How ELVISH WERE THE ALFAR? Terry Gunnell! Var pad pa sem vid lidum Was ic chen that we passed hvert inn { annad { fyrstu each into the other at first en sundur sidan and then apart og loks a dreif and then finally scattered inn into i stokka og steina? the stocks and stones? (Porsteinn fri Hamsi, 1992)" Wilhelm and (especially) Jacob Grimm in the field of Old Norse mythology and folklore, it might be argued that several current misconceptions about the subject area can be traced back to their original ideas. One of Jacob Grimm’s most attractive and (hence) influential arguments was thac there had been a single and consistent early Germanic ‘religion’ which was known throughout the Germanic countries in pre-Christian times, an idealizing notion driven to a considerable degree by a romantic nationalism born largely of Napoleonic humiliation. This position reflects a possibly subconscious Fé all the praise that has rightly been afforded to the pioneering work of " Twould like to express my gratitude to Olga Holownia for a number of discussions on the subject of difar over the years. I have no doubt that these have had more than a little influence on the arguments made in this article. Thanks are also due to Heimir Palsson and Sverrir Témasson for reading the article (and for drawing my attention to the Porsteina fra Hamri poem quoted above); and to Stefan Brink for advice and encouragement in relation to the placename section. Any errors that remain are, of course, the responsibility of the author. in Sefarinn sofandi (Reykjavik: Bunn, 1992), p. 42. * Porsteinn frd Hamsi, “Alfabj6d"m2 Terry Gunnell political agenda which had powerful implications in its own time.’ A second recutrent and essentially romantic idea in the Grimms’ work on mythology and religion was the conviction that the folk beliefs of their day were directly connected to the pagan beliefs of the past. Both of these notions made good sense in the spirit of their times and, no doubt in part out of respect for the reputation and authority accorded to the Grimms, came to be adopted by many later scholars. This approach (and especially firse idea) can be still seen in many overviews of Old Norse mythology or religion which attempt to present a single coherent and unified vision and structure, suggesting that such a structure and vision was known over a wide area of territory (such as most of the Nordic countries) and over a broad period of time.‘ Inevitably, hough, modern research > See Jacob Grimm, Deutsche Mythologie, 4th edn, 3 vols (Berlin: Diimmler, 1875~78),5, pp. 1-10 (‘Einleitung’), and pp. v-xlii (Vorrede’); also Jacob Grimm, Teutonic Mythology, trans. by James Steven Stallybrass, 4 vols (London: Bell, 1882-88), especially 1, 12 (Introduction), and Il, pp. v-lv (Preface to the 2nd Edition). While Grimm notes that ‘Unter allen stimmen des deutschen volks geben sich zahllose abweichungen der mundartkund, denen gleiches recht gebiihre; ebenso sind in dem volksglauben manigfalte unterschiede anzunehmen” (Deutsche Mythologie, ', p. xvii) (‘Among all branches of the Teutonic race there shew themselves innumerable varieties of dialect, each possessing an equal right; so likewise in the people's religion we must presuppose a good many differences’: Teutonic Mythology, il, p. xxii), he regularly returns to the idea of a single faith where ‘Deutschland befinder sich in einer besonderen nicht ungiinstigen mitte’ (Deutsche Mythologie, }, 7) (‘Germany holds a middle place, peculiar to herselfand not unfavourable’: Teutonic Mythology, 1, 8). The idea of unity of is effectively summed up by Grimm's English translator, James Stallybrass, when he + that: ‘Jacob Grimm was pethaps the first man who commanded a wide enough view of the whole field of Teutonic languages and literature to be able to bring into focus ehe scattered facts which show the prevalence of one system of thought among all the Teutonic nations from Iceland to the Danube’: see Stallybrass, ‘Translator’s Preface’, in Teutonic Mythology1, p.v.See also Tom Shippey, ‘A Revolution Reconsidered: Mythography and Mythology in the Nineteenth Century’, and ‘Afterword: A Chair, A Sock, and Language’, in The Shadow- Walkers: Jacob Grimm's Mythology of the Monstrous, ed. by Tom Shippey (Tempe: Arizona Center for Medieval and Renaissance Studies, co-publication with Brepols, 2005), pp. 1-28 (pp. 11-13), and pp. 379-88 (p. 381) respectively. * See Grimm, Deutsche Mythologie, 1, pp. xi-xii and xxiv-xxxv in particular (Teutonic Mythology, tl, pp. xiii-xvi and xxxiv-xlvii). Similar blending of legendary and mythological material is particularly evident in Grimms chapter on "wichte und elbe’ (‘wights and elves’): see Deutiche Mythologie, 1, 363~428, and especially 365-69 (Teutonic Mythology, 11, 439-517, 442-47). 5 Another clear example of the same approach from the nineteenth century is Nikolai Grundwvig, Nordens Mytologi eller Sindbilled-Sprog, historisk-poctisk udviklet og oplyst, 3rd edn (Copenhagen: Schubaches Boghandel, 1870), which attempts to reclaim ‘Germanic’ mythologyHOW ELVISH WERE THE ALFAR: 13 in folkloristics, archaeology, and oral tradition continues to highlight a number of key features in the Grimms’ remarkably influential analysis that still need revisiting and reconfiguring. The present article seeks to make some additional contributions to this re-examination, The proposition that there was ever a unified religion or mythology accepted and known throughout the aiicient Germanic world (or even the Nordic world asawhole) has long been unsustainable, Indeed as the Grimm brothers’ Deutsche Sagen confirms, religious ideas and beliefs in these geographical areas have always varied by time, place, fashion, cultural and social environment, and the general demands of society.‘ Snorri Sturluson’s’ suggestion chat Odinn was the accepted leader of the Nordic pantheon is seriously challenged not only by place name evidence in Norway, Sweden, and Iceland but also by the fact that Freyr is called Freyr (‘Lord’), that bérr has pride of place amongst the gods in both Uppsala’ and Mare,? and that Odinn is not mentioned in the twelfth-century Landndmabok, the early Icelandic Book of Settlements which provides genealogical information and legends about the first settlers. The notion that there was a for the North. The most obvious example of this same method in more recent times must be Georges Dumézil’s Les diewx des Germains (Pais: Presses Universitaires de France, 1959), translated as Gods of the Ancient Norshmen, ed. and trans. by Einar Haugen, Publications of the UCLA Center for the Study of Comparative Folklore and Mythology, 3 (Berkeley: University of California Press, 1973). Here the area of belief is extended still further into the Indo- European world as a whole. (We may note in passing Haugen’s interesting translation of ‘Germains’ into ‘Northmen’.) Even more recently, see, for example, Lotte Motz, The King, The Champion and The Sorcerer: A Study in Germanic Myth (Vienna: Fassbaender, 1996), where regular mention is made of the ‘Germanic pantheon’; and Folke Strém’s otherwise highly useful Nordisk hedendom: Tro och sed i forkristen tid, 2nd edn (Géteborg: Akademieforlaget, 1961), which makes similar assertions about the beliefs of the Nordic area. 5 A similar point has been made independently in two recent articles: see Shippey, ‘Afterword: A Chair, A Sock, and Language’, and ‘Alias Over Habeo: The Elves as a Category Problem’, in The Shadow- Walkers, pp. 157-88 (pp. 184, 187, 382, 388). 7 Of course, there are still questions about whether Snorti personally wrote the Prose Edda alone, or had it compiled or written for him. Nonetheless, following tradition within the field, Iwill continue to accredit the work to Snorti in the pages that follow. * Adam of Bremen, The History of the Archbishops of Hamburg-Bremen, trans, by Francis J. Tschan (New York: Columbia University Press, 1959), pp. 207-08. » Snorri Sturluson, Heimskringla, ed. by Bjarni Adalbjarnarson, fslenzk fornrit, 26, 3 vols (Reykjavik: Hid islenzka forntitafélag, 1941-51), 1, 317. * Icis believed that the first version of Landndmabék was composed by Ari frd6i Porgilsson (possibly with help from others) in the early twelfth century. Nonetheless, the earliest extant4 Terry Gunnell single, unified belief that the world had been created from the giant Ymir (given in VafPridnismdl, stanza 21" and Gylfaginning, chs VIl-VIII") seems contradicted by the statement in Valuspd, stanza 4 that the earth rose from the sea (an idea deftly avoided by Snorti). As John McKinnell has effectively demonstrated in his book Borh One and Many, the extant mythological texts in the Poetic Edda and Prose Edda suggest that there were clearly several different images of Loki over time, and also a variety of accounts of Bérr's fishing trip (some of which ended with Pérr killing the Middle-Earth sea serpent long before Ragnardk).* The range of conflicting myths that must have been in existence within the remarkable multicultural gathering of people that settled in Iceland in the late ninth century is particularly evident in the difficulty that Snorri Sturluson has in trying to construct a single image of the Nordic cosmological world in the Prose Edda; and also in the words of the Icelandic editor of the Sigurdr poems found in the main Codex Regius manuscript of the Poetic Edda, as he attempts to piece together the differing accounts of how Sigurdr Féfnisbani died (in the prose Frd dauéa Sigurbar). Written history demands facts and, ideally, uniformity, whereas oral culture has always tended to be quite satisfied with differing degrees of variation within spoken or sung ‘texts’."* The same can be said for folkloristics. manuscripts (Sturlubdk and a fragment of Melabst), which contain slightly differing texts, come from the late thirteenth century. " Allreferences to Eddic poetry (exceptwhere otherwise stated) are to Eddadigte, ed. by Jon Helgason, Nordisk filologi, serie A: tekster: 1: Voluspa. Havamdl, and edn, 3vols (Copenhagen: Munksgaard, 1964); Il: Gudedigte, 2nd edn (Copenhagen: Munksgaard, 1971); ii: Helzedigte, forse del Copenhagen: Munksgaard, 1961). * Snorti Sturluson, Edda: Prologue and Gylfaginning, ed. by Anthony Faulkes, and edn (London: Viking Society for Northern Research, 2005), pp. 10-12. * John McKinnell, Both One and Many: Essays on Change and Variety in Lase None Heathenism (Rome: Editrice il Calamo, 1994), pp. 13-86. ™ See, among numerous other works, Albert B. Lord, The Singer of Tales (New York Atheneum, 1974); John Miles Foley, How to Read an OralPoem (Urbana: University of Hlinois Press, 2002); Ruth Finnegan, Oral Poetry: Its Nature, Significance and Social Context (Cambridge: Cambridge University Press, 1977); David Buchan, The Ballad and the Folk (East Linton: Tuckwell, 1997); and Isidore Okpewho, Aftican Oral Literature: Backgrounds, Character and Continuity (Bloomington: Indiana University Press, 1992). As these works demonstrate, oral cultures (which naturally vary) tend to have a different view to that of written culturesas to the ‘correctness’ of the precise wording of any given account.smell ven ted dea his ind ver led ‘he the ate has the ms pts che ‘ly, ith da nse HOW ELVISH WERE THE ALFAR? us Icis quite clear thar if the Prose Edda had been written in northern Norway or south-eastern Sweden, it would have been a very different work. It is also quite possible that the worldview that lies behind individual Eddic poems differs greatly. Attempting to recreate a single cosmology from all these poems, as Snorri tries to do, is thus a questionable process. As both Neil Price and Thomas DuBois have argued, it is time we ceased using the Prose Edda as a starting point for neat structural analyses of a set Nordic cosmology (where whole ideas are sometimes based on a single reference) and started thinking instead of more pluralist, changeable ‘belief systems’.'® Along with the problem of relying on inflexible mythological models, we also need to challenge the tendency of scholars to apply a ‘top down’ template to Old Norse patterns of belief, applying modern concepts to chose of earlier times. For many, it is hard to abandon the idea (based largely on later folk beliefs influenced in part by ecclesiastical propaganda) that the jémar must all be malignant, ugly, and stupid, like the trolls of Nordic folklore that replaced them in [ater times. Dvergar are often depicted as bearded and small; while the archetypal valkyrjur ride horses in armour as part of a kind of operatic wild hune; and the Old Nordic difar are seen as being ‘elvish’, something underlined by the way in which they regularly appear in English indices under the heading of ‘elves’."” As I seek to show in the following discussion, the original difar seem to have had very little to do with ‘elves’ as most people saw them before The Lord The worldview of Grimnismdl, where Odinn rules supreme and, as noted, the vertically layered world (see Grémnismdl, stanza 31) seems to be created from the body of che giant Ymir, might be compared to that of Véluspd, where the world initially cises from the sea (stanza 4), or Skirnismal, which implies that Freyt, rather than Odin, views the world from the high-seat Hlidskfalf, and that the world of the jémar exists on the same horizontal plane as that of the gods. % Sec Neil S. Price, The Viking Way: Religion and War in Iron Age Scandinavia, Aun: Archaeological Studies, 31 (Uppsala: Department of Archaeology and Ancient History, 2002), pp. 26, 54-55; and Thomas A. DuBois, Nordic Religions in the Viking Age (Philadelphia: University of Pennsylvania Press, 1999), pp. 7-8; 10-12. See also McKinuell, Both One and Many, pp. 9-10, and 20-26. On the difficulties with teuscing Snorri’s presentation of the mythological world, see Alaric Hall, “The Meanings of Elf and Elves in Medieval England’ (unpublished doctoral dissextation, University of Glasgow, 2004, accessible at [accessed 15 May 2007]), p. 53; and Jon Hnefill Adalsteinsson, ‘Giants and Elves in Mythology and Folktales’, in A Piece of Horse Liver: Myth, Ritual and Folklore in Old Icelandic Sources (Reykjavik: Haskéladtgafan, 1988), pp. 129-41 (pp. 129, 151~32)- ” As pointed out in n. 4 above, the blending of mythological and folkloristic macerial is particularly evident in Jacob Grimm's chapter on ‘elves’ in Deutsche Myshologie.06 Terry Gunnell of the Rings came to be written."" Furthermore, it seems clear that the extant early Nordic sources point to a range of different understandings of difar which varied over time and in accordance with the worldviews of the writers. This is, of course, not surprising when the hear¢ of belief in the original difar seems to have centred in Sweden rather than in England or Iceland (where most written sources originated). The sources also indicate quite clearly a gradual development (largely under the influence of the Church and thirteenth- and fourteenth- century translations of French romances such as Tristrams saga” and Méttuls saga’*) in which the dffar gradually began to be blended with the early landvettir (nature spirits) that appear to have been deeply rooted in the popular consciousness of those living in the Nordic countries from an early point." 8 With regard co this image, see further the images accompanying Andrew Lang's Fairy books, or the early translations of the works of Hans Christian Andersen, See also Tolkien's own complaints about the way ‘fairies’ and ‘elves’ were presented in J. R. R. Tolkien, ‘On Faity Stories’, in his Tree and Leaf (London: Unwin, 1988), pp. 9-73 (pp- 1-14). ° See Riddara sigur (hereafter RS), ed. by Bjarni Vilhjélmsson, 5 vols (Reykjavik: Islendingasagnadtgafan, 1954), 1, 174. The Saga af Tristram ok lend and Méttuls saga are both Arthurian romances, based on French originals, and were translated into Old Norse at around the same time in the mid-thirteenth cencury at the behest of King Hakon of Norway. They thus brought new ideas about the supernatural world into the Nordic countries. Fragments of these sagas from the fourteench and fifteenth centuries exist. However, the main texts of the sagas come from the seventeenth and eighteenth centuries. This might mean that some of the vocabulary found in the texts of these sagas also comes from that time. * See RS, 1, 259. On the dating of the saga and its background, see previous note. ¥ While ‘land spirits’ might be a more direct translation for landvattir, |have made use of the expression ‘nature spirits’ here to stress, first of all, che implied connection between the landscape and these spirits (suggested elsewhere), and secondly to underline a difference between the lendvettir mentioned in Landndmabék and the sagas (see below), and the four figures that are said to protect Iceland against Haraldr Gormsson’s prospective atcack in Heimskringla, These four figures (also found on modern Icelandic coinage) are what most modern Icelanders associate with the word landvettir, even though the original text in which they occur does not actually use this word to describe them. Heimskrixgla simply states that the visiting ‘wise man’ sent by the Danish king ‘sé, a8 fjéll il ok hélar véru fullir af landvértum, sume stért, en sumt smatt’ (‘saw that all the mountains and hills were full of Jandvattir, some big and some small’), It then goes on to talk ofa giant, dragon, eagle, and bull which frighten the visitor away, stressing that the bullis itself followed by landuettir (while the giant is followed by giants, and the eagle by birds, and the dragon by various insects and 5): Snorri Sturluson, Heimskringla, 1, 271. Whether these four beings were originally viewed as being Lendvettir themselves is open to question. Most scholars today sce them as symbols originating in the Christian symbols of the evangelists rather than anything in the local «nell tly ed of we ity Ys ity HOW ELVISH WERE THE ALFAR? 017 The following discussion offers a brief examination of several key features that need to be considered in any discourse of how the scribes, poets, and storytellers of the Nordic countries originally conceived the difar. Though some of the arguments presented echo those recently made by Alatic Hall in his excellent thesis on ‘The Meanings of E/fand Elves in Medieval England’ (2004), and Tom Shippey’s ‘Alias Over Habeo: The Elves As A Category Problem’ (2006), others represent different approaches or concentrate on different information. As will be noted, I deliberately leave references to the Prose Edda until last. Ail too often scholars begin their discussions with Snorti’s classification of the difar, apparently forgetting the fact that he, like Grundtvig and the Grimms, was essentially drawing conclusions from older, varying source materials, and had a particular agenda in mind.” It is evident thar in early twelfth-cencury Iceland (prior to Snorri’s birth), at east, there was still a clear conceptual difference between nature spirits and difar. This can be seen, for example, in Ulffidtsldg? and Landndmabék, which say nothing about difar but make clear reference to both ‘landveettir’ (literally ‘nature spirits’) and ‘bergbuiar’’ (‘rock-inhabiters’), and to che beliefs of some people (mainly in the west of Iceland) to the effect that they ‘deri { hélana’ (‘died folk belief of the time. Iris questionable whether we should base our general understanding of the word landvattir on these figures. Nonetheless, the association of landvattir with ‘fll [...] 0g hélar’ is worth remembering, since it is reiterated elsewhere, and underlines a connection with later folk belief connected with the later “élfar’ and Auldufolk. * The approach is particularly evident in Jacob Grimm's discussion of the difar (see n. 3 above), and that of Nikolai Grundrvig (see Noxdens Myrologi, pp. 229-34), which goes on to suggest (on the basis of later folk belief) that the difar were small, but also that they were connected with angels and water spirits (pp. 230, 234). » See, forexample, Flateyjarbdk, ed. by SigurBur Nordal, 4 vols (Akranes: Flateyjardiegéfan, 1944-45): 1 274-75. Flateyjarbék was written c. 1390. However, the text of Ufjtslig (some of which is believed to come from pre-Christian times) is also contained in the Hauksbék version of Landndmabék written soon after 1300. With regard to the discussions about the dating and nature of the text in Ulfijbrslig, see, for example, Sverrit Témasson, ‘Ulfijstr (Uidljétslog)’, in Reallexikon der Germanischen Altertumskunde Bd. 31, ed. by Heinrich Beck and others (Berlin: de Gruyter, 2006), pp. 404-05. *4 Alll references to Landndmabék are wo Landndmabék, ed. by Jakob Benediktsson, {slenzk fornrit, x (Reykjavik: Hid islenzka forntitafélag, 1968), pp. 29~397. Specific references will be to chapters from the Searlubok MS (S). In this case, see especially Landndmab6k, S 329-30. + See ibid., S 329.u8 Terry Gunnell into the hills’) or into mountains like Helgafell and Mzelifell.* The absence of any use of the word difar to refer to nature spirits can also be seen in the late fourteenth-century manuscript of Pidranda pdttr ok Pérhalls,7 which instead refers to ‘kvikvendi’ (‘creatures’) living in hélar (hills or mounds), and in a translated homily in Hauksbék (written around 1300),"" which talks again of ‘land vertir’ when discussing those beings that received food in piles of rock or under flat stones in return for their support of farms. The account of Egill Skalla-Grimsson’s nfsténg activities in Egils saga” mentions ‘landvettir bar, er land petta byggva’ (‘those nature spirits that populate this land’), and also two verses uttered earlier by Egill which Almqvist sees as originally being associated with this act) a more godly being referred to as a ‘landass’ and “Tandalfe’” Ie seems that it was only in later times that difar started becoming associated with rocks and Aélar by Icelanders.” The Kormédks saga reference to a special blood-sacrifice being made to the difar in a héll as a means of getting assistance *6 See ibid., S 85, $ 110, and S 197. » See Islendinga sigur, ed. by Bragi Halldérsson and others (Reykjavik: Svart 4 hvitu, 1987), pp. 2254-55. The der is also preserved in Flateyjarbok (see above). ** Hauksbo 1892-96), p. 167. » The passage in question describes how Egill curses the king of Norway by placing a horse's head on a pole carved with runes, and by uttering a magical incantation designed to call con the spicits of che country, asking them to reject the king. See Egils saga Skalla-Grimssonar, ed. by Sigurdur Nordal, fslenzk fornrit, 2 (Reykjavik: Hid islenzka fornritafélag, 1933), pp. 163-72. The saga was probably composed in the early thirteenth century, the oldest fragmentary manuscript of the saga coming from the mid thirteenth century: see Lslensk bokmenntasaga, 1, ed. by Bédvar Gudmundsson and others (Reykjavik: Mal og menning, 1993), pp. 87-88. % See Bo Almaqvist, Norrin niddikening: Traditionshistoriska studier i versmagi, I: Nid mot Farstar (Stockholm: Almqvist & Wiksell, 1965), pp. 89-93; and Jon Hnefill Adalsteinsson, Under the Cloak: A Pagan Ritual Turning Point in the Conversion of Iceland, 2nd edn (Reykjavik: Héskeladegafan, 1999), pp. 153-57. Temightbe noted that Jn Hel ees these words as applying to two different beings. ” Gongu-Hrélfi saga (fifteenth-century manuscript), in Fornaldarségur Nordurlanda (hereafter FN), ed. by Gudni Jénsson and Bjarni Vilhjélmsson, 3 vols (Reykjavik: Bokatirgéfan forni, 1943-44), 1, 390-91; and Jarlmanns saga og Hermanns (fifteenth-century manuscript) in RS, V1, 220; of. Saga af Tristram ok lsénd in RS, 1, 174 (regarding the dating of this saga, see n. 20 above). ed. by Finnur Jénsson (Copenhagen: Kongelige nordiske oldskrift-selskab,vell of we ad of or ill er 1B id al ce de HOW ELVISH WERE THE ALFAR: ug with a cure for injury probably represents an earlier stage in the transition of beliefs from active worship or sacrifice to folk belief (as in the account in Hauksbék mentioned above). By the time of Grettis saga Asmundssonar,” Bésa saga oh Herrauds3* and Sigurbar saga pigla,® the transition seems to be nearly complete, with difar now placed alongside lesser ‘éveettir’ (‘evil spirits’), such as ‘moldbiiar’ (‘earth dwellers’), ‘fjandur’ (‘fiends’), ‘bergrisar’ (‘rock giants’), ‘nornir’ (‘witches’), and ‘troll’, Similar uses are found in romances like the Saga af Tristram ok sind,” Méteuls saga,* Elis saga ok Résamundu,® and Samsons saga * See Kormdks saga, ed. by Einar Olafur Sveinsson, Islenzk fornrit, 8 (Reykjavik: Hid ‘slenaka fornritafélag, 1939), p. 288. The earliest complete version of the saga is preserved in Mobdruvallabdk from the mid-fourveenth century: [slensk békmenntasaga, 11, p. 95. » Gretts saga Asmundssonar, ed. by Gudni Jonsson, {slenzk fornrit, 7 (Reykjavik: Hid {slenzka fornritafélag, 1936), p. 204. The saga was probably written in the fourteenth century, although the earliest manuscripts are from the fifteenth century: /slensk bokmenntasaga, Il, p. 144, » FN, 1, 474 (Bfteenth-century manuscript). % RS, 101, um (late founeenth-century manuscript) % Te might be noted that the locally based beliefs from Snefellsnes about difiek (‘defecation’) given in the late fourteenth century Bardar saga Snafellsdss (the earliest manuscript of which comes from the fifteenth century), ed. by Pérhallur Vilmundarson and Bjarni élmsson, fslenzk fornrit, 13 (Reykjavik: Hid islenaka fornritafélag 1991), p. 1; the late thirteenth-century Eyréyggia saga (the earliest manuscript of which was preserved in the fourteenth-century Vatnshyrna manuscript), ed. by Einar Olafur Sveinsson and Matthias Pordarson, Islenzk fornrit, 4 (Reykjavik: Hid islencka fornritafélag, 1935), p. 10; and Landndmabok (S 85), all come from the same area where we find beliefs about forefathers ‘dying into’ hills and mountains (sce above). On one side, they reflect commonly encountered later Nordic beliefs whereby urine and faxces were sometimes put on anchor ropes to dissuade water spirits from climbing into fishing boats; or eyes were washed in urine to break supernaturally created illusions. On another side, they might be seen to be part of an older local belief connected with the forefathers ‘becoming’ difar or local gods, and the need to keep ‘their’ areas unsoiled, to ensure their continuing presence. See, for example, Jén Arnason, hlenzkar pjddsigur og avintjri, 6 vols (Reykjavik: Dj6Bsaga, 195461), 1, 518 and Ill, 396. ” RS, 1, 174. (On the dating of this saga, see n. 19 above.) % RS, 1, 259. (On the dating of this saga, see n. 19 above.) » RS, 1V, 62, 128. (While the saga was translated in the thirteenth century, the oldest extant manuscript comes from the fifteenth century.)120 Terry Gunnell ‘fagra,® and other fornaldarsigur such as Norna-Gests pdter," Gingu-Hrélfi saga,” and Hrélfi saga kraka,” as well as Pidriks saga.” Here the difar are starting to take on their later internationally folldoric ‘elvish’ form, slipping through solid walls, stealing children, luring innocent young men and women off safe moral pathways, and having recurring problems with child-bearing. None of these later works make any mention of practical worship or sacrifice. The idea of making sacrifices to difar (something generally quite different to the approach usually taken to the more lowly landvactir)® is supported by Sighvarr PérBarson’s often quoted reference in Austrfararvisur'® to the Swedish woman in ‘Gautland’, who said that she would not let him and his companions into her abode because her household was conducting a private difablée (‘elf- sacrifice’) in the late autumn (not at jé/, as Strém argues).”” Similar ideas may be reflected in the Flateyjarbék" account of the offerings made by people to the grave mound of King Olafr Gudrddarson ‘Geirstadadlfr’ (‘the Alfr of Geirstadir’), who only gained his by-name after his death, when people started making sacrifices to him for ‘dr’ (‘a good year’). These activities all find parallels in the account in Ynglingasaga” of how the god/king Freyr received offerings in his grave ‘til drs ok fridar’ (‘for a good year and peace’); and perhaps also in Rimbere’s comment from the [ate ninth century about how the Swedes of Birka * RS, 1s, 384 (fourteenth century). * EN, 1, 167 (from Flateyjarbét, c. 1390). * EN, 11, 390-91 (fiftcenth-century manuscripy) © EN, 1, 25-27 (seventeenth-century manuscript). “ Didriks saga af Bern, ed. by Hensik Bertelsen, 2 vols (Copenhagen: Moller, 1905-11), 319. (The saga is believed to have been put together in Norway in the late thirteenth century, with the earliest manuscript dated to much the same period: /slensk bokmenntasaga, It, p. 213.) * See, nonetheless, the account in Hauksbsk mentioned earlier. “6 The Auserferarotsur describe a journey that the poet takes to Sweden on behalf of his king: for this section, see Snorti Sturluson, Heimskringla, 11, 137. See also Hans Kuhn, ‘Philologisches zur altgermanischen religionsgeschichte’, in.Kleine Schrifien, 1V, pp. 223-321 (p. 267). ¥ Seram, Nordisk hedendom, pp. 91.177 * Flateyjarbék, 11, 74~78; Kuhn, ‘Philologisches zur altgermanischen religionsgeschichte’, pp. 268-69. * Snorti Seucluson, Heimskringla, 1, 23-25.vell Is, ral er to sh ns Ie be he ng. he ris in ka HOW ELVISH WERE THE ALFAR’ 121 started worshipping a dead king as a god at the time when his predecessor, Bishop Ansgar, was trying to convert them.” The idea of active worship of figures known as difar (albeit only as supported by the above references) certainly suggests that these beings were viewed by some people as having the power to influence the world around them, almost like gods." As several scholars Rave noted, this idea is supported still further by a range of other early sources, notably the Eddic poems which, far from regarding them as mere landvattir, regularly place the difar alongside the sir gods and the jomar, Havamdl, stanza 159; Grimnismdl, stanza 4; Skirnismdl, stanza 7; and Lokasenna, stanzas 2, 13, and 30, all make use of the oral formula ‘sa ok alfa’ (Asir and difar). Other poems place the difar alongside the sir in lists, as in Hévamdl, stanzas 143 and 160; Skirnismdl, stanzas 17-18 (‘Hyvat er pat élfa / né dsa sona/ né vis[s}a vana’; ‘Is it the son of an diff, As or wise Vanr’?); Prymskvida, stanza 7, and Véluspd, stanza 48 (‘Hvat er med asom? / Hvat er med dlfom?’ in both and translating as ‘How goes with the sir, how goes with the dlfar?’); Féfnismdl, stanza 13; (‘sumar ero éskungar,/ sumar Alfkungar’; ‘some are of the family of the sir, some are of the family of difar); and Sigrdrifumadl, stanza 18 (per ro med dsom,/ per ro med dlfom / sumar med visom vénom’; ‘they are with the Asir, they are with the difar, some are with the wise Vanir). In Hévamddl, stanza 159, the dlfar are clearly listed among the tivar (gods). As Strém, Schjodt, Ellis Davidson and, most recently, Hall and Armann Jakobsson have all noted,® several Eddic poems seem to use the word difar to refer specifically to the Vanir race of gods, particularly Lokasenna, stanzas 2, 13, and 30 (the talk of ‘asa ok alfa,/ er hér inni ero’; ‘those sir and dlfar that are in here’, which underpins the introductory prose statement ‘Mart var par 4sa ok aif’; ‘there were many #sir and difar there’); Hdvamdl, stanza 159; and ® Rimbert, Ansgars liv, trans. into Swedish by Eva Odelman, in Boken om Ansgar (Stockholm: Proprius, 1986), p. 54. » Jacob Grimm observes this feature (see Deutsche Mythologie, 1, 366; and Teutonic ‘Mythology, lt 443), but concentrates on noting parallels with later folklore. The same feature is noted by Shippey, ‘Alias Over Habeo', p.179- ® Strim, Nordisk hedendom, p. 199; Jens Peter Schjodt, ‘Relationen mellem aser og vaner ogdens ideologiske implikationer’, in Nordisk hedendom: Ett Symposium, ed. by Gro Steinsland and others (Odense: Odense universitets forlag, 1991), pp. 303-19 (pp. 306-07); H. R. Ellis Davidson, Myths and Symbol: in Pagan Europe: Early Scandinavian and Celtic Religions (Syracuse: Syracuse University Press, 1988), pp. 105, 173: and, most recently, Hall, “The ‘Meanings of Elfand Elves’, pp. 35~53; and Armann Jakobsson, ‘The Extreme Emotional Life of Vélundr the Elf, Scandinavian Studies, 78 (2006), 233-38.122 Tony Gunnell Grimnismal, all of which refer to the difar but make no use of the word Vanir in lists of beings (in spite of their apparent knowledge of the Vanir gods’ presence). This applies especially to Lokasenna and Grimnismdl. The same idea might also conceivably lie behind the Skdidskaparmd claim that men can be compared to Assir, jétnar, and difar, Wt might also be seen in Egill’s verses uttered in connection with the x/dsténg, where the possibly interchangeable terms ‘landdss’ and ‘landélfy’ are deliberately used for single godlike beings, the first being listed alongside Odin, Freyr, and Njérdr.* Connections becween the Vanir and the alfar are underlined still further by the statement in Grimnismdl, stanza 5, that Freyr received Alfheimr (‘the world of the difar’) as ‘tannfé’ (a ‘tooth-gift’); as well as in the implicit connections between Freyr, his ‘servant /alter-ego Skirnir (lit. ‘Shining’) and the sun (known as ‘Alfrodutt’ (‘the elf wheel’) in Vafpridnismadl, stanza 47, and Skirnismdl, stanza 4); in the associations between the Vanir and whiteness (cf. the description of Heimdallr in Prymskvida, stanza 45); and in the parallels noted earlier between the grave mound worship of Freyr (also known as Ingvi-Freyr) and Olafr Geirstadadlfr, These connections were evidently noted by J. R. R. Tolkien, who seems to have gone out of his way to underline the idea for his ur-mythology that ‘Ingwé’ was ‘the most high lord of all the Elvish race. He [...] sits at the feet of the Powers and all Elves revere his name [...] The Vanyar were his people; they are the Fair Elves’, In Tolkien’s elvish, ‘Vanyar’ means ‘the fair’, something which, in the light of Tolkien’s celebrated philological awareness, might well be meant to reflect the Indo- European word *albh, meaning ‘white’ or brightness, cf. Old English Ailficjne.* ® See Snotti Sturluson, Edda: Skéldskaparmal 1: Introduction, Text and Notes, ed. by Anthony Faulkes (London: Viking Society for Northern Research, 1998), p. 40. See Egils saga, pp. 163-65; Almavist, Norrie middikening, pp. 89-933 and Jén Hnefill ABalsteinsson, Under the Cloak, pp. 53-57, where the suggestion is made that che ‘landalfi’ is actually Freyr. 4]. RR. Tolkien, The Silmarillion (London: George Allen and Unwin, 1977), pp. 52-53- On Ingvi-Freyr’s importance as a god in Anglo-Saxon England, see Richard North, Heathen Gods in Old English Literature (Cambridge: Cambridge University Press, 1997), pp. 26-77, 133-307. On Tolkien’s iacerest in extending the role of this figure in his mythology, see T. A Shippey, The Road to Middle-Earth, new edn (London: Gr&fton 1992), p. 271. As Shippey also notes here (pp. 56-57), Tolkien was nonetheless far from free of Snorri’s influence (or that of Jacob Grimm): while he saw the connections with the Vanir, he also felt bound to follow Snorri’s division of elves into dark and light categories. See Grimm, Deutsche Mythologie, 1, 365-66: and Teutonic Mythologie, l, 442; Hall, “The Meanings of Elf and Elves’, pp. 56-57, 71-76: Kuhn, ‘Philologisches zur altgermanischenHOW ELVISH WERE THE ALFAR? 133, The evidence presented above shows that for many in the Nordic world at the turn of the first millennium the difarand the Vanir appear to have been scen as synonymous. However it is important not to make any broad generalizations about this matter. As noted above, it should come as no surprise that different Eddic poems seem to originate in — and reflect — slightly different belief systems. For example, it is questionable whether Sk/rnismd] originated in an area where Odinn was seen as the highest god, in spite of the reference to Draupnir in stanzas 21~22, and the slightly dubious mention of Freyr illegally sitting in Odinn’s seat in the prose introduction.” Some of the Eddic poems make it clear that, unlike the performer-creators® of the recorded Lokasenna and Grimnismal, the creators of these other poems seem to have seen the difar as having been an individual godly race, different to both the Vanir and the £sir, since they mention all three as existing alongside each other as equals; see Sigrdrifumdl, stanza 18 and Sktrnismdl, stanzas 17~18 (both mentioning ‘visir vanir’ and ‘alfar’), Alvissmdl (throughout), VafBridnismal (which talks of Njérdr returning to the ‘visom vénom’ in stanza 39 but which also mentions the “élfrddull’ in stanza 47), Prymskvida (which mentions first Esir and difar alone as godly powers in stanza teligionsgeschichte’, pp. 266-69; and Shippey, The Road to Middle-Earth, p. 53. It might also be worth remembering that despite their prevalence in early sources, the Eddic difarare never mentioned by name, unlike the sir, jotmar, dvergar, and valkyrjur. Ic seems to be expected that we know who they are. This, to my mind, adds further support for the idea that they were seen as synonymous with the Vanir. See also Grimm, Deutsche Mythologie, 1, 3670, and Teutonic Mythology, N,, 444n, where the suggestion is made that the word Vanir may well carry a meaning related to ‘whiteness’; and Prymskvida, stanza 15, where Heimdallr is described as being both ‘hvitastr dsa’ (‘the whitest of the 4sir’), and one of the Vanir. The poetry says nothing about Freyr usurping Odinn’s seat. The Prose Edda (which seems to lie behind many of the prose comments accompanying the mythological poems in the Eddic manuscripts: see Terry Gunnell, The Origins of Drama in Scandinavia (Cambridge: Brewer, 1995), pp. 223-35) nonetheless goes to some pains to underline the idea that Freyr’s demise at Ragnardk is the result of his ‘immoral’ behaviour described in this mythological account. The Prose Edda stresses that he loses his sword as a result of his lust for the giant's daughter Gerdr, and simultaneously commits an act of hubris by taking Odinn’s seat, which provides him with sight over the world and the vision of the gitl in question. See Snorti Sturluson, Edda: Prologue and Gylfaginning, p. 31. As noted above, the poem says nothing about the seat being Odinn'’s: after all, Freyr is literally “The Lord’. # Many scholars today believe that the poems were to some degree relatively fluid in form, and that they were regularly recreated in performance rather than composed with pen in hand, similar to the works analysed by Lord and Parry in The Singer of Tales: see also Gisli Sigurdsson, ‘Inngangur’, in Eddukvadi, ed. by Gisli Sigurdsson (Reykjavik: Mal og mening, 1998), pp. ix-haii,124 Terry Gunnell 7, and then in stanza 15 describes Heimdallr as one of the Vanir), and Voluspa (which talks of the war between the sir and Vanir in stanza 24, but later refers only to Bsir and differ) Alongside these references, we have Vélundarkvida which, as several scholars have noted, seems to originate in a belief system different to that lying behind the other Eddic poems, not feast because of its mention of swan-maidens, which are referred to in the prose as ‘valkytior’. Here the word ‘Alf’, used to describe ‘Vélundr by the narrator and by Nidudr in stanza ro (‘Alfa li6di’) and stanzas 13 and 32 (‘visi Alfa’), seems first and foremost to refer to ‘otherness’ rather than holiness or common ‘Alf’ associations, but especially the kind of dangerous, supernatural ‘otherness’ that was commonly connected to che Sdmi or ‘Finnat’ in both saga times and in tater folklore. There is no reference to either godly powers (over and above that of flight) or any of the qualities normally associated with the idea of the early landvattir. A further level of meaning running alongside those identified above is that found in the various histories of the early Norwegian kings, in which Alfheimar is not a mythical site," but rather a land ruled by Swedes (or at least non- Norwegians), situated souch-cast of where Oslo is today, and north of Géteborg, between the rivers Gautelfr (Gotilven) and Raumelfr (Glommen).* In those accounts where Alfheimar is mentioned (set in a period priar co the birth of the unifying King Haraldr Adrfagri, ‘the Fine-Haired’), Swedish territory to the south of Raumariki (present day Romerike) is seen as beginning with the contested area of Vingulmérk (around Oslo), and continuing south into Alfheimar, which is situated in present-day Bohustin, but also runs east into 8 The later reference occurs in stanza 48, a strophe which, interestingly, appears in the Codex Regius manuscript directly before the mencion of Freys’s later foe at Ragnardk, the giant Surer. & See, for example, John McKinnell, ‘The Context of Vélundarkvida’, Saga-Book, 23 (1990), 1-27 (especially p. 3), and Jakobsson, “The Extreme Emotional Life’. “ Admittedly there is nothing in Grimnismal to suggest that the reference to Alfheimar being inherited by Freyr — forefather of Swedish kings > should not be taken to refer toa ‘geographically known area that was annexed by the Ynglingar in their move west. © Iris noteworthy that Jacob Grimm makes litle use of this material in his discussion of ‘elves’ in Deutsche Mythologie. On references to this setting, see Ynglinga saga, in Snorti Sturluson, Heimskringla, 1, 79; Ségubrot af fornkonungum, in FN, tl, 133: and the fifteenth- century Borsteins saga Vikingssonar, in FN, 1, 185-86.vell pd os ats id be te 1 HOW ELVISH WERE THE ALFAR? 125 Dalsland and Vastergétland).® It might be noted that while this wooded lowland territory would still have been relatively wet at this time (containing a number of dluleffar or rivers),** it is also the area that contained the greatest concentration of Bronze Age tock carvings related to agricultural fertility, including human figures, boats, and sun images (and a great many so-called a/vkvarnar [lit. ‘elf- |s'], or cup-marks carved into rocks which were still used for local offerings to spirits in later times). There can be little doubt that the people in the early © See Ortnamnen i Alusborgs lin, utgivna av Kungl. Ortnamnskommissionen, ed. by Ivar Lundahl and others, 20 vols (Stockholm: Norstedt, 1906-48), Li, 16-17 (on the area name, Alvsborgs in): 1.i, 33 and Li, 19 (on the use of the word alu for river in this area); Il, 162~63 (on the locally important Géwaaly river); and, 72 (on the present-day village name Alvhem, north of Gateborg — earliest form ‘Alveim’ from 1377: no mention is made here of the saga references). Itis noteworthy that the editors make no attempt to connect the word alv with alv. However, nor do they make any connections with difar, preferring to connect it to a layer of earth beneath the vegetable soil (‘jord under matjorden’), something which, even ifthisis right, cannot have applied to the wider area of Alfheimar as a whole. It is nonetheless interesting to note that Alvhem includes a central farm called ‘Kungsgirden’ (the King’s Farm), and close to this are some gravemounds, one of which is called ‘Kungshégen’ (the King’s Mound). Not far away isa place called Tingberg. All of these details testify to the age and earlier importance of the site. We may also note that nearby is another settlement called Frover (oldest form ‘Frowidh’: 1425), which presumably might have some connection with the god Freyr, for vet was then connected with vé (‘shrine’; although the writers of the relevant place-name survey underline connections between general fertility and the word fri, rather than a reference to the god: see Ortnamnen i Alusborgs lan, ll, 76-77). See also Svensks ornamns-lexikon, ed. by Mats ‘Wahlberg (Uppsala: SOFI, 2003), pp. 21~22, 82; and, concerning the ninth-century kingdom of Alfheimar, Erland Hjarne, ‘Alfheimar’, in Land och Ledung: Ur Erland Hijarnes historiska firfastarskap, 1-11, ed. by Gésta Aqvist, Skrifter utgivna av Instituret for Rattshistorisk forskning, grundat av Gustav och Carin Olin: Rattshistoriske bibliotek, 31-32 (Stockholm: A.-B. Nordiska bokhandelen, 1979-81), i, 29-381. With regard to Icelandic placenames including the element ‘AIF? (apparently including Almannagié at Pingvellie), and referring to spirits (probably in later times), see further Bjarni Einarsson, ‘Vettatri og nokkur islensk Srnefni’, in Melt mdl og form feed: Safn ritgerda eftir Bjarna Einarson gefid st d sfotugsafimali hans 11. april.zy87 (Reykjaviki Stofnun Arna Magnissonar 4 fslandi, 1987), pp. 9-25. “* Indeed, Ségubrot af fornkonungum, in FN, 11, 133, suggests that the names of the rivers come from the name of Alft inn gamli and his family (see below). Regarding the parallels berween the word elfr for river and the word dffr, Sverrit Témasson has noted 10 me in conversation (17 January 2007) that both might originally be connected to the idea of whiteness/brightness (the proposed Indo-European root: “albh). * See, for example, Anne-Sophie Hygen and Lasse Bengtsson, Rock Carvings in the Borderlands: Bobuslin ard Ostfold, trans. by Virginia Siget and Maidie Kloster (Goteborg: Warne, 2000); and HH. R. Ellis Davidson, Gods and Myths of Northern Europe126 Terry Gunnell medieval period knew of these images, and that the sites around them (often closely associated with Bronze Age grave-mounds) retained an element of sacredness or at least mysticism, just as they did in later times. There is also good reason to believe that when Sighvatr encountered an difzblét on his way through “Gautland’, he was in an area close to Alfheimar, or an area where these difar were well known. Alfheimar, however, was more than just a simple place name. Historical sagas suggest that it drew its name from that of a regal forefather named ‘Alfr konungr in gamli’ (‘King Alfr the Old’), who, according to Hversu Noregr byggdist,®* headed the pedigree on the maternal side of Haraldr Adrfagri’s line, while Haraldr’s father’s people are said to be descended from Odinn (here said to be the father of Freyr). This choice of mythological forefathers can be no coincidence. The importance of the name Alfr is also seen in its use as a prefix, not only in the names of Alfr’s descendants but also in that of an Anglo-Saxon king (lfred) and those of a number of prominent Icelanders in later times. ‘The first Alfr is thus said co be father of Alfgeir/Alfarin and grandfather of Gandilfr/Alfr, who ‘ré fyrir Alfheimar (‘ruled Alfheimar’). Alfarin/Alfgeir’s daughter is the famous Alfhildr, first wife of King Gudrédr Haélfdanarson and accredited mother of Olafr Geirstadadlfr (and sometimes Ragnarr lodbrdk) according to Heimskringla,* Pdttr Olafi Geirstadadlfi, Porsteins saga Vikingssonar,?° and Sagubrot af fornkonungum. This, in addition to providing similar genealogical information, directly connects the people or rulers of this area to the supernatural difar, underlining at the same time their fairness: pat er kunnige { é!lum fornum ségum um pad folk er Alfar hécu, en pat var miklu fridara en engi énnur mannkind 4 Norduréndum, bvi atallt foreldri Althildar, médur hans [Ragnars lodbrékar], ok allt attbélkr var kominn fra Alf gamla. Pat voru ba kalladar Alfa eecir.”” (Harmondsworth: Penguin, 1964), p. 107. © In EN, U, 144~45 (from Flateyjarbok). © On the use of these names in Anglo-Saxon, see also Shippey, ‘Alias Over Habeo’, pp. 172-73. See also Kuhn, ‘Philologisches zur altgermanischen religionsgeschichte’, p. 267 See Snorti Sturluson, Heimskringla, I, 79, 87. © In Flateyjarbok, 1, 74-78. 7 In FN, I, 185-86. ™ Ibid., 1,133. ” Tid. 13.HOW ELVISH WERETHE ALFAR? 127 it is known in all ancient sagas of those peopie who axe called Alfar, and they were much more beautiful than any other human beings in the Nordic countries, because the parents of Alfhilde, his (Ragnarr /odbrék's] mother, and indeed the entire extended family goes back to Alfr the Old. They were called the gens of Alfar. Everything about these accounts suggests that the people of this area seem to have seen themselves — or at least their rulers — as going back to ‘Alfr’, and that the name itself could carcy a degree of sacredness. Obvious parallels can be seen with other personal names based on the names of favourite deities, as with bérdlfr or Freydis. This points to a key difference between the words dis” and aff, with the former serving as a group description while the latter seems to be based on a name. The ‘name’ Alfr, however, might itself have roots in descriptive euphemisms such as ‘Tyr’, ‘Freyr’, or ‘Odinn’, here describing the bright qualities of the god as a means of avoiding direct naming.” As noted earlier, 1 have deliberately avoided discussing Snorri Sturluson’s famous account of Hésdifar (‘light/bright difar) and dokkdlfar (‘dark dffar’) and his implied connections between the dlfar and the dvergar (‘dwarves’) as given in Gylfaginning and Skéldskaparmdl,” on the grounds that most scholars now regard this account as spurious, based to a large extent on the image of angels given in the Old Norse Elucidarius (itself a translation from Latin).” No mention is made of such a division in types of difar in any of Snorri’s known & kind of family guardian spit: see further Terry Gunnell, “The Season of the Disir: ‘The Winter Nights and the Désarblét in Early Scandinavian Belie® , Cosmi0s,16 (2000).117-49, and the references contained there. % Hall follows Turville-Petce in seeing the désiras a female equivalent of the dffar. See E. 0.G. Turville-Petre, Myth and Religion of the North: The Religion of Ancient Scandinavia (New York: Wiedenfeld and Nicolson, 1964), p. 231; and Hall, ‘The Meanings of Eifand Elves’, pp. 37-39, and 196. % See also Kuhn, ‘Philologisches zur alegermanischen religionsgeschichte’, pp. 268-69; and Hall, ‘The Meanings of Elf and Elves’, pp. 36, 45. % See Snorti Sturluson, Edda: Prologue and Gylfaginning, pp 19, 28; and Edda: Skdldskaparmdt, p. 45. ” See Elucidarius in Old Norre Translation, ed. by Evelyn Sherabon Firchow and Karen Grimstad (Reykjavik; Stofnun Arna Magniissonar 4 Islandi, 1989), pp. 8, 12-14. See also, for example, the discussion in Hall, “The Meanings of Elf and Elves’, pp. 33-343 on the question of duergar parallels, see further Loree Mote, ‘Of Elves and Dwarfs’, Arv, 29-30 (973-74), 93-127. On this and other Christian influences on Snorti’s image of the elves, secalso Anne Holtsmark, Studier { Snorres mytologi, Skrifter utg. av dec Norske Videnskaps- Akademi j Oslo, 11, Hist.filos. Klasse, n.s. 4 (Oslo: Universitetsforlaget, 1964), pp. 35-38.ng Terry Gunnell sources.” In part it echoes the medieval Church’s deliberate attempt to equate all popular nature spirits with demons or fallen angels, an idea reflected in numerous exempla-based folk legends found all over the Nordic area which tell of the origin of nature spirits or their longing for salvation.”” However, as a number of scholars have suggested, this attempt at classification might be another example of Snorri trying to construct a whole picture out of the conflicting views of the d/far which occur in different sources. Some of these seem to connect the difar with death and grave cults (as in the statement in Hdvamdl, stanza 142, that Daina [lit. ‘Dead’] was the leader of the dlfar and in the earlier noted account of Olafr Geirstadadlft), while others seem to stress their beauty, brightness, and connection with the sun." Rejecting De Vries’ suggestion" that the worship of the difar should simply be scen as a death or forefather cult, Strém, Schjodt, Simek, and Steinsland have all underlined that associations with forefather worship need not rule out connections with brightness or fertility.” Indeed both elements are clearly involved in the accounts 7 Admittedly, mention is made of ‘dackélfar’ alongside ‘gygiur ok pursar,/ nai, dvergar’ Cogressesand gianss, dead bodies, dwarves’) in the poem Hrafnagaldr Odins, otherwise referred to as Forspalbjéd, stanza 25: see Norren fornkvadi: Islandsk samling af folkelige oldtidsdigte om nordens guder og heroer almindelig kaldet Samundar Edda hins frida, ed. by Sophus Bugge (Christiania: Maling, 1867), pp. 375-76; also p. 371 (stanza 1) where difar are mentioned alongside the Vanir, nornir, pursar and valkyrjur, and p. 373 (stanza 6), where [unn is said to be of the race of difar. 7° See, for example, the much later exempla-based legend translated from Evald Tang Kristensen, Danske sagn som de har lydt i folkemunde, 7 vols (Athus: Silkeborg, 1892-1901), |, 3, in Scandinavian Folk Belief and Legend, ed. by Reimund Kvideland and Henning K. Sehmsdorf (Oslo: Norwegian University Press, 1991), p. 168, which tells of how ‘When our Lord expelled the fallen angels from heaven, they fell on the earth and became the troll folk we know. Some fell on the roof tops and became nisse; some fell in the water and became water sprites; some fell on the hills and became the hill folk; and some fell into the moors’, For translated examples of salvation legends, see Scandinavian Folk Beliefand Legend, pp. 206-07 and 255-56. * See, for example, Sdgubrot af fornkonungum noted above (n. 62); and the earlier noted references to Aificjne and the ‘alfiédull’ ® Jan De Vries, Aligermanische Religionsgeschichte, 2nf cdn, 2 vols (Berlin: de Gruyter, 1956-57), |, 257-60. On this suggestion, see also Kuhn, ‘Philologisches zur altgermanischen religionsgeschichte’. p. 269. * See Strém, Nordisk hedendom, pp. 198-993 Schjodt, ‘Relationen mellem aser og vaner’, pp. 306-07; Rudolf Simek, Hugtok og heiti inorrenni godafradi, trans. into Icelandic by Ingunn Asdisardéecir (Reykjavik: Heimskringla, Mals og menningar, 1993), pp. 25-26: and Gronell ate in tell ole ar leo. HOW ELVISH WERE THE ALFA? ng ofthe worship of Freyr’s and Olafr Geirstadadlfr’s grave mounds, both of which find parallels in ancestor worship in many other societies, and also in the ways in which the Nordic farm guardian spirits such as the gardvord, haugtusse or rudningskarl (who received offerings on their gravemounds up into the early twentieth century) were seen as protecting all aspects of farm life," This makes sense, at least for some of the early understandings of the difaras deities, whether they are seen as being Vanir or separate entities. All in all, the source material available to us underlines that we should be wary about ever referring to the earlier manifestations of the difar as ‘elves’ unless. we use the cerm in the Tolkienian sense, whereby they represent a form of godlike entity associated with the land. Tolkien, of course, was making a deliberate attempt to wrest the original difar away from latter-day Faériedom, and to restore them to their original mythological importance." At the same time, however, we should be wary about taking all the various early references to the d/feras referring to one and the same concept, over and above the idea that these entities are clearly powerful ‘others’, commonly seen as having the potential to harm ata distance (hence the later, widespread idea of the ‘elf-shot’, which is also often connected to a number of ather Nordic nature spirits). As I have sought to show here, everything about these references suggests that they stem from a variety of different belief systems originating in different times and different environments, though this does not eliminate the possibility that ail of these different beliefs could have lived side by side in a multicultural settlement such as Iceland before they gradually blended into the latter-day Icelandic d/far Steinsland, Norron religion: Myter, riter, samfunn (Oslo: Pax, 2005), pp. 338, 345 © See, for example, Haakon Shetelig, ‘Folketro om gravhauger’, Maal og minne (191), 206-12; and the Gulapingslig from before 1184 which suggests thac this practice was well known in medieval Norway: see Den Eldre Gulatingslova, ed. by Bjorn Eithun, Magnus Rindal, and Tor Ulset, Norronne tekster, 6 (Riksarkivet: Oslo, 1994), p. 52: ‘Blot er oss oc kvidiat at ver scolom eigi blota heidit Gud. ne hauga. ne horga’ (‘With regard to sacrifice, itis stated that we must nor make sacrifices to a heathen god or [grave] mound or shrine’). For translated examples of Nordic legends dealing with these beings, see Scandinavien Foik Belief and Legend, pp. 238-248, See Tolkien ‘On Faity Stories’, pp. 9~73, and especially p. 31, which underlines Tolkien's view that the stories of the Vanir and the others gods contain a mixture of myth and history.0 Terry Gunnell and buldufélk,* a process that can be seen occurring in works created from the late thirteenth century onwards. As for Jacob Grimm, while his romantic nationalism (noted at the start of this article) might have been the source for some of the misunderstandings still encountered in generalized, ‘top down’ studies of Old Nordic and Germanic mythology, we would do well to recall that the Grimm brothers were also the founding fathers of modern folkloristics in which, as noted above, the concept of local variation is fundamental. Their pioneering collection of source materials (both local and comparative) remains awe-inspiring. In a sense, it might be said that, even if after revealing the detailed lay of the landscape with regard to the difar and the other beings of Nordic mythology, Jacob Grimm may have led us up the wrong road, he and his brother also unconsciously provided us with the map and the methodology to get back on the right track. % Parallels to these figures are found in the latter-day Nordic huldre (like buldufélk, meaning ‘the hidden ones’), underjordiske (‘underground ones’), and diver (elves). These ae the end results of a similar process ia meaningMAKING THE MIDDLE AGES “THE CENTRE FOR MEDIEVAL STUDIES UNIVERSITY OF SYDNEY Volume CONSTRUCTING NATIONS, RECONSTRUCTING MYTH Essays in Honour of T. A. Shippey d by Andrew Wawn with Graham Johnson and John Walter & BREPOLS.ss rary Cataloguing in Publicvion Date salons, reconstructing myth essays in hinour of TA. Shippey - (Making che Middle Ages. 9) tc Literacure, Medieval -Hitory and exc 2 English 00 - History and 4 Myth i 5.Na T.Shippey, TA. I, Wawn, Andrew il Johazon, Graham IV, Wale, John ISBN-1y: 9782508523934 © 2007, Bepole Publishers 2, Ts Belgian All rights severved, No pact of his publ stored in cal ystem,o¢tamsmited, in clecronic, mechanical photocopying, recording o ohexw without the p permission ofthe publisher. Dizsorlo09s!99 ISBN: 978-2509-s974 Printed ithe E.U. on acid-free paper ‘Tom Shippey