[In the following excerpt, Mitchell discusses early Eddic poems from which the Volsunga Saga and other Scandinavian heroic prose narratives derive.]

… The saga writers were informed about Nordic mythology through many different conduits (for example, Snorra Edda, Icelandic ‘learned history’), and one of these was surely eddic poetry. This traditional narrative verse was the vehicle by which heroic adventures were recounted before the existence of the extant fornaldarsogur, as in the case of Hálfr of Hálfs saga ok Hálfsrekka and Sigurðr of Volsunga saga. For many of the heroes whose careers and adventures fill the Icelandic fornaldarsogur, origins can be postulated which stretch beyond the realm of the historical in even its most extended definition—where such a reference might indicate nothing more concrete than a postulated text—to a kind of inspiration which borders on the archetypal.¹ Before the written fornaldarsogur came into being, some, at least, of these heroic lives were recounted in verse,² which increasingly came to be embedded in a prose matrix, as in the case of Volsunga saga. Here the compiler of the extant fornaldarsaga has employed the eddic poems that treat the lives of Sigurðr, Brynhildr, and Guðrún and from their testimony—and to some extent from quotations of this material—formulated the most famous of the fornaldarsogur. Although such a process accounts for the extant Volsunga saga and probably for several other works, it should not be assumed that the transition from poetry to prose was ubiquitous.

What this process was like, how it began, and why it should have taken place at all are difficult questions—and yet some propositions can be offered. As a point of departure, we might well consider such transitional texts, works moving between the poetic and prosaic categories by means of the gradual accretion of prose inquits, a kind of process which describes several of the fornaldarsogur. Among the most famous accounts of sagas being told is that in Þorgils saga ok Hafliða (chap. 10), which reports how several sagas are delivered at a wedding at Reykhólar in 1119: Hrólfr af Skálmarnesi sagði sogu frá Hrongviði víkingi ok frá Óláfi liðmannakonungi ok haugbroti Þráins berserks ok Hrómundi Gripssyni, ok margur vísur með […] Þessa sogu hafði Hrólfr sammansetta. Ingimundr prestr sagði sogu Orms Barreyjarskálds ok vísur margar ok flokk góðan við enda sogunnar, er Ingimundr hafði ortan ‘Hrólfr of Skálmarness told the saga about Hrongviðr the Viking and about Óláfr the Warriors' King and the mound-breaking of Þráinn the Berserkr and about Hrómundr Gripsson, together with many verses […] Hrólfr himself composed this saga. Ingimundr the priest told the saga of Ormr Barreyjarskáld, including many verses and a good poem (flokkr) at the end of the saga which Ingimundr had composed.’ Although we will never know with certainty what forms these texts had, in trying to describe what such recitations might have been like one scholar likens them to the eddic Helgi lays, that is, “verses set in a simple narrative framework of prose.”³ Many of the fornaldarsogur probably had beginnings of this sort. Such a compositional profile may also imply that the poetry antedates the prose, and, to be sure, this situation frequently obtains in the corpus (for example, the “Battle of the Goths and Huns” in Heiðreks saga, which may be among the oldest pieces of Old Norse heroic poetry).⁴ Yet a relationship of this sort between prose and poetry is far from standard in the fornaldarsogur: some of the sagas contain poetry written in order to amplify and comment on action not otherwise reflected in the text, as in Hálfs saga ok Hálfsrekka.⁵ In still other cases, the sagas contain paraphrases of the preexisting poetry, rather than the poems themselves; such is the situation in the famous case of the Bjarkamál of Hrólfs saga kraka.⁶

Something of this process of prose accretion may be observed in the case of The Second Lay of Helgi Hundingsbani, to which the scene in Þorgils saga ok Hafliða was compared.⁷ The Helgi texts point toward the apparent growth of prose “connectors” between passages of poetry, at least one means by which some sections of the fornaldarsogur evolved. At certain junctures in Helgaqviða Hundingsbana II, for example, the events reported in poetry would make little sense without a mental dexterity that might tax even a knowledgeable audience familiar with the tradition. Even such an informed public might be confused when the stanzas of the extant poem move erratically from one scene to another, without any indication of the scene change. Such is the case, for example, following a conversation between Helgi and Sigrún in which she declares love for him and hatred for her betrothed, Hoðbroddr. In the subsequent stanza (25) she states:

“Muna þér Sigrún                     frá Sefafiollom,
Hoðbroddr konungr,                     hníga at armi;
liðin er ævi—opt náir hrævi
gránstóð gríðar—Granmars sona.”

“Never, King Hoðbroddr, shall you hold Sigrún of Sefafiall in your arms; that time is past; the gray steeds of the giantess [wolves] shall feast upon carrion—Granmarr's sons.”

It is generally held that members of the audience were capable of making all the correct assumptions about the activity that intervenes between the two stanzas, as they were familiar with the plot of the story. Yet the collector of the Edda seems dissatisfied with such a presentation and consequently inserts a prose explanation in which he tells of the battle fought between Sigrún's aggrieved family and suitor and the forces Helgi commands, concluding it with the statement Sigrún gecc í valinn oc hitti Hoðbrodd at kominn dauða ‘Sigrún went out among the slain and found Hoðbroddr near death’ (p. 154). In this case, the prose insert forms a bridge between the units of versified dialogue and we may well ask whether or not at some earlier point in the poem's existence, the same purpose was served by poetry in Helgaqviða Hundingsbana II (or *Volsungaqviða in forna ‘The Old Lay of the Volsungs,’ the work to which the collector refers in the Edda).

At other points in the text, however, the redactor's asides function differently; they do not clarify the actions of the stanzas but merely serve to prepare the audience for that which could be inferred from the poetry itself. The prose and poetry are here no longer in complementary distribution, but instead amplify and enlarge, or simply duplicate, each other. Thus, when Helgi is later killed by Sigrún's brother and buried, the redactor tells us: Ambót Sigrúnar gecc um aptan hiá haugi Helga oc sá, at Helgi reið til haugsins með marga menn ‘Sigrún's slave-woman went near Helgi's burial mound one evening and saw that Helgi rode to the mound with many men’ (p. 159). The woman then speaks (st. 40):

“Hvárt ero þat svic ein,                     er ec siá þicciomz,
eða ragna roc,                     ríða menn dauðir?
er ióa yðra                     oddom keyrit,
eða er hildingom                     heimfor gefin?”

“Is this a deception I seem to see, or the Fate of the Gods [end of the world]—do dead men ride? Do you with spurs urge your steeds; are heroes given leave to come home?”

Helgi responds to the servant, who runs off to report the news to Sigrún, including the fact that up er haugr lokinn, / kominn er Helgi ‘the burial mound is open, Helgi has come’ (st. 42). The dialogue given in verse relates the same information already amply provided in the prose of the collector, rendering the verse nearly redundant. Unlike the previous case, the redactor has here unnecessarily cluttered the text and added prose more or less for its own sake, perhaps as a means of exerting his own authorial presence on the preexisting materials. The effect is that the prose tends to impede the flow of the tale rather than to assist it. From the evidence of the Helgaqviða Hundingsbana II, it may be possible to see the process of prose accretion whereby traditional poetic texts become mixed prose and poetry narratives. At times such embellishments are useful, if somewhat overly ample, narrative bridges; at other times, they are little more than agrestic intrusions on the preexisting material.⁸

The life of Helgi Hundingsbani is known not only from the eddic poems, but also from Volsunga saga, the only fornaldarsaga in which the relationship between traditional eddic poetry and the prose form can be explored in some detail. Scholarship on the Volsunga saga, and the rest of the corpus of texts dealing with the Nibelungs, is nothing less than voluminous, as are the theories that seek to account for the relationships between the several extant texts, to say nothing of the hypothetical works that have been inferred over the years.⁹ In addition to the eddic poems and Volsunga saga, medieval texts that treat the Sigurðr-Nibelung cycle include Snorra Edda, Þiðreks saga, and the Middle High German Nibelungenlied. The popularity of the tradition is further indicated by the wealth of supporting plastic representations of the tale throughout northern Europe (for example, the Ramsund petroglyph in Sweden, the wood-carvings in Norwegian stave churches, the petroglyph from the Isle of Man) and by the fact that the story of Sigurðr the dragon slayer appears among the literary traditions of other Germanic peoples, such as the reference to the story of Sigemund the dragon slayer in Beowulf (ll. 874-97).¹⁰

There are good indications too that Sigurð's youthful exploits and death were once the subject of their own prose saga. Just such a *Sigurðar saga Fáfnisbana is mentioned in Norna-Gests þáttr (sem segir j sogu Sigurdar Fafnisbana ‘as is related in the saga of Sigurðr the dragon slayer’ [chap. 287]) and in Snorra Edda (eptir Sigvrþar savgv ‘according to the saga of Sigurð’ [p. 231]). Finnur Jónsson, after an exhaustive examination of the prose inquits in the early Sigurðr poems in the Poetic Edda, concludes that these inserts were in fact based on the redactor's knowledge of the *Sigurðar saga, although this view has not been universally accepted.¹¹ It is reasonably assumed that the writer responsible for Volsunga saga had access to the materials of the Poetic Edda: the citations from the texts testify to this fact.¹² At times, it seems that only the most meager discrepancies separate the prose of Volsunga saga and the poems of the Edda. Thus, for example, when Sinfjótli and Granmarr engage in a flyting in Helgaqviða Hundingsbana I, the text (sts. 37-42) closely mirrors the structure and essential contents of the eddic poems, as in the following example (st. 37):

“Þú vart volva í Varinseyio,
scollvís kona, bartu scroc saman;
qvaztu engi mann eiga vilia,
segg bryniaðan, nema Sinfiotla.”

“You were a witch on the island of Varin, a deceitful woman, you created falsehoods, said that you wanted no other man, no mailed warrior, only Sinfjotli.”

This passage may be compared to the corresponding section of Volsunga saga (chap. 9), where the compiler writes: Eigi mantu glokt muna nu, er þu vart volvann i Varinsey ok kvazt vilia mann eiga ok kaust mik til þess embęttis at vera þinn madr ‘You cannot now remember when you were a witch on the island of Varin and said that you wanted to have a man and chose me for the service of being your husband.’ The same information is covered in the two narratives, but the results for the flyting as a whole are striking. The prose version no longer has the volume or range of invectives found in the poetry and loses much of the vigor contained in the abusive language of the original.

The difference in the resulting text—with its tendency to blunt the tale as given in the poetic original—is one exhibited throughout the saga. For example, using the two Atli poems of the Edda, which cover the same materials but in very different styles, the Volsunga saga author attempts to mediate the differences between the two versions. The baroque and grotesque scene of the screaming thrall being chased about is retained, but lacks the deceit that justifies the motif in Atlaqviða; the consequence is that the episode in the saga merely dulls the heroic confrontation between Gunnar and Atli over the Niflung treasure and Hogni's fate (chap. 37). Yet despite this and other difficulties, it would be wrong to assume that the prose saga is consistently inferior to its originals. It is true that the saga does not in general possess the terseness characteristic of the eddic poems, preferring instead a more descriptive and reflective style; but as Wieselgren comments, the author is not nearly so bad as his reputation would seem to make him.¹³ Wieselgren further points out that the degree of fidelity with which the Volsunga saga author has held to the poetic materials of the Edda is highly variable: in some instances these texts are merely paraphrased, at other times, copied quite faithfully; at still others, lexical substitutions are common.¹⁴ For unknown reasons, the author nowhere quotes the Helgi poems.

The conditions, considerations, and practical constraints under which the redactor of the Edda and the author of Volsunga saga worked—and our inexact knowledge of these problems—should make us wary of using the example of the Helgi poems, and Volsunga saga more generally, to draw conclusions as to how the authors of the fornaldarsogur made use of traditional eddic poetry. Yet certain points of comparison—especially with regard to a growing preference for prose narration—help us understand how traditional poetic material often gave rise to long prose sagas, or was taken wholesale as the basis for saga composition. In the case of Volsunga saga (at least), the process appears to have been something very much like Heusler's characterization of the book-prose theory, when he writes that the saga author's main source was tradition, supplemented by the poetic heritage and by written works.¹⁵ It is unlikely that any single method will ever account for the entire body of fornaldarsogur, and certainly there is no reason to believe that every one of the extant fornaldarsogur was preceded by an elaborate set of poems. Still, this process—the transformation of traditional poetic materials into basically prose sagas—is no doubt the historical and technical reason the fornaldarsogur display a uniform preference for eddic, as opposed to skaldic, poetry. Skaldic verse accommodates occasional poetry, whereas it is to the various eddic forms of poetry, such as málaháttr, ljóðaháttr, and fornyrðislag, that all early lengthy narrative poetry in Scandinavia looks. Where poetry was composed for the express purpose of accompanying a fornaldarsaga text, it always imitated one of these verse forms, since the connection between eddic poetry—and the world conjured up by the poems of the Edda, with their legendary heroic figures and mythological characters—and the topics of the fornaldarsogur must already by an early point have been very strong.

Notes

1. So, for example, de Vries (1963), who takes up several of the figures from the fornaldarsogur.

2. An older generation of scholars believed in the primacy of poetic forms of the fornaldarsogur, an idea which has been under much attack in recent years. Heusler (1955, 151), for example, maintains with reference to the Nibelung materials that what is usually called heroic legend (Heldensage) ought to be called heroic poetry (Heldendichtung), the product of a specific poet's labor. Schneider's dictum, that heroic legend is heroic poetry (“Heldensage ist Heldendichtung”), is implicit throughout his Germanische Heldensage (cf. de Boor 1929). It is a view that can boast fairly deep roots in literary interpretation. Ker 1908 divides the medieval world into an heroic age followed by an age of chivalry, with the parallel in literature of an age dominated by “epics” (e.g., Beowulf) followed by an age of romance. It is not difficult to imagine how the two came to be regarded as periods of poetic and prose composition respectively. Especially important in refuting this view have been Genzmer 1948 and Kuhn 1952. Genzmer and Kuhn argue against the idea that the majority of fornaldarsogur (and other Germanic heroic texts) are necessarily prose reworkings of earlier lays; thus, Kuhn maintains (1952, 276), “The sagas originated from widely known materials (aus den bekannten Elementen), although not necessarily in the form of lays (Lieder).”

3. Þorgils saga ok Hafliða, ed. Dronke, p. 75.

4. Tolkien 1955-56.

5. Cf. Genzmer 1948, 16, but see also Jón Helgason (1934, 203-4), who states that the poetry is not particularly old, but older than the prose of the saga at any rate.

6. Any discussion of Bjarkamál must begin with Olrik's daunting project (1903-10, I, 28-114), in which he bravely attempts to piece together the poem from the various medieval sources (Saxo, Snorri's Heimskringla, Snorra Edda, and Hrólfs saga kraka) and thereby provide a vernacular edition of the poem.

7. See the remarks and bibliography contained in J. Harris 1983.

8. For parallel examples from other cultures, see Godzic and Kittay 1987.

9. For a comprehensive review of the scholarship in this area, see the discussion and bibliographic materials in Andersson 1980.

10. The many sources of plastic information on the Sigurðr cycle are covered in Margeson 1980 and Sigurds saga i middelalderens billedkunst. A convenient review of textual sources is provided in Andersson 1980, 20-22.

11. Finnur Jónsson (1917, 36) credits the lost prose story of Sigurðr's life with having been the basis for chaps. 1-8 of Volsunga saga. In contrast, Heusler (1919, 47-48) feels that the Helgi material was an original contribution by the author of Volsunga saga. Thus, both regard the Helgi material of chap. 9 as falling outside the scope of items encompassed by the lost saga.

12. Wieselgren 1935, 16-30.

13. Ibid., p. 14.

14. Ibid., p. 12. See also Finch (1962-65), who carefully reassesses the contributions and techniques of the compiler, as well as his more recent study (1981), which regards the saga as a synoptic prose version of the poems. The general question of the relationship between prose and poetry in the fornaldarsogur has been addressed by Davið Erlingsson (1987).

15. Heusler 1909a, 445. Basing his remarks on the writing of Finnur Jónsson and others, Heusler characterizes the view with the “oft-used formulation.” “The author's main source was tradition […] an important further source consisted of verses or the ‘poetic tradition’; a third consisted of written works.”