It has by now been amply demonstrated that in England, English persisted as a literary language across the Norman Conquest and throughout the twelfth century.Footnote 1 The works of Ælfric of Eynsham (c. 955–c. 1010) in particular have been central in illustrating the continued copying and use of Old English.Footnote 2 However, they also provide evidence for the importance of English in another sphere: as part of a multilingual Anglo-Scandinavian literary community during the twelfth and early thirteenth centuries. An Old Norse translation of Ælfric’s sermon De falsis diis is known to survive in the early fourteenth-century Icelandic manuscript compilation Hauksbók, along with other material partially translated from Ælfric’s sermon De auguriis. In his useful facing-page edition of the Old English and Old Norse versions of De falsis diis, John Frankis takes these translations to be isolated efforts completed in England, since he assumes that Old English would have had little reach in Scandinavia.Footnote 3 Yet it is not just that the translation of De falsis diis fits with other evidence for extensive English influence on Old Norse homily-writing amid larger Anglo-Scandinavian Latin exchanges.Footnote 4 The sermon’s euhemerizing account of the Classical gods, whom Ælfric equates to their Norse equivalents, must also be seen alongside the translation into Old Norse of Old English vernacular genealogies including pre-Christian Germanic gods in the early thirteenth century. This indicates that English likely had a more international reach than is commonly assumed, and that it played a sustained part in medieval North Sea literature alongside Latin and, later, Old French.
In the context of the Danelaw in England, Matthew Townend has established that Old English and Old Norse were probably at least partly mutually intelligible between the ninth and eleventh centuries.Footnote 5 Yet despite the well-known links between English and Old Norse homily-writing, as well as extensive evidence for the presence of Latin manuscripts from England in Norway and Iceland,Footnote 6 scholars have so far only identified few and relatively isolated pieces of evidence for wider knowledge of English in medieval Scandinavia.Footnote 7 Indeed, it has become a critical commonplace to remark that ‘English is only spoken in England’,Footnote 8 and that ‘English-language literature did not travel’.Footnote 9 This is of course not entirely inaccurate. It is notably reflected in a recent study on the survival of medieval vernacular manuscript texts, which shows that English-language manuscripts tend to follow a much more insular preservation pattern than manuscripts containing French, German, or Dutch, with surviving Old and Middle English manuscripts in Europe almost exclusively concentrated in British repositories.Footnote 10
All this has significantly complicated attempts to retrace the transmission of Ælfric’s De falsis diis into Old Norse. The translation survives in a single copy as part of Hauksbók, compiled by and for the Icelander Haukr Erlendsson (c. 1260–1334), where it is titled ‘vm þat huaðan otru hofst’ (‘on the origin of false belief’).Footnote 11 The same part of the manuscript also contains another section on pagan superstitions that includes translations from Ælfric’s De auguriis. Footnote 12 In view of Hauksbók’s Icelandic provenance, Arnold Taylor assumed that the translation must have been completed in Iceland,Footnote 13 although the relevant part of the manuscript is written in a Norwegian hand and was probably copied in Norway during Haukr’s stay there between 1302 and 1310.Footnote 14 Due to the lack of evidence for the required knowledge of English and availability of English-language manuscripts in Scandinavia, however, Frankis instead attributes the translation to ‘presumably a cleric resident in England’.Footnote 15 Since the Old English De falsis diis and De auguriis appear together in several surviving manuscripts, Frankis naturally concludes that they must have been translated together.Footnote 16 The same applies to the two Old Norse texts on the phoenix demonstrably based on the Old English Prose Phoenix, neither of which is found in Hauksbók:Footnote 17 since the Prose Phoenix sometimes circulated with copies of Ælfric’s homilies, Frankis characterizes this too as part of the same isolated translation effort from a single putative manuscript.Footnote 18 It is only when compared to further evidence for English-Norse vernacular contact that the Old Norse translation of De falsis diis takes its place in a larger, multilingual Anglo-Scandinavian literary community.
The Old English De falsis diis
It is worth noting that Ælfric’s De falsis diis is an interesting, if not peculiar, choice for a sermon to be translated into Old Norse. The sermon was probably written between 993 and 998, during Ælfric’s time at Cerne Abbey.Footnote 19 It begins with a section drawing largely on Martin of Braga’s sixth-century De correctione rusticorum,Footnote 20 recounting the origins of the worship of the Classical gods Saturn, Jupiter, Juno, Minerva, Venus, Mars and Mercury. Ælfric’s account provides a euhemerizing explanation for pagan beliefs, which is to say that it describes these gods as powerful historical figures that were eventually worshipped as gods: ‘Ðas manfullan men wæron þa mærostan godas þe þa hæðenan wurðodon’ (230–2; ‘these wicked men were the most famous gods that the heathens worshipped’). But this is only a relatively minor part of the sermon. Ultimately, De falsis diis is centrally concerned with opposing these and other false gods to the true Christian God, especially through a lengthy later section on the biblical prophet Daniel,Footnote 21 who explicitly expresses this contrast: ‘Ic nelle wurðian þa geworhtan godas ac ic geleue on ðone lyfigende God’ (686–9; ‘I will not honour the man-made gods, but I believe in the living God’). This contrast is even encoded into the language of the sermon: as N. M. Robinson points out, Ælfric reserves the verb form lifiende (‘living’) exclusively for the Christian God, as opposed to the form libbende. Footnote 22 In this way, throughout the sermon, Ælfric draws a sharp distinction between various false pagan gods, including long-dead humans, inanimate idols, or a dragon, and the eternally living Christian God.
Yet the aspect of De falsis diis that has generally attracted the most attention is the way in which Ælfric adapts the euhemerization of Classical gods from Martin of Braga’s account. Most notably, he pointedly adds comments specifically equating different Classical gods to their Norse equivalents.Footnote 23 Thus, he notes that Jupiter ‘hatte Þor betwux sumum ðeodum, þone þa Deniscan leoda lufiað swiðost’ (241–4; ‘was called Þórr among some peoples, and him the Danish people love most strongly’). Of Mercury, Ælfric says that ‘Ðes god wæs arwurðe betwux eallum hæðenum, and he is Oðon gehaten oðrum naman on Denisc’ (270–3; ‘this god was venerated among all heathens, and he is called Óðinn by another name in Danish’), and of Venus that she is called ‘Fricg on Denisc’ (326; ‘Frigg in Danish’). John C. Pope notes that these comments ‘reveal neither intimate knowledge of the forms of worship nor much fear of their being adopted’.Footnote 24 Instead, these seem to be topical references made in the context of renewed Viking raids in the 990s.Footnote 25 Whether this description of Scandinavian paganism was by then still fully accurate is questionable, since the Christianization of Denmark and Norway was well underway by the late tenth century,Footnote 26 but Ælfric seems determined to depict paganism as an external rather than internal threat by purposefully aligning pagan beliefs with the foreign invaders.
At the same time, Ælfric shows a ‘deliberate reticence’ to remind the English of their own pagan past, avoiding any mention of the equivalent pre-Christian English gods Þunor, Woden, or Frig.Footnote 27 While he adopts Martin of Braga’s description of how pagans named the days of the week after their gods, he only lists ‘Sunandæg’ (310; ‘Sunday’) and ‘Monandæg’ (311; ‘Monday’) as being named after the sun and the moon. Mercury and Jupiter, however, are linked to the ‘feorðe dæg’ (316; ‘fourth day’) and the ‘fiftan dæg’ (319; ‘fifth day’) instead, without any mention of the English names for Wodnesdæg (‘Wednesday’) or Þunresdæg (‘Thursday’).Footnote 28 It is of course possible that this is due to a fear of spreading superstitions that may no longer have been widely known in England.Footnote 29 But as David F. Johnson points out, Woden also features prominently as an ancestor of English kings in early medieval genealogies, notably those included in the Anglo-Saxon Chronicle, so that presenting him as a pagan god would also have been politically inexpedient.Footnote 30 In short, Ælfric pursues a dual strategy in De falsis diis: he both obscures pre-Christian English paganism and displaces it onto contemporaneous Scandinavian raiders.
As a result, De falsis diis also becomes a targeted ‘attack on specifically Norse paganism’, whether real or imagined, in the wake of recent Viking raids.Footnote 31 This is even heightened by repeated references, in a later passage on the war between the Israelites and the Philistines, to the ‘fif burga’ (458; ‘five boroughs’) of the Philistines, which seem to allude to the five boroughs of the Danelaw.Footnote 32 The phrase also features prominently in an earlier poem, included in four manuscripts of the Anglo-Saxon Chronicle, on King Edmund’s Capture of the Five Boroughs, or ‘Burga fife’ (5b; ‘five boroughs’), from Viking invaders in 942.Footnote 33 This poem also frames Edmund’s reconquest in starkly religious terms, as a rescue of the Christian inhabitants of the Danelaw, who were:
Ælfric’s reference to the five boroughs of the Philistines thus implicitly aligns the English with the Israelites and the Vikings with their pagan adversaries.Footnote 34
So strong is the anti-Scandinavian dimension of De falsis diis that it becomes the main feature of a later adaptation of the sermon by Wulfstan of York (d. 1023). Wulfstan’s De falsis deis for instance carefully reproduces Ælfric’s assertion that Mercury ‘is Oðon gehaten oðrum naman on Denisce wisan’ (72; ‘is called Óðinn by another name in the Danish manner’),Footnote 35 but otherwise heavily abbreviates the sermon and omits most of the later material to place the focus squarely on the condemnation of the Norse gods.Footnote 36 Wulfstan also includes further criticism of Danish beliefs, taken from additions to a later revision of Ælfric’s sermon that appear in some manuscripts of De falsis diis:Footnote 37
Nu secgað sume þa Denisce men on heora gedwylde þæt se Iouis wære þe hy Þor hatað, Mercuries sunu, þe hi Oðon namiað, ac hi nabbað na riht, forðan þe we rædað on bocum, ge on hæþenum ge on Cristenum, þæt se hetula Iouis to soðan is Saturnes sunu. (72–7)
(Now some Danish men say in their error that this Jupiter, whom they call Þórr, was Mercury’s son, whom they name Óðinn, but they are not right, because we read in books, both heathen and Christian, that the hateful Jupiter is in truth Saturn’s son.)
Even among pagan beliefs, Norse religion is thus presented as not only theologically but also historically incorrect, and therefore inferior even to Classical paganism. Writing under an even more direct threat of Viking invasions,Footnote 38 Wulfstan evidently recognized the nationalist, anti-Scandinavian potential of Ælfric’s De falsis diis and capitalized on it.
The Old Norse Translation
And yet, Ælfric’s De falsis diis clearly also found fertile ground in Scandinavia. The date of the Old Norse translation itself is difficult to determine. Frankis concludes that the translator was likely working with a text ‘that contained some readings peculiar to twelfth-century copies of the homily but did not correspond exactly to any one extant manuscript’, and he eventually suggests a date ‘probably in the second half’ of the twelfth century.Footnote 39 Beyond that, since the Old Norse translation only survives in an early fourteenth-century manuscript as part of a section compiling various encyclopaedic materials,Footnote 40 there must have been several stages of copying and possible revision as part of the transmission process.Footnote 41 Yet that does not mean that the translation does not show what kind of interest De falsis diis might have held for Scandinavian audiences. Frankis characterizes the translated text as ‘less a target than an accidental outcome’, but he is quick to ascribe any deviation from the Old English text to a later reviser instead of the translator.Footnote 42 Medieval translations, however, rarely aim to reproduce their source text exactly, and instead usually provide some level of interpretation and adaptation.Footnote 43
For comparison, the Old Norse adaptations from Ælfric’s De auguriis are a very clear example of this process. It has long been accepted that a short sermon in Hauksbók on superstitious practices and rituals partially draws on Ælfric’s De auguriis,Footnote 44 although the Old Norse adaptor ‘carefully changed and adapted his material for a different audience situated in different cultural circumstances’.Footnote 45 The Old Norse sermon begins with an attribution to ‘Hin helgi byskup er heitir Augustinus’ (‘the holy bishop called Augustine’),Footnote 46 reprising Ælfric’s own reference to the Latin De auguriis, then ascribed to Augustine of Hippo but now attributed to Caesarius of Arles.Footnote 47 Importantly, however, Ælfric augments his Latin source with the example of an obscure ritual practised by ‘sume geƿitlease ƿíf’ (148; ‘some witless women’), who ‘teoð heora cild þurh ða eorðan’ (149; ‘draw their children through the earth’). This example is distinctive of Ælfric’s version of the sermon, so that the Old Norse text’s reference to women who draw their children ‘i giognum iorð’ (‘through the earth’) makes clear its dependence on the Old English.Footnote 48 Yet the Old Norse adaptor in turn also augments Ælfric’s sermon. Thus, Ælfric’s description of men bringing offerings ‘to eorðfæstum stane · ⁊ eac to treoƿum’ (130; ‘to an earthfast stone and also to trees’) is transformed into an account of women who dedicate these offerings to ‘land vettum’ (‘spirits of the land’).Footnote 49 Since the belief in such landvættir in Norway was still explicitly forbidden by the Younger Gulaþing Law of 1267,Footnote 50 this addition appears designed to add local relevance to the sermon.
Otherwise, the Old Norse text drastically abbreviates the theological discussion in Ælfric’s De auguriis and ‘adds a completely different ending’ that explains the symbolic meaning of the cross for what seems to be a broader, less theologically informed audience.Footnote 51 The subject matter is common and is notably also taken up in an Old Norse sermon that occurs in both the Old Norwegian and Old Icelandic Homily Books.Footnote 52 Yet even this new ending takes up a brief comment in the Old English sermon, immediately preceding the example of women drawing their children through the earth, that ‘se reða deofol ƿearð þurh ða rode ofer-sƿiðed · | ⁊ heo is ure sige-beacn ongean þone sceoccan á’ (146–7; ‘the wicked devil was overcome through the cross, and it is forever our sign of victory against the fiend’). This clearly provided an impetus for the Old Norse text’s concluding explanation of the cross, which begins with an account of how Christ ‘bant diofulenn’ (‘bound the devil’) with the cross ‘oc bauð oss með þui sigr marke at veria oss dioflum oc illum vettum’ (‘and bade us with this sign of victory to defend ourselves against devils and evil spirits’).Footnote 53 It is certainly not necessary to suggest, as Frankis does, that a full Old Norse translation of De auguriis must have existed, which was then reworked by a later reviser.Footnote 54 Instead, the Old Norse text proceeds with the Old English De auguriis just as Frankis himself describes Ælfric’s own practice in reworking Caesarius’ Latin sermon: it ‘translates or paraphrases some passages, omits others, occasionally reorders portions and adds original passages or extracts from other sources’.Footnote 55
Of course, the Old Norse version of De falsis diis provides a much more extensive translation of Ælfric’s Old English sermon. But as in the case of De auguriis, it is these larger patterns of reinterpretation that may show in what literary context, and to what effect, De falsis diis was translated. Overall, the style of the Old Norse translation is quite plain and occasionally awkward, with no trace of Ælfric’s typical rhythmical prose style.Footnote 56 Its opening reproduces the Old English address ‘Eala ge gebroðra þa leofostan’ (1–2; ‘oh you most beloved brothers’) as ‘Liufir brœðr’ (1; ‘beloved brothers’), initially suggesting a similar monastic audience, yet such addresses are not necessarily reliable indicators of a sermon’s audience.Footnote 57 At the same time, much like the Old Norse sermon drawing on De auguriis, the translation of De falsis diis clearly shows significantly less interest in complex theological matters. Ælfric opens with an extended discussion of the Holy Trinity, who ‘synd ðry on naman, Fæder and Sunu and Halig Gast’ (40–1; ‘are three in name, Father and Son and Holy Ghost’), before moving on to the Creation and Fall of man. In contrast, the Old Norse text is much less theologically sophisticated,Footnote 58 simply asserting that ‘einn er Guð alz faðer, oc alz er gerande oc valdande oc raðande’ (8–10; ‘there is one God who is father of all, and he creates and rules and commands all’). Equally, the careful linguistic juxtaposition Ælfric establishes between false gods and the eternally living Christian God is lost,Footnote 59 as ‘se lyfigenda God’ (647; ‘the living God’) becomes simply ‘Guð almattegr’ (519–20; ‘God almighty’). The Old Norse sermon is also much shorter than its Old English source, although the loss of the text’s final section may be due, as Frankis suggests, to ‘later abridgement’.Footnote 60 The general effect is thus of a sermon aimed at a less sophisticated, more general audience than Ælfric’s De falsis diis.
From the beginning, this impression is then reinforced by further additions in the Old Norse translation that suggest a lay, or at least more diverse and less celibate, audience. One detail that so far seems to have gone unnoticed is that, in the initial account of the Creation and Fall of man, the translation consistently highlights Eve’s presence where the Old English sermon only mentions Adam. Ælfric thus notes that ‘wunode Adam swa orsorh on blisse’ (64–5; ‘Adam lived carefree in bliss’), but the Old Norse reads: ‘slict lif var Adame skapat, oc þeim hiunum baðom sua’ (27–9; ‘such a life was made for Adam, and for the couple both’). Similarly, the singular verb forms asserting that ‘he agylt hæfde and Godes bebod tobræc’ (86–7; ‘he had sinned and broke God’s commandment’) are adapted into the plural ‘þau hofðu brotet þat’ (55–6; ‘they had broken it’). In this way, the Fall is presented as an experience actively involving, and relevant to, both man and woman, of which the Old Norse translation is also keen to stress the universal consequences in the present day: ‘Mikit gerði Adamr oc þau beði hiun oss ollum’ (65–6; ‘much did Adam, and the couple both, do to us all’).
The translation also adds further contextual material to its account of the Creation and Fall. Taking advantage of the Old English sermon’s mention of the promise of eternal life, the Old Norse text includes a brief reference to the Fall of the angels by adding that Adam and Eve would ‘fylla himinriki þann lut er englar tœmdu firir mikileti sinu’ (21–3; ‘occupy that part of heaven that the angels left empty for their pride’). This addition notably parallels a sermon in the Old Norwegian Homily Book, compiled c. 1200, which probably collects material intended for both clerical instruction and preaching to wider, lay audiences.Footnote 61 In a passage likely adapted from Pirmin’s Scarapsus de singulis libris canonicis,Footnote 62 the Norwegian sermon De ammonitione bona similarly notes that God ‘gerði os or iorðu til þes at vér ſ(cy)ldum þann lut fylla himin-rikis er þæir tømdo er fyr(ſt) vꜵro ænglar en nu ero dioflaʀ. fyrir of-drambe þeirr(a) ok mykil-læte ok ágirnd’ (‘created us from the earth so that we should occupy that part of heaven that those who were angels and are now devils left empty, for their conceit and pride and ambition’).Footnote 63 The similarities in phrasing make it possible that the addition to the Old Norse version of De falsis diis is echoing the Norwegian sermon, but the subject is general enough to make this difficult to establish with any certainty. In any case, the result is a retelling of the Creation and Fall that is made both as applicable and as instructive as possible for a broad, general audience.
Importantly, a similar pattern is consistently evident in the rest of the Old Norse text. The translation is amplified throughout with additional contextual material that provides, as Diane Elizabeth Szurszewski argues, an ‘outline of religious history’, following Augustine’s advice on sermons addressed to unlearned audiences in his De catechizandis rudibus. Footnote 64 Occasionally, these additions seem driven by a need to make sense of ‘an only partially understood text’.Footnote 65 The Old English De falsis diis opens the topic of pagan worship with the assertion that ‘ne ræde we on bocum þæt man arærde hæðengild on eallum þan fyrste ær Noes flode (140–3; ‘we do not read in books that heathen worship was practised in all the time before Noah’s flood’). The Old Norse, on the other hand, ends up saying the complete opposite, because the translator has moved the negative statement into the subclause: ‘Sva segia oss heilagar bœcr at engi maðr skal blota heiðnar vettir firir þui at þat gerðu menn firir Noa floð’ (93–6; ‘holy books tell us that no man may worship heathen spirits because this is what men did before Noah’s flood’). This provides an immediate opportunity for the Old Norse text to expand on Noah’s flood by causally linking it to pagan worship: ‘en af þui varð su hin micla floð’ (97–8; ‘and therefore the great flood happened’).
But the Old Norse translation also generally takes the opportunity to amplify its source with fuller retellings of biblical history at points where no such confusion arises. To the Old English sermon’s statement that ‘entas worhton þone wundorlican stypel æfter Noes flode’ (144–6; ‘giants built the wondrous tower after Noah’s flood’), the Old Norse adds an explanation briefly summarizing the story of the Tower of Babel: ‘þeir vildu með þui koma i himinriki. En þa sa Drottenn var mikileti þeira’ (112–14; ‘they wished to get to heaven with it. And then our Lord saw their pride’). Later, while the Old English sermon asserts that the Ark of the Covenant contains ‘Aarones gyrd, þæs ærestan biscopas’ (404–5; ‘the rod of Aaron, the first bishop’), the Old Norse text instead calls it the rod of Moses, which makes possible a further digression on the crossing of the Red Sea, where ‘þeim vendi laust Moyses a Hafet Rauða þa er Farao konungr for eftir þeim með sinn her’ (305–8; ‘Moses used this rod on the Red Sea when King Pharaoh went after them with his army’). The Old Norse even adds that the Ark contains manna, which Ælfric does not mention, as an opportunity to bring up the Israelites’ exile for ‘fiorirtigir vetra i œyðimorc’ (327–8; ‘forty years in the wilderness’). Frankis ascribes these additions to a reviser working on the basis of the Bible,Footnote 66 but as Taylor points out, they ‘could all be general reminiscences from the author’s own knowledge of the scriptures’.Footnote 67 Most importantly, these amplifications otherwise perfectly align with the general translation strategy of providing a much more detailed overview of religious history in which to situate the matter of pagan worship. Jonas Wellendorf goes so far as to argue that the Old Norse translation of De falsis diis is ‘transformed into a historical text’ rather than a sermon.Footnote 68 This is also likely why this text was later included in Hauksbók in the first place, where its Norwegian scribe placed it in a broadly historical sequence of encyclopaedic and historical materials, often drawing on the Etymologiae of Isidore of Seville, and which include the stories of Noah and Moses.Footnote 69
Even the end of the Old Norse translation, which Frankis characterizes as an ‘independent conclusion’ written by a reviser for the abridgement of the text in Hauksbók,Footnote 70 seems to follow this historicizing impulse. It seizes on Ælfric’s assertion that ‘Fela we mihton secgan be swylcum leasum godum’ (918–19; ‘much could we say about such false gods’), but then adds: ‘en nu er þat lanct oc leiðent at [segia] huessu illfus þau voro’ (727–30; ‘but now it would be long and tiring to tell about how ill-willed they were’). In a manner familiar from other amplifications of the text, the conclusion expands on the Old English allusion to the Six Ages of the World, that ‘ure Hælend Crist com to ðissere worulde on ðære sixtan ylde’ (928–30; ‘our saviour Christ came to this world in the sixth age’), to add ‘þeirri sem verðr, en fimm voro aðr gengnar’ (745–6; ‘which is now current, and five had gone before’) before listing the other five Ages. Whether this conclusion is an independent addition to the translation or not, it clearly builds on the larger framing of the Old Norse translation of De falsis diis as an exploration of religious history.
And this is the framework within which the translation into Old Norse of Ælfric’s attack on Scandinavian paganism makes the most sense. Indeed, the translation takes the Old English euhemerization of the Norse gods even further, removing any traces of the fantastic or marvellous to further historicize them. Whereas Ælfric’s De falsis diis has people begin to worship ‘mislice entas and men’ (197–8; ‘various giants and men’), the Old Norse text specifies that they only ‘blotaðu menn’ (153; ‘they worshipped men’). It also further clarifies that these men were only worshipped ‘siðan er þeir voro dauðir’ (155–6; ‘after they were dead’), a point that is repeated when the translation specifies of Mars that they ‘blotaðu hann dauðan’ (200; ‘worshipped him when he was dead’). In this way, the Old Norse text effectively absolves these historical characters of complicity in their own worship. At the same time, the translation obscures other details. Ælfric introduces Saturn as living ‘on ðan iglande Creta’ (204; ‘on the island of Crete’), a detail he adds to Martin of Braga’s sermon,Footnote 71 but the Old Norse text instead places Saturn ‘i œy nokorre’ (161; ‘on a certain island’). This deliberate vagueness reduces the foreignizing effect of the euhemerization. In the terms established by Lawrence Venuti, this is a thoroughly domesticating translation,Footnote 72 which produces a historical account that seamlessly blends into Scandinavian history.
Indeed, the Old Norse translation of De falsis diis also reinforces the identification between Classical gods and Norse gods. It follows a domesticating translation strategy that aims ‘to bring back a cultural other as the recognizable, the familiar, even the same’,Footnote 73 in that it reclaims the pagan gods that Ælfric presents as manifestations of the cultural other as part of Scandinavian history. This begins with its introduction of Jupiter, ‘er sumir menn kalla Þor; en sa var allr einn’ (190–2; ‘whom some men call Þórr, and he was the same one’). The translation also reproduces Ælfric’s assertion that Mercury ‘het Oðenn a donsku’ (217–18; ‘is called Óðinn in Danish’). This is repeated at a later mention of Mercury, of whom the translation adds: ‘þann er ver kollum Oðenn’ (250–1; ‘whom we call Óðinn’). Again, Frankis attributes this amplification to a later reviser,Footnote 74 but again this fits perfectly with the Old Norse translator’s general practice, as this deployment of the first-person plural firmly establishes Óðinn’s common cultural relevance. What the Old Norse translation of De falsis diis produces is thus an account of religious history fully integrated with the pre-Christian Scandinavian past, which as Wellendorf has it ‘is better situated among historical than exhortative texts’.Footnote 75 By reclaiming Ælfric’s euhemerizing account of the Norse gods, the Old Norse text can present pre-Christian religious history from a distinctly Scandinavian perspective.
This perspective also fits with the possible translation errors visible in the Old Norse version of De falsis diis, which Frankis notes are ‘suggestive of a native speaker’ of Old Norse.Footnote 76 In one instance, in the later episode on the prophet Daniel, Daniel’s Old English greeting to the king ‘Þu leofa cyning, leofa ðu on ecnysse’ (614–16; ‘you beloved king, may you live forever’) is condensed into ‘Lifir þu konungr’ (494–5; ‘may you live, king’). The Old Norse translation is not wholly inaccurate, but it is likely prompted by a misreading of the adjective leof (‘beloved’) as the imperative form of the verb libban (‘live’). Leof is of course also part of the standard homiletic address in Old English,Footnote 77 within which it appears – and is correctly translated – at the beginning of De falsis diis, which could even indicate a translator not overly familiar with Old English homiletic conventions. In another instance, the Old English sermon states that the lions to which Daniel is thrown were previously fed ‘twa sceap to bigleofan and twegen leapas’ (859–60; ‘two sheep for subsistence and two corpses’). The Old Norse translation renders this as ‘tua laupa brauðs oc tiu sauði til fœslu’ (660–2; ‘two baskets of bread and ten sheep for food’), based on a mistranslation of the figurative use of the Old English leap (‘basket’, ‘corpse’) as the Old Norse laupr (‘basket’).Footnote 78 The Old Norse term has no such figurative use and therefore prompts the translator to supply the incongruous mention of bread in order to make sense of the presence of baskets. In this case, it is possible that the rare figurative use of leap in English may have been obscured by language change, but again the translator’s handling of the text suggests that ‘English was not his first language’.Footnote 79 Indeed, the translator, though inventive, may not have been particularly practiced. Overall, Frankis sees a distinction between the plain and occasionally awkward translations from the Old English and the more fluid style visible in independent additions to the sermon.Footnote 80 Instead of oddly localized instances of later revision, however, this is just as likely to reflect a certain uneasiness with the task of translation, which disappears when the Norse-speaking translator can independently expand on the source text.
On this basis, the translation could have been produced by one of the many Scandinavians who visited England in the twelfth century. Anglo-Scandinavian contacts remained ‘plentiful’ until the late thirteenth century,Footnote 81 and this demonstrably resulted in some literary exchange. Eysteinn Erlendsson (d. 1188), the Norwegian Archbishop of Niðaróss, stayed at Bury St Edmunds and Lincoln during his exile between 1180 and 1183.Footnote 82 The oldest surviving manuscript of his Latin life of Saint Óláfr Haraldsson originates from Fountains Abbey, and it is likely that he both brought Latin documents with him to England and took liturgical texts back with him to Norway.Footnote 83 There are further records of a noticeable Norwegian presence in English monasteries in the twelfth century,Footnote 84 as well as of shorter visits by Norwegian clergymen as envoys or merchants.Footnote 85 Similarly, two successive Icelandic bishops of Skálholt, Þorlákr Þórhallsson and Páll Jónsson, studied in Lincoln in the second half of the twelfth century.Footnote 86
But the translation’s characteristics just as easily place it directly in Scandinavia, where it fits into a larger pattern of English influence on religious writing. As Henry Goddard Leach has shown, the Christianization of Norway was powerfully driven by English churchmen, with the English church remaining a dominant influence on Norway until 1290.Footnote 87 Stavanger Cathedral was dedicated to the English saint Swithun and its first bishop Reinaldr (d. 1135) was English.Footnote 88 In the mid-twelfth century, English Cistercians from Fountains and Kirkstead Abbeys founded the Norwegian monasteries of Lyse and Hovedøya, respectively, and both foundations maintained close contacts with England.Footnote 89 In 1194, the Englishman Marteinn (d. 1216) became bishop of Bergen.Footnote 90 All this resulted in significant English influence on Norwegian liturgy and Old Norse homily-writing.Footnote 91 Christopher Abram has identified partial translations from Old English sources, including by Ælfric, in sermons included in the Old Norwegian Homily Book dating back to as early as the late eleventh century.Footnote 92 The late twelfth- or early thirteenth-century scribe responsible for the Old Norwegian Homily Book was himself probably trained in England,Footnote 93 and its production has been connected to Bishop Marteinn of Bergen, who may well have brought English books with him to Norway.Footnote 94 If the Old Norse version of De falsis diis does indeed echo the Old Norwegian Homily Book, it might even be tempting to locate its translation in the same milieu.
At the same time, English influence also clearly extended to Iceland. Two of the earliest missionary bishops in eleventh-century Iceland were likely English, and priests with English or Anglo-Norman names are recorded well into the thirteenth century.Footnote 95 Latin manuscript exemplars from England probably underlie parts of the Old Icelandic Homily Book from c. 1200.Footnote 96 English influence is of course also evident in Denmark, especially from the reign of King Cnut.Footnote 97 A collection of Latin texts assembled by Wulfstan of York, which even includes a short passage in Old English in Wulfstan’s hand, was probably sent to Denmark with Bishop Gerbrand of Roskilde after his consecration at Canterbury in 1022.Footnote 98 Odense Cathedral, a ‘centre of English influence’, was dedicated to Saint Alban, and the Benedictine Abbey of Evesham founded a priory there in 1095.Footnote 99 The English monk Ælnoth, a native of Canterbury, then played an important part in the production of the late eleventh- and early twelfth-century Latin texts on Saint Knútr, collectively known as Odense literature, that seem to follow the model of English royal hagiographies.Footnote 100 The translation of De falsis diis, along with the adaptation of De auguriis, must thus be set in the context of what Abram calls ‘a flow of literary ideas straight across the North Sea’,Footnote 101 as part of which Ælfric’s sermon on false gods was adapted and reframed as an exploration of pre-Christian Scandinavian history.
Old English Genealogies and Anglo-Scandinavian Literary Exchange
In fact, these Anglo-Scandinavian interactions must be placed within an even broader sweep of English literary influence that begins with the earliest writing in Scandinavia. There is extensive evidence for the presence of Latin manuscripts of English origin in Norway, especially between the later twelfth and the early thirteenth centuries, as well as scribes evidently trained in England.Footnote 102 Latin manuscripts of English origin are also recorded in Iceland from the early twelfth century,Footnote 103 and later Icelandic church records show the presence of a large number of manuscripts from England, some of which may even have been in English.Footnote 104 As a result, Old English left a marked trace in Old Norse, both in vocabulary and scribal practice. For instance, the development of the Old Norse term ríta (‘to write’) was likely influenced by its Old English cognate writan, and Old Norse stafróf (‘alphabet’) is a direct loan from Old English stæfræw. Footnote 105 Additionally, both Norwegian and Icelandic scripts were heavily influenced by English practices. The earliest twelfth-century Norwegian scripts adopted many features of Old English vernacular minuscule for writing in Old Norse,Footnote 106 and while Icelandic script was instead adapted from Caroline minuscule, it did early on adopt the Old English rune thorn ⟨þ⟩.Footnote 107 This adoption of runic letters for representing Old Norse sounds is explicitly addressed in the twelfth-century Icelandic First Grammatical Treatise, in which it is justified by comparison with English practices.Footnote 108 Indeed, Ælfric’s own Old English Grammar was known in some form in Iceland. Kari Ellen Gade has identified a late fourteenth- or early fifteenth-century manuscript fragment providing Old Norse explanations for the inflections of the Latin verb amo that is clearly patterned on Ælfric’s Grammar, and she further suggests that Ælfric’s Grammar may have influenced the Third Grammatical Treatise written by the Icelander Ólafr Þórðarson around 1250.Footnote 109
Most importantly, however, it is within this larger Anglo-Scandinavian literary exchange beyond religious writing that another parallel with the Old Norse translation of Ælfric’s De falsis diis emerges, which provides further evidence for the importance of translations from Old English in Scandinavia. Frankis notes the existence of a ‘community of interest in the Scandinavian world in the twelfth and thirteenth centuries’ concerning the euhemerization of pre-Christian gods,Footnote 110 in which De falsis diis naturally finds its place. But another significant part of this literary current is the inclusion of pre-Christian Norse gods in Scandinavian genealogies, a practice that begins in the early thirteenth century but is clearly influenced by earlier English models.Footnote 111
The earliest-known of these thirteenth-century genealogies, identified by Anthony Faulkes, is a set of regnal lists that today only survives in a transcription by the antiquarian Árni Magnússon (1663–1730) of a thirteenth-century manuscript from the library of the Danish scholar P. H. Resen (1625–88), which was destroyed in the Copenhagen Fire of 1728.Footnote 112 It contains lists of English kings, continued to King Henry III’s accession in 1216, which must ultimately draw on a version close to the genealogies that survive in an eleventh-century Old English manuscript: London, British Library, Cotton MS Tiberius B V.Footnote 113 The names of two of the kings in Árni’s transcription, ‘[K]eaƿlim’ and ‘ƿitta’, contain the characteristically English wynn rune ⟨ƿ⟩ to represent ⟨w⟩, which is otherwise represented by ⟨v⟩.Footnote 114 Although ⟨ƿ⟩ was adopted by early Norwegian vernacular scripts and later also appears in Icelandic scripts under Norwegian influence,Footnote 115 this inconsistent usage suggests that the regnal lists originally drew on an Old English exemplar.
Even more significant, however, is the fact that the Old Norse regnal lists begin with the name ‘Sescef’.Footnote 116 Of this name, which also appears in later Old Norse genealogies, Eduard Sievers already pointed out in 1892 that it is clearly a misreading of an Old English genealogy like that in the Tiberius manuscript.Footnote 117 There, the entry ‘Bedƿig. sceafing’ (‘Bedwig, son of Scef’) is followed by a prose paragraph beginning with the sentence ‘Se scef ƿæs noes sunu’ (‘this Scef was Noah’s son’).Footnote 118 The Old Norse adaptor evidently misread the Old English demonstrative pronoun se as part of Scef’s name. This is an understandable error, given that in the Tiberius manuscript, the ‘Se’ that opens the prose paragraph on Scef begins with the enlarged initial otherwise reserved for the first letter of the listed kings’ names. At the same time, one of the Old Norse regnal lists also relatively closely translates a prose paragraph from the Old English genealogy of West Saxon kings, showing that the Old Norse translator must have had a reasonable command of the language and its script. The Old English text in Tiberius states:
Ingeld ƿæs ines broðor ƿestseaxna cyninges. ⁊ he heold rice .vii.⁊xxx ƿintra. ⁊ he getimbrade þœt beorhte mynster æt glæstinga. byrig. ⁊ æfter þam fyrde to sancte petres. ⁊ þær his feorh asealde. ⁊ on sibbe gerest. ⁊ hi begen broðra ƿæron cenredes suna.Footnote 119
(Ingeld was the brother of Ine, king of the West Saxons. And he held the kingship for 37 years. And he built that bright monastery at Glastonbury. And after that he went to St Peter’s and there gave up his life and rests in peace. And both brothers were sons of Cenred.)
In the Old Norse regnal list, this is reproduced as:
Ingeld broðir vestr Saxa k(onongs). hann var k(onongr) xxxvi· hann let gora mustari i Glestinga bvri. siðan foro þeir badir til rvms oc ǫnduðuz þar þessi ero nofn langfeðga þeira.Footnote 120
(Ingeld, brother of the king of the West Saxons. He was king for 36 [years]. He had the church at Glastonbury built. Afterwards they both went to Rome and breathed their last there. These are the names of their ancestors.)
As in the Old Norse translation of De falsis diis, there is some abbreviation and adaptation. There can, however, be no question that the Old Norse is a direct translation of a closely related Old English text.
Finally, and most significantly in regard to De falsis diis, the Old English genealogies in the Tiberius manuscript also list ‘Ƿoden’ among the ancestors of West Saxon kings.Footnote 121 The corresponding Old Norse regnal list, using the same domesticating phrase as the translation of De falsis diis does of Mercury, seizes on this opportunity to link Woden to Óðinn and emphasize his importance to Scandinavian history: ‘Voden þann kollum ver Odinn fra honum eru comnar flestar kononga ęttir i norðr halfu heimsins’ (‘Woden, whom we call Óðinn, from whom are descended most royal families in the northern part of the world’).Footnote 122 This integration of Woden or Óðinn into royal genealogies subsequently proved highly influential in Scandinavia. The tradition of such Scandinavian genealogies is complex, but the Icelander Snorri Sturluson (1179–1241) at least must have drawn on regnal lists very similar to those translated from Old English for both his Heimskringla and his Edda: notably, parts of the genealogies he gives in the prologue to Gylfaginning in the Edda closely follow the translated regnal lists, and he even repeats the same phrase when he includes ‘Voden, þann kǫllum vér Óðin’ (‘Woden, whom we call Óðinn’).Footnote 123 Just as the Old Norse translation of De falsis diis had re-claimed Ælfric’s euhemerization of Norse gods to transform Ælfric’s sermon into an exploration of religious history from a firmly Scandinavian perspective, the Old Norse regnal list appropriates Woden in translation and makes his genealogy part of Scandinavian history.
These translations from Old English then also take their place in a larger, multilingual Anglo-Scandinavian literary exchange. The same kind of domesticating, historicizing translation is more widely found in early thirteenth-century Old Norse translations of Latin texts from England, such as the early thirteenth-century translation of Geoffrey of Monmouth’s Prophetiae Merlini by the Icelandic monk Gunnlaugr Leifsson (d. 1218/19), which presents the prophecies of Merlin in the framework of Icelandic history.Footnote 124 Another example is the Breta sǫgur, an early thirteenth-century translation of Geoffrey’s De gestis Britonum. Footnote 125
Notably, the Breta sǫgur also survive in Hauksbók, immediately following Trójumanna saga, which begins with a euhemerizing prologue equating the Classical gods to their Norse equivalents, though following a different tradition than De falsis diis. Footnote 126 The Breta sǫgur then address pagan gods in their account of the arrival of the Saxons under Hengest to Britain. In the Latin De gestis, Hengest lists among his gods ‘Mercurium, quem Woden lingua nostra appellamus’ (‘Mercury, whom we call Woden in our language’) and notes that a day of the week is named after him.Footnote 127 The Old Norse text in Hauksbók significantly amplifies this passage on Mercury: ‘þann kalla svmir Oðinn ok hafa varir forellrar mikin trvnað a hanvm haft sva ok a Þór ok Ty Frig ok Freyiv’ (‘whom some call Óðinn, and our ancestors have had great faith in him, as well as in Þórr and Týr and Frigg and Freyja’).Footnote 128 The insertion of additional Norse gods makes the episode even more relevant to Scandinavian history, as does the end of the Hauksbók Breta sǫgur, which gives a list of West Saxon kings down to Æthelstan, who ‘fostradi Hakon svn Haralz konvngs harfa(g)ra’ (‘fostered Hákon, son of King Harald fairhair’).Footnote 129 Thus, the history of Britain naturally flows into the history of Norway. This ending, unique to the Hauksbók text, may be a later addition,Footnote 130 but it highlights how fluidly a domesticating and historicizing translation like the Breta sǫgur can fuse the history of Britain and Scandinavia.
These translations can then be set alongside the translation of Ælfric’s De falsis diis and of the Old English genealogy. Together, they attest to the continuity of multilingual Anglo-Scandinavian literary contacts after the Norman Conquest and before the series of Old Norse translations from Anglo-Norman French commissioned by King Hákon IV Hákonarson of Norway (1204–63).Footnote 131 Indeed, this continuity even offers some support for the claim made in the prologue to Olif ok Landres, part of the late thirteenth-century Karlamagnus saga largely translated from Old French, that it was translated from a now lost Scottish source ‘ritaða ok sagða í ensku máli’ (‘written and told in the English language’).Footnote 132 What emerges here is a sustained pattern of literary exchange across the North Sea and as far as Iceland, which does not only include Old Norse, Latin and Old French, but English as well.
Conclusion
These separate instances of translation demonstrate that Old English writings, and Old English manuscripts, clearly still held literary importance in late twelfth- and early thirteenth-century Scandinavia. They also shed additional light on the knowledge of English beyond England, such as what may underlie a well-known comment on the relationship between English and Old Norse in the First Grammatical Treatise. The author of the Treatise invokes the English to justify the adoption of runes and other new letters for Old Norse ‘eptir þeira dæmvm’ (‘following their example’), before adding: ‘ver ervm æinnar tvngo þo at giorz hafí miǫk onnvr tveggía ęða nakkvað bááðar’ (‘we are of one tongue, though one of the two has changed greatly, or both somewhat’).Footnote 133 This comment has given rise to various interpretations: Hreinn Benediktsson dismisses it as probably ‘based on second-hand information’ without any actual knowledge of English,Footnote 134 but it has alternatively been seen as evidence for broader knowledge of English and its similarity to Old Norse in twelfth-century Iceland.Footnote 135 The latter case does not seem nearly so unlikely in view of widespread English influence, along with the translations both of De falsis diis and De auguriis as well as of the Old English genealogy, which showcase the broader literary reach of English in Scandinavia.
Nor can this Anglo-Scandinavian network of literary exchange easily be reduced to single manuscripts, or even single centres of transmission. Abram proposes the Abbey of Bury St Edmunds, which entertained close links with the Norwegian church, as a possible source for the Latin texts underlying the Old Icelandic Homily Book,Footnote 136 and Worcester Cathedral Priory as a centre for the transmission of both the Old English sources of the Old Norwegian Homily Book and De falsis diis. Footnote 137 Lincoln Cathedral has also been noted as a likely centre of transmission of Latin manuscripts to Iceland, including of the sources for the Old Norse Physiologus, Thómas saga erkibyskups, Merlínusspá, and possibly Breta sǫgur. Footnote 138 Meanwhile, the Tiberius manuscript containing the Old English genealogy closest to the Old Norse regnal list was probably copied in Winchester but was kept at Battle Abbey during the twelfth century.Footnote 139 Yet a collection of genealogies similar to that in the Tiberius manuscript also appears in the Textus Roffensis, written between 1122 and 1124 at Rochester Cathedral Priory,Footnote 140 which is coincidentally where the Old English manuscript closest to the Old Norse translations of De falsis diis and De auguriis was copied.Footnote 141 Most of these centres, besides Bury St Edmunds and Lincoln, have no established links to Scandinavia. These cases mirror the difficulties of retracing the transmission of other Latin manuscripts from England to Scandinavia,Footnote 142 but together they showcase the broad nature of Anglo-Scandinavian literary exchange.
The lack of surviving manuscripts containing English in Scandinavia should still not be too surprising. As time went on, they would be much less likely to be preserved if no one could read them anymore.Footnote 143 Árni Magnússon, largely responsible for the extensive preservation of Old Norse manuscripts in Iceland, notably also showed much less interest in Latin manuscripts, resulting in their comparatively poor preservation.Footnote 144 This is relevant insofar as the Tiberius manuscript that includes the Old English genealogy contains a miscellany of items in Old English and Latin,Footnote 145 and the genealogy could have been transmitted to Scandinavia in a similar manuscript compilation. In fact, the same is theoretically possible in the case of De falsis diis, as a brief account of pagan gods excerpted from Ælfric’s De falsis diis was, in the eleventh century, added to a Latin text of the Etymologiae that is now kept in Paris.Footnote 146 Evidently, Old English did occasionally travel alongside Latin texts in manuscripts, as in the case of the Copenhagen Wulfstan collection.Footnote 147 Similarly, an Old English glossed psalter belonging to Gunhild (d. 1087), sister of King Harold Godwinson, was likely taken to Bruges with her after the Norman Conquest.Footnote 148 These are only the faintest traces of what may once have been the reach of English across the medieval North Sea.
It may therefore be time to reassess the perceived insularity of English in the high Middle Ages. The Old Norse translation of Ælfric’s De falsis diis, along with the adaptations from De auguriis, should not be reduced to an isolated enterprise undertaken by ‘one translator in one place’.Footnote 149 Instead, taken together with the broader English influence on Old Norse homily-writing and the use of Old English genealogies in Scandinavia, they become evidence for continuing close Anglo-Scandinavian cultural connections and vernacular literary contacts beyond Latin and Anglo-Norman French. Only then does the international reach of English become visible, as a vernacular playing a small but nevertheless sustained part in a larger, multilingual North Sea literature.
Acknowledgements
This article was written during my tenure of the E. K. Chambers Studentship at Corpus Christi College, Oxford. It has greatly benefitted from discussions at the 2025 Ælfric’s Afterlives conference at Leiden University, as well as at the 2025 ISSEME conference at Heinrich-Heine-Universität Düsseldorf, where my attendance was supported by a TOEBI Conference Grant. I am grateful to the anonymous reviewers for their helpful comments and suggestions; any remaining errors are, of course, my own.
Open access
