo-Saxon England, Beowulf, Archetype `The Lapidge, Michael, 29 5-41 of Relations Late Antiquity Roman Foreign in Frontiers: Information D., Lee, A. and

(Cambridge: Cambridge University Press,1993)

258

Lehmann, Winfred P., `The Grouping of the Germanic Languages', in Ancient IndoEuropean Dialects: Proceedingsof the Conferenceon Indo-European Linguistics Held California, Los University Angeles, April by Henrik the of at 2f z7 1963, ed. Birnbaum and Jaan Puhvel (Berkeley: University of California Press, 1966), pp. 13`27

Levison, Wilhelm, England and the Continent in theEighth Century (Oxford: Oxford University Press, 1946)
Lewis, Charlton T., and Charles Short, A Latin Dictionary (Oxford: Oxford University Press, 1879; repr. 1933)

Lindow, John, `Norse Mythology and Northumbria: Methodological Notes', in AngloScandinavianEngland: Norse-EnglishRelationsin thePeriod before Conquest, the ed. by John D. Niles and Mark Amodio, Old English Colloquium Series,4 (Lanham: University Pressof America, 1989),pp. 25-40 Liuzza, Roy Michael, `Anglo-Saxon Prognostics in Context: A Surveyand Handlist of Manuscripts', Anglo-SaxonEngland, 30 (zoos), 181-230
Looijenga, Tineke, and Arend Quak, eds, Frisian Runes and Neighbouring Traditions: Proceedingsof the First International Symposiumon Frisian Runes at the Fries Museum, Leeuwarden, 26-29 January 1994, Amsterdamer Beitrge zur lteren Germanistik, 45 (Amsterdam: Rodopi, 1996) Lucy, Sam, and Andrew Reynolds, eds, Burial in Early Medieval England and Wales, The Society for Medieval Archaeology Monograph Series, 17 (London: The Society for Medieval Archaeology, zooz)

Lutz, Angelika, `A thelweard's Chroniconand Old English Poetry', Anglo-SaxonEngland,

2-9(2-000)) 177-2I4

Magnus, Bente, `The Firebed of the Serpent: Myth and Religion in the Migration Period Mirrored through Some Golden Objects', in The Transformation of the Roman World AD 4oo-goo, ed. by Leslie Webster and Michelle Brown (Berkeley: University of California Press,1997),pp. 194-202 Integration Heer ihre im `Alemannen Max, Martin, und rmischen - eine verpate (496/97), bis "Schlacht dei Zlpich" Alemannen die Franken Die in Folgen', zur und de Gruyter, 1998) (Berlin: Geuenich Dieter by PP-407-22 ed. Seen Dusburg', by Peter Prussians in Frontiers: `Mental I., Vera Matuzova, von as by Alan Frontier V. Murray Baltic Conversion Crusadeand zzyo-Ifoo, ed. on the (Aldershot: Ashgate, zoos), pp. 253-59

259

McKinnell, John, `Norse Mythology and Northumbria: A Response',in AngloScandinavianEngland: Norse-EnglishRelations in thePeriod before Conquest, the ed. by John D. Niles and Mark Amodio, Old English Colloquium Series, 4 (Lanham: University Pressof America, 1989), pp" 41-5Z
McKitterick, Rosamond, The Frankish Church and the Carolingian Reforms, 789-89y, Royal Historical Society Studies in History (London: Royal Historical Society, 1977)

Meaney, A. L., `Woden in England: A Reconsideration the Evidence', Folklore, of 77

(1966),
IO5-I5

Meaney, Audrey L., `Bede and Anglo-Saxon Paganism',Parergon, (1985), n.s. 3 I-29 Moltke, Erik, Runesand Their Origin: Denmark and Elsewhere, Peter G. by Foote trans. (Copenhagen: National Museum, 1985) Motz, Lotte, The King the Championand the Sorcerer: A Study in GermanicMyth, Studia Medievalia Septentrionalia, i (Vienna: Fassbaender, 1996)
Moulton, William G., `Mutual Intelligibility among Speakersof Early Germanic Dialects', in Germania: Comparative Studies in the Old Germanic Languagesand Literatures, ed. by Daniel G. Calder and T. Craig Christy (Cambridge: Brewer, 1988), pp. 9-28 Murray, Alan V., ed., Crusade and Conversion on the Baltic Frontier m5o-iyoo (Aldershot: Ashgate, zoos)

Myhre, Bjorn, `Germanic Kingdoms Bordering on Two Empires', in Kings of theNorth Kramer, Ingrid Stoumann Evert Greg by Andrew Sea,AD 25-o-85-o, ed. and (Newcastle: Tyne and Wear Museums, Zooo), pp. 41-54 Nelson, Janet L., "`... sicut olim gens Francorum nunc gensAnglorum": Fulk's Letter ... Studies Wise: Honour Aled in Janet Bately in Revisited', Alfred the of to on the by Jane Roberts fh Birthday, Janet Nelson L. Sixty-F her Occasion ed. and with of Malcolm Godden (Cambridge: Brewer, 1997))pp. 135-44 Newton, Sam, The Origins of Beowulf and thePre-Viking Kingdom of East Anglia (Cambridge: Brewer, 1993) Scandinavia: Studies Germanic Language in Runic Early The Frede, Hans Nielsen, of (Heidelberg: Carl Winter, Zooo) Dialect Geography
Anglo-Scandinavian Brunanburh', England: in in Technique D., `Skaldic Niles, John Conquest, before John Period by D. Niles in Relations Norse-English the the ed. and (Lanham: Series, University Colloquium Press English Old Amodio, Mark 4 of America, 1989), pp. 69-78

260

Niles, John D., and Mark Amodio, eds,Anglo-ScandinavianEngland: Norse-English Relations in the Period beforethe Conquest,Old English Colloquium Series, 4 (Lanham: University Pressof America, 1989) Ninck, Martin, Wodan und germanischer Schicksalsglaube (Jena:Eugen Diederichs, 1935) Nordland, Odd, `Vaihall and Helgafell: Syncretistic Traits of the Old Norse Religion', in Syncretism:Basedon PapersRead at the Symposium Cultural Contact, Meeting on of Held at Abo on the &b-loth of September, Religions,Syncretism Sven S. by 1966, ed. Hartman, Scripta Instituti Donneriani Aboensis, 3 (Stockholm: Almqvist & Wiksell, 1969), pp. 66-99 North, Richard, Heathen Godsin Old EnglishLiterature, Cambridge Studies in AngloSaxon England, zz (Cambridge: Cambridge University Press,1997) Noy, David, Foreignersat Rome: Citizensand Strangers (London: Duckworth, zooo) O'Flynn, John Michael, Generalissimos University Western (Alberta: Roman Empire of the Press, Alberta 1983) of
Verse, Old English Transitional Literacy O'Keeffe, Katherine O'Brien, Visible Song. in " Cambridge Studies in Anglo-Saxon England, 4 (Cambridge: Cambridge University Press, 199o) Olsen, Magnus, Minner om Guderne og Deres Dyrkelse i Norske Stedsnavn(Kristiania: Steensball, 1923 offprint from P. A. Munch, Norrmne Gude- og Heltesagn]) Olsen, Magnus, Farms and Fanes of Ancient Norway: The Place-Names of a Country Discussedin their Bearings on Social and Religious History, trans. by Th. Gleditsch, Instituttet for Sammenlignende Kulturforskning, Serie A: Forelesninger, 9 (Oslo: Aschehoug; Leipzig: Harrassowitz; Paris: Champion; London: Williams & Norgate; Cambridge, MA: Harvard University Press, 19z8) (Newton David Charles, Abbot: Anglo-Saxons & Religions Rites R., Gale Owen, of the and 1981) (London: British Myths Memorials Records, Vikings: Chronicles Page, R. I., in and of the Museum Press, 1995)

Sjtiu Jakobi helgaar Kristjnsson, Jonas G., Einar Petursson, ritger6ir eds, and Magnussonar Stofnun Arna (Reykjavik: Rit jli Benediktssyni, iz r977, zo. Islandi, r The Transformation Roman States', in Successor Barbarian the of Pohl, Walter, `The (Berkeley: Brown Michelle Webster Leslie by and World AD 4oo-goo, ed. Press, California University of 1997),PP-33-47

261

Prokosch, E., A Comparative Germanic Grammar (Philadelphia: Linguistic Society of America, 1939) Quast, Dieter, `Opferpltze und heidnische Gtter: Vorchristlicher Kult', in Die Alamannen, ed. by Karlheinz Fuchs and others (Stuttgart: Theiss, Zooi), pp. 43340

Quast, Dieter, `Vom Einzelgrab zum Friedhof. Beginn der Reihengrbersitteim 5. Jahrhundert', in Die Alamannen, ed. by Karlheinz Fuchs and others (Stuttgart: Theiss, 2ooI), pp. 171-90 Raby, F. J. E., A History of Christian-Latin Poetry: From theBeginnings Close to the of the Middle Ages(Oxford: Clarendon Press,'9z7) Rausing, Gad, `Barbarian Mercenaries or Roman Citizens?, Fornvnnen,82 (1987),126-3I
Reuter, Timothy, ed. and trans., The Medieval Nobility: Studies on the Ruling Classes of France and Germany from the Sixth to the Twelfth Century, Europe in the Middle Ages: Selected Studies, 14 (Amsterdam: North-Holland Publishing Company, 1978) Reynolds, Andrew, `Burials, Boundaries and Charters in Anglo-Saxon England: A Reassessment', in Burial in Early Medieval England and Wales, ed. by Sam Lucy Series, Monograph for Archaeology Society Medieval The Reynolds, Andrew and Medieval for Archaeology, The Society (London: zooz), pp. 171-94 17

Roberts, Jane, and Janet L. Nelson with Malcolm Godden, eds,A red the Wise: Studiesin (Cambridge: Sixty-Fifth Birthday her Honour ofJanet Bately on the Occasion of Brewer, 1997)
Rohr, Christian, `Hagiographie als historische Quelle: Ereignisgeschichte und Wunderberichte in der Vita Columbani des Jonas von Bobbio', Mitteilungen des (1995), Geschichtsforschung, Instituts fair sterreichische 103 229-64 Forschungen Bereich im Fragen in Heliandsprache', die `ber Erik, Rooth, und und Frings Ar Theodor Festgabe Philologie: der Umkreis zum 7o. germanischen Sprache Literatur, fr deutsche Instituts des 8 Verffentlichungen Geburtstag, und (Berlin: Akademie Verlag, 1956), pp. 40-79

Skygger Anden Virkelighed, Hedeager, Lotte Clunies, of en Ross, Margaret review of

Alv(ssmdl, 8 1998), 113-15

Cult Woden for AngloPoetry from in Evidence England: the of a Ryan, J. S., `Othin in (1963), Folklore, England', Saxon 460-80 74 Science Library Onomatology Studies in Negotium: Otium and Folke, Sandgren, et ed., Stockholmiensis, Regiae Bibliothecae Acta Feilitzen, Olof i6 Presented to von (Stockholm: Norstedt, 1973)

262

Sawyer, Birgit and Peter Sawyer, Medieval Scandinavia: From Conversion to Reformation, Series, (Minneapolis: Nordic University The Minnesota Boo-ifoo, 17 of circa Press, 1993)

Sawyer,Birgit, The VikingAge Rune-Stones:Customand Commemorationin Early Medieval Scandinavia(Oxford: Oxford University Press,zooo) Aufstze See,Klaus von, `Christliche Skaldendichtung', in Edda, Saga,Skaldendichtung: Mittelalters, Literatur des by Klaus See, von ed. zur skandinavischen SkandinavistischeArbeiten, 6 (Heidelberg: Carl Winter, 1981), pp. 389-4o5 [originally published in Studien zur christlichen Dichtung der Nordgermanen,roooLange, Wolfgang Palaestra, (Gttingen: by Vandenhoeck & zzi rzoo, ed. Ruprecht, I9S8)] des Literatur Aufstze See,Klaus von, ed., Edda, Saga,Skaldendichtung. " zur skandinavischen Mittelalters, SkandinavistischeArbeiten, 6 (Heidelberg: Carl Winter, 1981)
See, Klaus von, `Hskarla hupt. Nochmals zum Alter der Bjarkaml', in Speculum Norroenum: Norse Studies in Memory of Gabriel Turville-Petre, ed. by Ursula Dronke and others (Odense: Odense University Press, 1981),pp. 421-31

Shaw, Philip A., `Miracle as Magic: Hagiographic Sourcesfor a Group of Norse Mythographic Motifs?, in GermaniaLatin V Miracles and theMiraculous in Medieval Latin and GermanicLiterature (forthcoming)
(Cambridge: Hall by Angela Mythology, Northern Dictionary A Simek, Rudolf, trans. of Brewer, 1993)

AngloPapers British Academy Genealogies', in Royal `Anglo-Saxon Sisam, Kenneth, on Press, University Oxford (Oxford: Stanley G. E. by Saxon England, ed. 1990), pp.
145-Z04

Saxostudier, Placering', in Genremaessige Danorums `Gesta ed. Skovgaard-Petersen,Inge, Tusculanum, Museum (Copenhagen: Grxcolatina, Opuscula Boserup, by Ivan z
1975)) pp. 20-27

First Three Books Study A in Byzantium: Way `The the of to Skovgaard-Petersen,Inge, Author Medieval between A Grammaticus: Saxo in Denmark', Saxo'sHistory of (Copenhagen: Museum Friis-Jensen Karsten by Culture, Norse and Latin ed. Tusculanum Press, i98i), pp. 111-33 Old Norse Names', in Religion Change `The and and the of Sorensen,John Kousgrd, Symposium Papers Read Based Place-Names: Cultic at the on Finnish Religionsand Cultic Place-Names Nordic Times Old in Religions Between and on Encounters on Scripta by Tore Ahlbck, August Abo, Finland, on the 19th-2sst 1987,ed. of Held at Wiksell, Almqvist (Stockholm: & Aboensis, 199o), pp. 13 Instituti Donneriani
394-403

263

Southern, R. W., Western Views of Islam in the Middle Ages(Cambridge, MA: Harvard University Press, 196z) Stanley, E. G., ed., British Academy Papers on Anglo-Saxon England (Oxford: Oxford University Press, 199o) Starkey, Kathryn, `Imagining an Early Odin: Gold Bracteates as Visual Evidence?, Scandinavian Studies, 7! (1999)) 373-92-

Stern, Henri, Le Calendrier de 354:Etude sur sontexte et ses illustrations,Institut Francais d'Archeologie de Beyrouth: Bibliotheque Archeologique et Historique, SS(Paris: Imprimerie Nationale, 1953)
Stoklund, Marie, `The Ribe Cranium Inscription and the Scandinavian Transition to the Younger Reduced Futhark', in Frisian Runes and Neighbouring Traditions: Proceedingsof the First International Symposiumon Frisian Runes at the Fries Museum, Leeuwarden, 26-29 January 1994, ed. by Tineke Looijenga and Arend Quak, Amsterdamer Beitrge zur lteren Germanistik, 45 (Amsterdam: Rodopi, 1996), pp. i99-109 Storms, G., `How Did the Dene and the GeatasGet into Beowufi', English Studies, 8o 6999)) 46-49

Stroheker, Karl Friedrich, Germanentumund Sptantike,ed. by Olof Gigon, Die Bibliothek der alten Welt (Zrich: Artemis, 1965)
Strm, Ake V., `Berserker und Erzbischof: Bedeutung und Entwicklung des Religious Ecstasy:Based on PapersRead at the in Berserkerbegriffes', altnordischen Abo, August Finland, Held Symposiumon Religious Ecstasy the 26th-28th of on at . (Stockholm: Almgvist & Wiksell, 1982), pp. 178-5 Holm G. Nils by. 1981,ed.

from Homilies Catholic jElfric's Exploitation The Source: the Swan, Mary, `iElfric as of University doctoral (unpublished Centuries' Twelfth thesis, Late Tenth to of Leeds, 1993) Century, Twelfth Old English in Rewriting Treharne, M. the Swan, Mary, and Elaine ed., (Cambridge: Cambridge England, Anglo-Saxon in Studies Cambridge 30 University Press,zooo)
Gender Writing England', in Anglo-Saxon in Veronica and Swan, Mary, `Remembering Texts, Middle English by Old Approaches Literature: Medieval to Genre in and ed. Brewer, (Cambridge: Treharne Elaine zooz), pp. 919-39 Gaul Germany, Coloured Pottery from Dark Fine Wares: and Symonds, R. P., Rhenish (Oxford: Oxford Monograph for Archaeology, Committee 23 Oxford University for Archaeology, Committee 1992) University

264

Thompson, E. A., The Huns, rev. by Peter Heather, The Peoplesof Europe (Oxford: Blackwell, 1996)
Thorpe, Benjamin, Northern Mythology: Comprising the Principal Popular Traditions and Superstitions of Scandinavia, North Germany and the Netherlands, (London: 3 vols Lumley, I85I-I85z)

Townend, Matthew, `Viking Age England as a Bilingual Society', in Cultures in Contact: ScandinavianSettlementin England in theNinth and Tenth Centuries,ed. by Dawn M. Hadley and Julian D. Richards, Studies in the Early Middle Ages, z (Turnhout: Brepols, 2,0oo),pp. 89-105 Townend, Matthew, `Contextualizing the Knutsdrapur: Skaldic Praise-Poetry at the Court of Cnut', Anglo-SaxonEngland, 30 (zoos), 145-79 Toynbee, Jocelyn M. C., Roman Medallions, Numismatic Studies, s (New York: The American Numismatic Society, 1986)
Treharne, Elaine, ed., Writing Gender and Genre in Medieval Literature: Approachesto Old and Middle English Texts (Cambridge: Brewer, zooz) Tripp, Raymond P., `Odin's Powers and the Old English Elegies', in The Old English Elegies: New Essaysin Criticism and Research,ed. by Martin Green (Cranbury, NJ: Associated University Presses,1983),pp. 57-68 Turville-Petre, E. O. G., Myth and Religion of the North: The Religion ofAncient Scandinavia (London: Weidenfeld and Nicolson, 1964)

Turville-Petre, E. O. G., ScaldicPoetry (Oxford: Oxford University Press, 1976)

Turville-Petre, Joan, `Illustrations of Woden and his sons in English genealogical (1988), Queries, Notes 158-59 233, n. s. 35 and manuscripts',

Urbanczyk, Przemyslaw, `The Lower Vistula Area as a "Region of Power" and its Continental Contacts', in Topographiesof Power in the Early Middle Ages,ed. by Mayke de Jong and Frans Theuws with Carine van Rhijn, The Transformation Brill, (Leiden: World, Roman 6 zooi), pp. so9-32 of the Othin Especially His Relation in Agricultural Study Contributions de, to Vries, Jan to the of Practices in Modern Popular Lore, Folklore Fellows Communications, 94 (Helsinki: Societas Scientiarum Fennica, 1931)

Inhoud, Nederlandsche Bijdragen Mythologischen Skaldenkenningen De de, Vries, J. op met (Haarlem: Linguistiek, H. Philologie D. Germaansche Gebied het 4 en van Tjeenk Willink, 1934)
Grundriss der Philologie, Religionsgeschichte, germanischen Vries, Jan de, Altgermanische de Gruyter, 1935-1937) (Berlin: iz, z vols

265

Vries, Jan de, Altnordisches Wrterbuch,2nd edn (Leiden: Brill, 1977) etymologisches Waas, Manfred, Germanenim Dienst Chr., A. Habelts im rmischen 4. n. Dissertationsdrucke: Reihe alte Geschichte, (Bonn: Rudolf Habelt, 3 1961) Wallace-Hadrill, J. M., The Long-Haired Kings Other History Studies Frankish in and (London: Methuen, 1962)
Wallace-Hadrill, J. M., The Frankish Church (Oxford: Oxford University Press, 1983; repr. 1985)

Watts, Duncan J., Small Worlds: The Dynamicsof Networksbetween Order and Randomness (Princeton: Princeton University Press,1999) Wawn, Andrew, The Vikingsand the Victorians:Inventing the Old North in NineteenthCentury Britain (Cambridge: Brewer, zooo)

Webster, Leslie, and Michelle Brown, eds, The Transformation of the Roman World AD (Berkeley: University California Press, 400-900 of 1997)
Wessen, Elias, `Schwedische Ortsnamen und altnordische Mythologie', Acta Phslologica Scandinavica: Tidsskrftfor Nordisk Sprog forskning, 4 1929-3a), 97-I! Whitelock, Dorothy, The Audience of `Beowulf (Oxford: Clarendon Press, i95i; repr.

1967)

Wilson, David, Anglo-SaxonPaganism(London: Routledge, 199z) Winterbottom, M., `The Style of Aethelweard', Medium zEvum, 36 (1967), io9-i8 Wolfram, Herwig, History of the Goths,trans. by Thomas J. Dunlap, 2nd edn (Berkeley: University of California Press,1988) Wood, Ian, `The Franks and Sutton Hoo', in Peopleand Placesin Northern EuropefooSawyer, by Ian Wood Niels Lund Peter Hayes Honour Essays in ed. and of rfoo: (Woodbridge: Boydell, 1991), pp. 1-i4 Northern Places Europe in Essays in Wood, Ian, and Niels Lund, eds, People and yoo-rf-oo: Honour of Peter HayesSawyer(Woodbridge: Boydell, i99i)
Wood, Ian N., `Fredegar's Fables', VerffentlichungendesInstituts fr sterreichischen (1994), Geschichtsforschung, 359-66 32Wood, Ian, ed., Franks and Alamanni in the Merovingian Period: An Ethnographic (Woodbridge: Archaeoethnology, Boydell, Historical Studies in Perspective, 3 1998) Dying Lord in Germania Men in Ideal `The their the Woolf, Rosemary, with the and of (1976), England, 63-81 Anglo-Saxon Battle of Maldon', 5