The Medieval Icelandic Saga and Oral Tradition: A Discourse on Method

Citation: Sigurðsson, Gísli. 2004. The Medieval Icelandic Saga and Oral Tradition: A Discourse on Method. Trans. Nicholas Jones. Milman Parry Collection of Oral Literature 2. Cambridge, MA: Milman Parry Collection of Oral Literature. http://nrs.harvard.edu/urn-3:hul.ebook:CHS_SigurdssonG.The_Medieval_Icelandic_Saga_and_Oral_Tradition.2004.

8. Implications for Saga Research

The overall conclusion to emerge from this study is that, by assuming the existence of a living oral tradition in Icelandic society in the 12th and 13th centuries, the perspective of our research shifts fundamentally. This change applies equally to how we interpret both the historical facts and the individual texts themselves. For instance, in Part I reasons were given for thinking that the lawspeakers of the ancient Icelandic Commonwealth did not necessarily view the written codification of the law as ‘a step in the right direction’: this introduction of writing into what had previously been an oral preserve served primarily the interests of the Church and might well have been viewed by the lawspeakers themselves as a threat to their position of authority within society. Similarly, the analysis of the verse examples in Óláfr Þórðarson hvítaskáld’s Third Grammatical Treatise led to new insights regarding the extent and nature of the oral poetic tradition, specifically that the poetic knowledge of a man brought up in the oral tradition of poetry and storytelling in the 13th century appears to have been confined principally to the parts of the country in which his and his family’s influence was felt most strongly, while his citations from the common poetic heritage of previous centuries suggest that this heritage was restricted largely to court poetry produced by Icelandic skalds on their travels abroad. The domestic tradition of oral poetry therefore appears to have been much more highly localized than was previously thought.

Part II turned to the implications of assuming a living oral tradition of storytelling and historiography behind the written prose literature of medieval Iceland. This approach enables us to identify features of the style and meaning of the sagas that can be traced to an interplay between the texts and the audience’s prior knowledge of the characters and events described, a knowledge that can only have come from oral tradition. Hitherto, these features have generally passed unnoticed among scholars, or been condemned as structural flaws and put down to incompetence on the part of the authors. By applying the findings of research into modern oral traditions and using them to cast light on the culture of medieval Iceland and frame new questions of our limited sources, it becomes possible to construct an image of how these ancient writings might have worked upon their original audiences—which should always be the first task of the literary scholar, before it is permissible to venture out onto the thin ice of textual interpretation through the methods of contemporary psychological and critical theories.

This section looked in some detail at a number of examples from the sagas set in the east of Iceland in which the same event is described in more than one source. The examination of these parallel passages suggested that scholars have been overly ready to postulate literary relations between texts (rittengsl) and have claimed to detect such relations far more widely than the material justifies; if we adopt as a working principle that such connections can only be accepted as valid where there are concrete verbal correspondences between texts, the number of places where it seems at all probable that there are direct literary relations among the sagas of the east becomes very small indeed. On the other hand, it was possible to show that in many places the writers clearly assumed that their audiences were already familiar with material that was critical to an understanding of the texts, such as the general ‘biographies’ of men like Þorkell Geitisson. In the case of Gunnarr Þiðrandabani, assuming orally based connections enabled us to cut through the insoluble tangle of textual relationships proposed and disputed by previous scholars.

Once it has been shown that there are strong possibilities that the written sagas tapped into a rich fund of living accounts preserved in oral tradition, it becomes tempting to try to assess the age of this tradition: How far back is it possible to trace the thread connecting the texts to the events they purport to describe? This question provided the center of interest in Part III, using the Vínland sagas, Eiríks saga rauða and Grœnlendinga saga, as an example. These sagas preserve memories of voyages made from Iceland and Greenland to the mainland of North America at a time when archaeology tells us that men and women from these lands but with connections to the British Isles built three huts at the northern tip of Newfoundland and traveled on south from there to gather items of the local vegetation. Since it has been demonstrated that previous theories of a written literary relationship between these two sagas no longer hold water and that both must have been written down independently of each other based on oral accounts (see Ólafur Halldórsson 1978:293-400), these sagas provide an excellent testing ground for the methodology advocated in this study. By using the sagas as a source for the oral tradition from which they arose, it proved possible to reconstruct a visual image that we can suppose constituted a living part of the tradition relating to these distant lands. This is rather different from the way that these texts have been treated hitherto, viz. by superimposing them onto a real landscape in the manner of a ship’s logbook, and, if they fail in this, rejecting them in part or in whole as admissible sources. What emerged was that the overall picture of the lands of the Vínland voyages provided by the tradition bore a perfectly recognizable resemblance to a rough sketch map of the east coast of North America, from New England north. This mental or ‘immanent’ map corresponds well to the regions we can suppose Viking travelers passed through and built up a reasonable knowledge of in the course of a number of several-year voyages of exploration south from Newfoundland, with their eyes open for the resources of the new lands they came to and their potential for settlement.

As noted, the Icelandic sources—particularly the sagas of Icelanders and Landnámabók—contain several examples of variant accounts of the same events. Very often one account is more circumstantial and detailed than another, but it is generally fairly easy to detect when there has been direct use of sources, i.e. when one written source is based on another, be it another saga or Landnámabók. A clear example of this kind of use of written sources occurs in chapter 26 of Reykdœla saga ok Víga-Skútu (‘Saga of the Men of Reykjadalur and Killer-Skúta’) and chapter 16 of Víga-Glúms saga (‘Saga of Killer-Glúmr’). The correspondences point unequivocally to a written literary relationship, though not to which saga is using which. Jónas Kristjánsson (1978:299), for instance, works on the assumption that the ‘Skúta episode’ in Víga-Glúms saga was the source for Reykdœla saga, while Hofmann (1972) inverts the relationship and considers the episode to be original to Reykdœla saga on the grounds that it is better integrated into the narrative there.

Whatever the case, the literary relationship is incontrovertible. A short example from the conversation between Skúta and the accomplice he sends to lure Glúmr into the open will suffice:

Reykdœla saga ok Víga-Skútu, chapter 26 (ÍF X:231)

Víga-Glúms saga, chapter 16 (ÍF IX:50)

‘Þú skalt fara á hans fund ok mæla þessum orðum, at þú þykkisk þurfandi, at hann gerisk forsjámaðr ráðs þíns, ok seg at þitt vandræði er mikit, er þú hefir beðit af vígaferlinu. Ok get ek, at þann veg beri til um fund ykkarn, at Glúmr sé í þingreið. En hans skaplyndi er þat, at hann er maðr þrautgóðr, ef menn þurfu hans, ok enn mætti svá verða, ef þú gerir þitt mál líkligt, enda viti hann at þú ert hjálplauss, at hann mæli, at þú farir til Þverár ok bíðir hans þar, til þess er hann kemr heim af þinginu.’
(‘You shall go and meet him and say these words, that you feel you’re in need of him making himself the man responsible for your interests, and say that the difficulties are great that you have suffered as a result of the feud. And I predict that your encounter will come about this way, that Glúmr will be riding to the thing. And his nature is so, that he is a man of great resource, when people need him, and might turn out so to be still if you make your case sound plausible, especially as he knows you are vulnerable, so that he tells you to go to Þverá and wait for him there, until he comes home from the assembly.’)

‘Þú skalt fara sendifǫr mína til Víga-Glúms ok mæla þessum orðum við hann, at þú þykkisk þurftugr, at hann sé forstjórnarmaðr þíns ráðs. Ek get, at nú beri svá til um fund ykkarn, at hann sé í þingreið. Hann er þrautgóðr ef menn þurfu hans, ok má vera at hann mæli at þú farir til Þverár ok bíðir hans þar.’
(‘You shall go on my errand to Víga-Glúmr and say these words to him, that you feel you’re needful of him to be the man in charge of your interests. I predict that your encounter will now come about in this way, that he will be riding to the thing. He is of great resource when people need him, and it may be that he tells you to go to Þverá and wait for him there.’

Equally uncontroversial are the literary relations between the Sturlubók version of Landnámabók and the early chapters of Grettis saga, as detailed by Guðni Jónsson (1936:xvii) in his introduction to Íslenzk fornrit VII. Jónsson’s view is that the saga must be the recipient. As an example, we may take a passage from the account of Bjǫrn Hrólfsson and Eyvindr the Easterner, in which there are repeated verbal correspondences and full consistency in the names in both sources, a sure sign of textual relations. This should be compared with the very different nature of the relationship between the parallel passages in Hrafnkels saga and Landnámabók discussed earlier (see pp. 26 f, 197 f).

Landnámabók (S 217)

Grettis saga, chapter 3 (ÍF VIII:8)

Bjǫrn hét maðr ágætr á Gautlandi; hann var son Hrólfs frá Ám; hann átti Hlíf, dóttur Hrólfs Ingjaldssonar, Fróðasonar konungs. Eyvindr hét son þeira. Bjǫrn varð ósáttr um jǫrð við Sigfast, mág Sǫlvars Gautakonungs, ok brenndi Bjǫrn hann inni með þremr tigum manna. Síðan fór Bjǫrn til Nóregs með tólfta mann, ok tók við honum Grímr hersir son Kolbjarnar sneypis, ok var með honum einn vetr. Þá vildi Grímr drepa Bjǫrn til fjár; því fór Bjǫrn til Ǫndótts kráku, er bjó í Hvínisfirði á Ǫgðum, ok tók hann við honum. Bjǫrn var á sumrum í vestrvíking, en á vetrum með Ǫndótti, þar til er Hlíf kona hans andaðisk á Gautlandi. Þá kom Eyvindr son hans austan ok tók við herskipum fǫður síns, en Bjǫrn fekk Helgu, systur Ǫndótts kráku, ok var þeira son Þrándr. Eyvindr fór þá í vestrvíking ok hafði útgerðir fyrir Írlandi. Hann fekk Rafǫrtu, dóttur Kjarvals Írakonungs, ok staðfestisk þar; því var hann kallaðr Eyvindr austmaðr.
(There was a man of noble birth from Götaland [in modern southern Sweden] called Bjǫrn; he was the son of Hrólfr of Ár; he was married to Hlíf, the daughter of Hrólfr, son of Ingjaldr, son of King Fróði. Their son was called Eyvindr. Bjǫrn had a land dispute with Sigfastr, the brother-in-law of Sǫlvarr, king of the Gauts, and Bjǫrn burned him alive in his home with thirty men. Then Bjǫrn went to Norway with eleven others, and was received by Lord Grímr, son of Kolbjǫrn sneypir, and <he> was with him for one winter. Then Grímr wanted to kill Bjǫrn for his wealth, so Bjǫrn went to Ǫndóttr kráka, who lived in Hvínisfjǫrðr in Agðir, and he welcomed him. Bjǫrn spent the summers raiding in the British Isles and the winters with Ǫndóttr, until his wife Hlíf died in Götaland. Then his son Eyvindr came west and took over his father’s warships, and Bjǫrn married Helga, the sister of Ǫndóttr kráka, and they had a son called Þrándr. Eyvindr then went raiding in the British Isles and kept a force of ships off the coast of Ireland. He married Rafarta, the daughter of Kjarval, king of the Irish, and settled there; for this reason he was called Eyvindr the Easterner.)

Bjǫrn var faðir Þrándar ok Eyvindar, sonr Hrólfs frá Ám. Hann stǫkk ór Gautlandi fyrir þat, at hann brenndi inni Sigfast, mág Sǫlva konungs. Síðan hafði hann farit til Noregs um sumarit ok var með Grími hersi um vetrinn, syni Kolbjarnar sneypis. Hann vildi myrða Bjǫrn til fjár. Þaðan fór Bjǫrn til Ǫndótts kráku, er bjó í Hvinisfirði á Ǫgðum. Hann tók vel við Birni, ok var hann með honum á vetrum, en herjaði á sumrum, þar til er Hlíf kona hans lézk. Eptir þat gifti Ǫndóttr Helgu dóttur sína Birni, ok lét Bjǫrn þá enn af herfǫrum. Eyvindr hafði þá tekit við herskipum fǫður síns ok var nú orðinn hǫfðingi mikill fyrir vestan haf. Hann átti Rafǫrtu, dóttur Kjarvals Írakonungs.
(Bjǫrn was the father of Þrándr and Eyvindr, and the son of Hrólfr of Ár. He fled from Götaland because he had burned Sigfastr, the brother-in-law of King Sǫlvi, alive in his home. Then he had gone to Norway in the summer and stayed with Lord Grímr, son of Kolbjǫrn sneypir during the winter. He wanted to murder Bjǫrn for his wealth. From there Bjǫrn went to Ǫndóttr kráka, who lived in Hvinisfjǫrðr in Agðir. He welcomed Bjǫrn well, and he stayed with him in the winters and went raiding in the summers, until his wife Hlíf died. After this Ǫndóttr married his daughter Helga to Bjǫrn, and then Bjǫrn left off raiding. Eyvindr had then taken over his father’s warships and had by now become a great chieftain in the west over the sea. He married Rafarta, the daughter of Kjarval, king of the Irish.)

The two examples above demonstrate excellently how manifest literary relations can be in clear cases of borrowing between written sources. The most salient feature of such passages is their patent and repeated verbal correspondences.

Things are rather different when we turn to more complex textual relations such as those we find in the similar circumstances surrounding the killings of Helgi Ásbjarnarson in Droplaugarsona saga and Þorgrímr Þorsteinsson in Gísla saga Súrssonar. These passages, and earlier discussions of their relationship, were analyzed in depth by Theodore M. Andersson (1969:28-39), who came to the conclusion that the thematic treatment of the incidents in the two sagas pointed to Droplaugarsona saga being influenced by Gísla saga rather than the opposite way around. However, despite the obvious general similarities in diction in the two passages, it is impossible to demonstrate a specifically literary relationship through tangible verbal correspondences. The relationship might thus be explained in other ways, such as through oral tradition rather than the direct use of written sources.

In many cases, scholars’ interpretations of parallels between individual passages in different sagas have been based less on objective criteria than on their initial working hypotheses of saga origins. Thus, advocates of the bookprose theory have been particularly inclined to identify literary relations on all conceivable occasions, since this fits in with their general conception of an individual author behind each saga constructing his work out of a collection of assorted written sources. On the other hand, freeprose men, and more recently those influenced by the ideas of formalist-traditionalism, have frequently chosen to interpret as evidence of oral preservation things that bookprose men have taken to be irrefutable proof of literary borrowing. One clear advantage of assuming an oral tradition in the background is that it frees us from the need to assume the existence of hypothetical lost written sources; instead, we can treat general similarities between the written sagas as being a consequence of the fluid nature of the oral tradition out of which they arose. Further justification for an interpretative approach to the sagas through oral tradition comes from our certain knowledge that people in the Middle Ages were quite capable of telling stories and performing poems without always having to turn to written sources. There is thus no need to prove the existence of an oral tradition. With the exception of Walter Baetke (1956) and his followers, the vast majority of scholars accept the existence of some kind of oral tradition behind the sagas; it is its degree, nature, and effects that are under dispute.

The Feud between Finnbogi the Strong and the Men of Hof

A particularly instructive example of how scholars’ initial working hypotheses can affect the findings of their research concerns the ways in which they have interpreted the differing perspectives toward the same events in two accounts of the feud between Finnbogi rammi (‘the Strong’) Ásbjarnarson of Borg in Víðidalur in the Húnavatn region of northern Iceland and the sons of Ingimundr inn gamli (‘the Old’) Þorsteinsson of Hof in the neighboring valley of Vatnsdalur. This feud is described in Vatnsdœla saga (ÍF VIII) and Finnboga saga ramma (ÍF XIV), and in both cases the events themselves are largely the same. In other respects, however, the two accounts are so different that it is by no means obvious whether the connection between them is best explained through oral tradition or some kind of written literary relationship. Though most discussions of the two sagas and the connections between them allow for some kind of oral tradition to account for their differences, the received opinion among recent commentators appears to be that one of them must have used the other in some way in written form. These sagas therefore provide a good litmus test of how general theories of saga origins have shaped scholars’ findings and their interpretations of individual sagas. Simultaneously, they provide an opportunity to assess the validity of the arguments adduced in favor of literary relationships and to ask whether there are perhaps other factors that point rather in the opposite direction, i.e. that suggest a specifically oral connection behind these two particular sagas. [1]

The value of this example is greatly increased by the fact that the dealings between Finnbogi the Strong and the men of Hof are also described in the Melabók version of Landnámabók (96). Three points relating to the locations of events are specified in greater detail in Landnámabók than in Vatnsdœla saga, though without this giving rise to serious doubts about the literary relationship (see Halldórsson, J. 1959:lix-lx). More generally, other material from Vatnsdœla saga appears to have been incorporated into various versions of Landnámabók; this matter was investigated by Einar Ól. Sveinnson (1939:xxxiii-xli), who came to the conclusion that the Landnámabók texts had drawn on a lost written version of Vatnsdœla saga rather than simply on oral tales about the people of Vatnsdalur. [2]

As was pointed out in the Introduction (p. 38 f.), our ideas about the artistic capabilities of oral narrative have undergone considerable change in recent years, and this has been reflected to some extent in research into old Icelandic literature. One thing that has become absolutely clear is that there is no necessary contradiction between stories being well made and showing signs of conceptual unity and their having been passed on through oral transmission—a preconception that, however, appears to color most recent general treatments of the sagas here under discussion. If the written sagas drew a significant amount of their material from an oral narrative tradition anything like the living oral traditions of more modern times, it becomes difficult to apply the same methods of analysis to them as have been developed in the criticism of modern novels by named authors. The laws applying to artistic presentation are different in oral narrative, for instance in matters such as shared background knowledge of the story material, prestructured diction, and the structuring of individual themes and narrative units. These rules call for critical methods designed specifically for works rooted in such a tradition.

Oral tales in a tradition of this type cannot be regarded as reliable sources for events of the 10th and 11th centuries, whether they turn up in Landnámabók or the sagas of Icelanders. Even though written genealogies may have existed in some form from an early period, most of the narrative content of Landnámabók is based on oral accounts by people who were not first-hand witnesses of the events they described. The compilers of Landnámabók could not confirm the narratives of their informants other than by comparing them with the equally unreliable accounts of other informants. Thus both the sagas of Icelanders and Landnámabók are based on traditional tales that would have been highly unreliable as historical sources for events of the 9th, 10th, and 11th centuries by the time they came to be used in written texts in the 12th, 13th, and 14th centuries. These considerations need to be kept constantly in mind in any attempt to assess the nature of the relationship between the accounts of the same events in Vatnsdœla saga and Finnboga saga ramma.

Vatnsdœla saga, chapters 31-35

Finnboga saga ramma, chapters 33-35

In Vatnsdœla saga the arrogance of a relative of Finnbogi, Bergr inn rakki, leads to ill feeling with Þorsteinn and Jǫkull, the sons of Ingimundr of Hof. Finnbogi and Bergr wade across the Vatnsdalur river in the height of winter on their way to a wedding. At the wedding they clash with the brothers from Hof, as a result of which Finnbogi and Bergr challenge Þorsteinn and Jǫkull to a duel. Bergr’s concubine, Helga, does not like the look of things and on the day of the duel there is a fierce blizzard with freezing conditions and snowdrifts, and Finnbogi and Bergr are unable to make it to the dueling ground. Helga is suspected of being behind this sudden onset of bad weather. However, Þorsteinn and Jǫkull do get there, together with their brother Þórir and a local farmer called Faxa-Brandr, who assists Jǫkull in raising a níð-pole against Finnbogi, a deeply denigrating construction symbolizing cowardice and sexual deviance. Finnbogi attempts to take revenge but Þorsteinn manages to muster enough men to hold him at bay without a fight. The dealings between the two sides come to an end with Finnbogi moving away to Trékyllisvík in the northwest and Bergr disappearing from the saga.

In Finnboga saga ramma the feud begins when Finnbogi gets a relative of his wife called Þorkell to propose to a woman who is generally known to be the mistress of Jǫkull Ingimundarson. Jǫkull resents this deeply but is unable to prevent the marriage from going ahead or to take revenge. His brothers react positively to Finnbogi’s invitation to the wedding but are reluctant to go against their brother’s wishes. A little later, Bergr, a relative of Finnbogi’s, arrives in Iceland together with his Hebridean wife, Dalla. When winter comes, Finnbogi and Bergr are invited to a wedding at Hof. On the way there they have to wade across the Vatnsdalur river. At the feast, Jǫkull shoves Bergr aside but Jǫkull’s brothers, Þorsteinn and Þórir, manage to restore peace for the time being. The following summer Jǫkull challenges Finnbogi to a duel, and Þorsteinn challenges Bergr. On the day set, Dalla conjures up a storm so fierce that Finnbogi and Bergr fail to get to the agreed place, and later it comes out that Jǫkull has raised a scathing níð-pole. The following summer, as Finnbogi and Þorkell are accompanying Bergr to his ship, they walk into Jǫkull’s ambush and Bergr and Þorkell are killed before Jǫkull’s brothers arrive and stop the fight. After some further inconclusive skirmishes with the men of Hof, Finnbogi is forced to leave Borg and move to Trékyllisvík in the northwest, but his feud with Jǫkull simmers on.

Judging from recent reference books, there seems to be a general consensus on the ages of these two sagas and the literary relations between them. Of Vatnsdœla saga, Rudolf Simek and Hermann Pálsson (1987) say that it was written before 1280, on the grounds that it was used by Sturla Þórðarson (d. 1284) in his redaction of Landnámabók, and after 1260, since the ‘author’ appears to have used Hallfreðar saga, Ǫrvar-Odds saga, and Laxdœla saga, all believed to have been written before, but not necessarily long before, 1260. In his entry on Vatnsdœla saga in Dictionary of the Middle Ages, Paul Schach (1989) says that the saga was probably written around 1270 in the monastery at Þingeyri (not far from Vatnsdalur) and that there is ‘evidence to suggest’ that the text we have is a rewriting of an older version that was used by Sturla Þórðarson when compiling his version of Landnámabók. [3]

Regarding Finnboga saga ramma, Simek and Pálsson (1987) say that it first appeared in the 14th century and that its main hero has historical roots as he is also known from Landnámabók, Íslendingadrápa, [4] and Vatnsdœla saga. Schach (1985:64-5) gives a date of around 1310 for the saga and says that although Finnbogi was a historical character (as proved by his name appearing in Landnámabók and Íslendingadrápa) ‘the first part of his story is a fantasy made up of popular saga and fairy tale (Märchen) motifs.’ Schach also discusses the saga’s relations with other sources, particularly Vatnsdœla saga, and points out that there are very few verbal correspondences that might suggest direct borrowing between the texts. On the obvious similarities and differences between the two sagas, Schach remarks: ‘Some scholars have felt that this contrast reflects preliterate differences in local traditions that developed in the districts of Vatnsdalr and Víðidalr, Finnbogi’s original home.’ However, Schach rejects this view on the typical bookprose grounds that ‘good’ equates with written and ‘bad’ or ‘clumsy’ with oral: ‘The consistent enhancement of Finnbogi in his saga and the corresponding denigration of Jökull, however, suggest authorial intent,’ i.e. consistency of characterization and authorial intent make it unlikely that the saga derives from oral tales. Toward the end of the entry, Schach repeats a frequently expressed view concerning the writing of Finnboga saga: ‘The author’s major purposes in composing this story seem to have been entertainment and the literary rehabilitation of Finnbogi of Víðidalr, who comes off rather badly in the saga of the Vatnsdalr chieftains [i.e. Vatnsdœla saga].’ [5]

These rather bald comments are based on a number of presuppositions: a) that we can only assume that the contents of a saga became known once it existed in written form, i.e. that the age of the written text is crucial; b) that Landnámabók is an (unimpeachable) historical source and the sagas of Icelanders are not; c) that authorial intent in a saga reduces the likelihood of oral origins; and d) that, where there are discrepancies between Landnámabók and a saga, the most probable explanation is that there was once an older version of the saga, now lost, which was slightly different from the one that has been preserved. In the case of the connections between Landnámabók, Vatnsdœla saga, and Finnboga saga ramma, all these assumptions are open to question, since the verbal correspondences between these three works are never such as might allow us to state with any certainty that the correspondences of subject matter go back to the use of written sources (see below). This is also the conclusion of Margét Eggertsdóttir in a recent general article on Finnboga saga: after discussing the conventional view of the relationship between Finnboga saga and Vatnsdœla saga through literary borrowing, she concludes, ‘but it is equally plausible that the two versions represent oral variants of the same story’ (1993:194).

In his introduction to the Íslenzk fornrit edition of Finnboga saga, Jóhannes Halldórsson (1959) compares the two sagas at some length, appearing to take for granted the existence of a real historical Finnbogi in the 10th century on the grounds that he is mentioned in Landnámabók and Íslendingadrápa. However, the fact that a person by the name of Finnbogi is mentioned outside just a single saga proves nothing about the existence of a real person in 10th-century Iceland; all it shows, in fact, is that in 12th- and 13th-century Iceland there were various stories in circulation about a man of this name and his origins and deeds, stories that did not necessarily agree in matters of chronology, genealogy, names, or individual details, and which most certainly are not to be regarded as historically reliable sources. [6] Halldórsson thinks it unlikely that Finnboga saga was written using Vatnsdœla saga and therefore proposes a common source used by both. At this point he comes very close to accepting the idea of an oral tradition behind the saga which might even have had some basis in actual events several centuries earlier: ‘Tvenns konar munnmæli af gömlum rótum geta hins vegar skýrt ósamræmi nafna í sögunum, hvort sem um er að ræða aflögun upphaflegrar sagnar eða fyllingu í eyður’ (‘On the other hand, two groups of oral tales with old roots might explain the discrepancies in names in the sagas, whether it be by distortion of an original tale or by filling in the gaps’) (ÍF XIV:lxiv). But having got this far, he immediately retreats from this idea by adding: ‘Gera má ráð fyrir, að munnlegar sagnir Víðdæla hafi verið heldur fáskrúðugar eins og sagnirnar í Vatnsdal og höfundur Finnboga sögu hafi lagt margt til frá eigin brjósti’ (‘We can suppose that the oral tales of the people of Víðidalur were rather meager and sketchy, as with the tales in Vatnsdalur, and that the author of Finnboga saga supplied much of his material out of his own head’ ) (ibid).

In spite of the several similarities between the two sagas, the differences are greater in number and even more compelling in nature. Apart from the fact that each saga takes the side of its own hero in order to elevate him at the expense of his rivals, they reveal striking differences in their artistic approach and handling of the material. With some justification, Vatnsdœla saga may be said to be the more polished piece of literature. Here we find mythological undertones and anything that does not serve the unimpeded unwinding of the narrative is suppressed, while Finnboga saga happily finds room to accommodate various extraneous pieces of information that appear to add little to the saga as a whole. The sagas thus represent two very different treatments of what is clearly a common core of material. Central to the thematic interests of Vatnsdœla saga are its descriptions of Finnbogi’s kinsman Bergr and the men of Hof’s struggle through the elements to get to the dueling ground, both incidents which are entirely absent from Finnboga saga. In Finnboga saga the main emphasis is on Jǫkull’s love life and his battles with Finnbogi and Þorkell.

Mythological Overlays in Hœnsa-Þóris saga

Hœnsa-Þóris saga (‘Saga of Hen-Þórir’) is one of the most intensively studied of all the sagas and has been investigated from just about all points of view that have enjoyed any favor within the field of saga research since scholars in the 19th century first started to take a serious interest in the source value of the ancient Icelandic texts. The particular interest in Hœnsa-Þóris saga stems from the fact that a part of its subject matter also appears in our oldest and most trustworthy source for the history of early Iceland, Ari fróði Þorgilsson’s Íslendingabók. According to Ari, the information he gives on the matter in question was told to him by Úlfheðinn Gunnarsson, the lawspeaker in the years 1108-16.

‘Discrepancies’ between Ari’s Íslendingabók and Hœnsa-Þóris saga

In chapter 5 of Íslendingabók, Ari tells of a feud between two chieftains from the west of Iceland, Þórðr gellir of the Dalir region and Tungu-Oddr of Borgarfjörður, which arose out of an incident in which Hœnsa-Þórir barricaded Þorkell Blund-Ketilsson in his home and burned him to death. The involvement of the chieftains stemmed from one of Tungu-Oddr’s sons having taken part in the burning and from Hersteinn, the son of Þorkell Blund-Ketilsson, being married to a niece of Þórðr gellir’s. Two battles ensued between Þórðr and Tungu-Oddr, first at the local assembly at Þingnes in Borgarfjörður, where a follower of Þórðr’s called Þórólfr refr was killed, and then at the Alþingi, where Tungu-Oddr lost some of his men. Hœnsa-Þórir was declared outlaw ‘ok drepinn síðan ok fleiri þeir <es> at brennunni váru’ (‘and later killed, together with more of those who were at the burning’) (ÍF I.1:12). This feud brought to light a flaw in the country’s judicial system and, as a result, in 965 the country was divided into quarters and quarter assemblies introduced, with three local assemblies in each quarter (except in the Northern Quarter, where it proved necessary to have four local assemblies because the men of the north demanded separate assemblies for, east to west, Húnavatn, Skagafjörður, Eyjafjörður, and Þingeyjarsýsla). It is this constitutional change that accounts for Ari’s interest in the affair.

Ari’s account and Hœnsa-Þóris saga disagree on three points:

In Íslendingabók it is Þorkell Blund-Ketilsson who is burned to death; in the saga it is his father Blund-Ketill.

In Íslendingabók, so far as can be seen, Hersteinn Þorkellson was already married to Þórðr gellir’s niece at the time of the burning; in the saga, Hersteinn Blund-Ketilsson contrives to win the support of the powerful Þórðr gellir by marrying his niece directly after and as a result of the burning. Hersteinn’s schemes are directed by his foster father, Þorbjǫrn stígandi, of whom it is said that he ‘væri eigi allr jafnan, þar sem hann var sénn’ (‘there was more to him than met the eye,’ i.e. he had control of supernatural forces) (ÍF III:24).

The sources disagree on the numbers of those killed in the battles between the chieftains. The saga speaks of four of Þórðr’s men in the first encounter (including Þórólfr refr) and one of Oddr’s, and of six of Oddr’s men later in the battle at the Alþingi.

Source value, literary relations, the part of the ‘author,’ and social comment

The superiority of Íslendingabók over Hœnsa-Þóris saga so far as historical source value is concerned was established as long ago as 1871 by the German legal expert, Icelandicist, and folklorist Konrad Maurer. Since then, the dominant view among scholars of the Icelandic school, notably Sigurður Nordal (1938) and Jónas Kristjánsson (1973; 1978:325-6), has been that the ‘author’ of the saga created his work based on a meager selection of pre-existing sources, especially Íslendingabók but also lost versions of Landnámabók and perhaps some written genealogies. Scholarly interest then focused on, first, why the author of the saga had deviated from Íslendingabók, and subsequently the nature of the relationship between the saga and Landnámabók, especially the Sturlubók and Hauksbók texts.

This research is based on a number of presuppositions, chiefly that the only way that information and verbal correspondences could have been passed on was through written texts, and that no knowledge was available until it was first written down on parchment, after which it would have been swiftly disseminated throughout the whole country. As has been demonstrated above, there is good reason to believe that these presuppositions are simply not true and highly misleading so far as research into ancient writings is concerned. It is only in a minority of cases that similarities between sagas are best explained through literary relations and appeal to the image of an individual author creating his saga on the basis of material taken from other written texts. It is generally more profitable to assume that just about any kind of information could have existed and circulated within an oral tradition of historical narrative, and that its form within this tradition would in large part have determined its eventual written form.

Very much in the spirit of the Icelandic school, Björn Sigfússon (1960–63) maintained that Hœnsa-Þóris saga could in part be read as a justification for a new legal provision included in Jónsbók, the lawbook imposed on Iceland by the king of Norway and ratified at the Alþingi in 1281, that chieftains had the right to sequester and distribute other people’s hay in times of need. This is precisely what the noble chieftain Blund-Ketill does in the saga, following which the devious Hœnsa-Þórir uses legal chicanery to have him declared a thief, with tragic consequences. According to Árni saga biskups (‘Saga of Bishop Árni,’ pp. 812-3) this provision was hotly disputed at the time and was among the items that the Icelanders asked in a petition to the king to have revoked. The new measure was not, however, something made up purely by the king himself but rather a reflection of an international movement in 13th-century theology concerning public duties of mutual assistance. It is not known whether learned men in Iceland were aware of these ideas earlier in the century, but there is no doubt that they are strongly represented in Jónsbók, and on this basis Sigfússon concluded that the saga was written shortly before 1281, when the preparations for and discussions around this change in the law were in full swing.

In a radical departure from the methods and preoccupations of the Icelandic school, Theodore M. Andersson (1964:104-11; 1967:111-21; 1978:152-8; see also p. 39 above) has provided a powerful critique of the scholarly premises behind the idea that all similarities between sagas are to be explained as the result of textual borrowing. In this particular case, Andersson presented a number of reasons to doubt the claim of Jónas Kristjánsson (1978:325) that the story related in the saga was unlikely to have existed in the form of oral tales. For example, we may consider the differences between the saga and Íslendingabók noted above: these consist in large measure of confusion between names and discrepancies in the numerical details and order of events, all matters that are very likely to undergo modification in oral preservation but which are generally transmitted accurately by authors who get their information from written sources: see p. 27.

Hœnsa-Þóris saga in interaction with oral tradition

Andersson’s interest in Hœnsa-Þóris saga as the product of an oral tradition arises from his attempt to build up a picture of the overall structure of the prototypical oral saga. But there are also thematic and structural elements within the saga that are instructive on the ways in which material from an oral tradition was shaped and reworked in the creation of the written sagas. Of particular interest here are the comments of Carol Clover (1982:123-4) on the literary techniques that characterize Hœnsa-Þóris saga when compared to Ari’s more historiographical account in Íslendingabók. As an example, we may consider the contrasting presentations in the two texts of how Hœnsa-Þórir eventually gets his just deserts. In Ari, the killing of Hœnsa-Þórir follows his sentencing at the Alþingi and is the simple and logical outcome of his earlier actions. The denouement in the saga is rather more complex. At this point in the saga the action unfolds on two levels, with two lines of narrative running concurrently to create a tension and at times thwart or delay the expectations of the reader in a way Clover considers characteristic of written as opposed to oral literature: while the chieftains wrangle at the Alþingi, Hersteinn Blund-Ketilsson takes the law into his own hands and kills Hœnsa-Þórir back at home in his own district, justifiably, to be sure, but without the formal sanction of the law. As will be shown below, this narrative technique forms part of a more general tendency found throughout the saga.

By assuming that Hœnsa-Þóris saga arose from a living oral tradition, new ways of investigating and interpreting it open up to us. Within this tradition a whole range of material coexisted side by side and in no particular order—stories, poems, traditional lore, and, by no means least, myths. The connections between the myths and the written sagas have been widely investigated; notable contributions include those of Magnus Olsen (1928), Anne Heinrichs (1970), John Lindow (1977), Joseph Harris (1976), Preben Meulengracht Sørensen (1986), and Haraldur Bessason (1977). Bessason, for instance, demonstrates how patterns and motifs from the myths are used in the sagas of Icelanders and kings’ sagas to add emphasis to the events described and broaden their perspective; thus, a saga dealing with ordinary people but containing mythological elements and allusions gains added significance from the mythological parallels and invites interpretation through the eyes of the myth. Bessason calls such features in the sagas ‘mythological overlays.’ More recently, this idea has been extended by Margaret Clunies Ross, who in two books (1994, 1998) investigated particularly how the mythological tradition resonates in the narrative techniques of the sagas and imbues ‘mundane’ works with special significations.

If Hœnsa-Þóris saga is read in this light, a number of interesting parallels to myths emerge and we find ourselves better able to focus on the main issues in the saga without being sidetracked by matters that have hitherto tended to obscure them, such as the preoccupation with its historicity vis-à-vis Íslendingabók, its dating with regard to the legislative background, and the rights and wrongs of the individual characters in light of the strict letter of the law. For instance, within the context of the saga it is quite clear that, however much right Hœnsa-Þórir may technically have to refuse to sell his hay, this right rests on extremely dubious moral grounds. Hœnsa-Þórir’s actions explain why it was necessary to introduce an amendment like the one in Jónsbók to allow chieftains to organize the allocation of hay in bad years, while the saga also provides a graphic illustration of the background to the amendment to the constitution designed to bring about greater equity in legal cases through the division of the country into juridical quarters. Thus, one theme of the saga is a belief in the law as an instrument for settling the kinds of problems that will always arise in an imperfect world that includes petty crooks like Hœnsa-Þórir and bullies like Tungu-Oddr.

However, the saga also illustrates a further attitude to the law, and one of no less importance, concerning its inextricable relationship with power and status: only chieftains are in the position to go to law to settle their disputes; lesser men generally need the support of greater men to get satisfaction, as when the farmers turn to Blund-Ketill in their hour of need, and when Hersteinn enlists the support of Gunnarr Hlífarsson and Þórðr gellir. For ordinary people, the only solution is often to turn to direct means, as when Hersteinn kills Hœnsa-Þórir while the chieftains are still arguing the case in more civilized fashion at the Alþingi.

By rejecting the working premises of the Icelandic School, we also free ourselves from the need to account for the knowledge of local geography displayed in individual sagas and what this says about their supposed ‘authors.’ The writer of Hœnsa-Þóris saga has generally been given rather poor marks in this respect, since the morning after the burning in Örnólfsdalur he portrays Hersteinn and Þorbjǫrn stígandi as setting out to Breiðabólstaður to elicit Oddr’s support, then back to Örnólfsdalur, and then on to Svignaskarð driving all Blund-Ketill’s livestock before them—a combined distance of over 40 km (25 miles) as the crow flies. By the evening, Hersteinn and Þorbjǫrn have made it all the way to Gunnarsstaðir in Hvammsfjörður. To people familiar with the area, this journey seems highly improbable, and as a result the author has been accused of ignorance or inaccuracy in his treatment of the material. But such remarks ignore the simple device employed by the writer, of giving Þorbjǫrn supernatural powers, especially in respect to travel, as his nickname ‘stígandi’ (‘Strider’) suggests. Such a man can be sent around the district at whatever speed is necessary; all that matters for the plot of the saga is that he and Hersteinn should be able to arrive at places before the news of Blund-Ketill’s burning.

The killing of Helgi and the death of Baldr

A particularly clear example of the use of mythological overlays in the saga occurs at the death of Helgi Arngrímsson. Helgi is a young lad who is only present by chance when Hœnsa-Þórir leads his party to Örnólfsdalur to summons Blund-Ketill for the theft of his hay. In the confrontation, Helgi happens to be struck down by an arrow shot into the air without intent to kill. This incident directly parallels the death of Óðinn’s son Baldr, ‘inn góði áss’ (‘the good god’), killed unintentionally by an arrow made of mistletoe shot by the blind god Hǫðr at the instigation of Loki. Pointing the parallel to the myth still further, Hœnsa-Þórir kneels over the boy’s body and claims to catch his dying words, heard by no one else, urging them to burn Blund-Ketill in his farm; visually, this is strikingly similar to the way Óðinn leans over Baldr’s body as it lies on the funeral pyre and whispers secret words into his ear of which he alone is witness.

Helgi is an innocent and noble-hearted young man, the mouthpiece of truth, conciliation, and nonviolence in the saga, a role echoed in his name, i.e. ‘holy.’ The parallels with the death of Baldr make it natural to interpret the significance of Helgi’s death to the saga as a whole in a similar light. With Helgi’s death, the good in the world is destroyed and cannot be recovered without atonement through great slaughter, just as happens in the world of the gods at Ragnarǫk (i.e. the apocalyptic battle at the end of the world). This motif of the unintended killing of an innocent man occurs widely in the sagas, even through an unpremeditated shot as here (cf. the death of Þiðrandi, p. 224 above, and Mageröy 1991), demonstrating the power that this image exerted over the minds of saga audiences and their awareness of its mythological overlays, without there being any reason to assume deliberate borrowing from a particular myth.

The root of evil and dual structure: Hœnsa-Þóris saga and Vǫluspá

There are striking parallels between Hœnsa-Þórir himself and the malevolent trickster god Loki. Loki is not part of the family of the gods, though he lives among them and seems to have considerable influence over them; Hœnsa-Þórir has no family connections and thus no status within society, but he is rich enough to have acquired considerable influence in human society. Loki, through his actions, is the instigator of evil; he is responsible for most of the difficulties and disasters that disrupt the world of the gods and provokes the gods into committing evil deeds themselves, generally through working on their avarice and lust for gold. Gold—wealth—is also at the root of evil in the saga, as, for instance, in the methods used by Hœnsa-Þórir to bribe his way to what he wants and destroy the unity of the chieftains of nobler background. Hœnsa-Þórir’s wealth arouses the envy and cupidity of Arngrímr, which leads to the loss of his son. Similarly, later in the saga it is Þorvaldr Tungu-Oddsson’s willingness to be bought by Hœnsa-Þórir’s money that leads him to his banishment and death.

With the killing of Helgi and the burning of Blund-Ketill, the powers of good are extinguished and the problem faced by those that remain is how to contain the evil that has been released and restore equilibrium and justice. After the burning, the feud ceases to center on Hœnsa-Þórir and his or others’ right to the hay, and thus specifically on the vengeance for Blund-Ketill. Matters start to escalate, bringing in people from outside who are not directly concerned. Two local chieftains, Þórðr gellir and Tungu-Oddr, compete for power and the issue now becomes one of the methods that are allowable in order to achieve justice when the law itself is imperfect. Here a parallel can be drawn from the structure of the edda poem Vǫluspá as analysed by Haraldur Bessason (1984, 1999). The poem, like the saga, presents its themes through two related levels. The entire action occurs on the level of the gods where the great events take place, but there are repeated glimpses into the world of man that show humans behaving in exactly the same way as the gods, only in rougher and coarser form; what is only suggested of the gods is said directly of mankind. Thus it is that while, in the saga, the chieftains and their supporters come face to face at the Alþingi and battle seems imminent, the instigators of the feud are elsewhere and settle matters among themselves in their home region through simple recourse to bloodletting. On the saga’s higher level, this is a conflict between great men and regions; on the lower level, it is a dispute between private individuals. However, there is an essential difference between the mythic world of Vǫluspá and the human world of the saga: in the myth, everything is destroyed in the fires of Ragnarǫk, from which only the innocent emerge unscathed; in the world of man, however, there is governance through law to act as a mediator and thus avert catastrophe.

Another aspect of the saga’s judicial message is that just laws do not work on their own when their enforcement is dependent on the whims of chieftains who cannot be relied on to act with wisdom and impartiality. The saga makes it quite clear that rules and laws are meaningless unless they are followed up and enforced by those who have the power to do so. Jesse L. Byock (1982) has put forward the view that the sagas of Icelanders can be read as a kind of didactic manual on the handling of disputes: see also the ideas of Eric Havelock (1963) discussed earlier, p. 57. Byock shows how disputes originate when ordinary people attempt to assert their rights and then escalate to a point where chieftains take over, just as in Hœnsa-Þóris saga. In the saga, characters are repeatedly forced to look to chieftains for support in their troubles and to ensure that their rights are safeguarded:

The Norwegian merchant Ǫrn is intimidated by Tungu-Oddr regarding the pricing of his goods, in blatant contravention of the legal provisions covering such matters, and so turns to Blund-Ketill for support.

Hœnsa-Þórir attempts to raise his social status to reflect his newly acquired wealth and to this end buys his way into the favor first of Arngrímr goði and later of Þorvaldr Tungu-Oddsson. But money leads only to evil—a fact copiously exemplified in the ancient myths and heroic legends.

Blund-Ketill’s tenant farmers are motivated by need and a sense of their rightful dues when they turn to him when their hay runs out.

Hersteinn is forced to use cunning and his foster father’s magical powers to elicit the support of chieftains in his own quest of justice, because the chieftain whose actual responsibility it is to ensure local justice, Tungu-Oddr, is ruled only by his personal interests and sentiments, having fallen out with Blund-Ketill in the matter of the Norwegian merchant and because his own son is implicated as one of the arsonists.

By focusing on the mythological overlays in the construction of Hœnsa-Þóris saga, we obtain a clearer appreciation of the conscious and deliberate way in which the structure of the saga operates as a whole, right from the prelude about the merchant Ǫrn to the marriage arrangements at the end. Thus the ground is carefully prepared for Hersteinn’s need to look beyond his local chieftains for support, and thus for the introduction of a chieftain from outside, Þórðr gellir, by creating an enmity between Blund-Ketill and Tungu-Oddr arising out of the case of the Norwegian merchant Ǫrn. This enmity also calls for a particular solution once the case against the arsonists has been brought to completion. The resolution is reached through negotiations among the great chieftains at the Alþingi, who arrange for the marriage of Tungu-Oddr’s son Þóroddr to Jófríðr, the daughter of Gunnarr Hlífarsson, who lived in Örnólfsdalur where he had built with timber originally brought to Iceland by Ǫrn. Thus both the beginning and the end can be seen as fully integrated into the overall structure of the saga and play essential parts in the train of events described, creating a tension between the characters that raises the saga to the political level, i.e. the level of great chieftains, and demonstrating how the warring parties can eventually be reconciled.

Conclusions

By identifying links in this way between Hœnsa-Þóris saga and the common tradition of learning passed on orally, it is not necessary to adopt a position on matters such as authorial use of sources and the relations between the text and the various versions of Landnámabók— except perhaps to reject the reliability of conclusions which have been arrived at through the scholarly means available. Similarly, this approach does not seek to justify oral origins for the saga by attempting to draw parallels between its structure and that of the prototypical oral saga or by identifying other aspects considered to be typical of the narrative techniques of oral storytelling. There would be no point in this, since the sagas of Icelanders are written literature and not oral tales set down in writing. The only presuppositions taken here have been that the subject matter of the saga was well known in oral tradition and that the saga was intended for an audience that was familiar with the full range of Norse myths and traditional tales, and that it must thus be read in the context of this tradition if we are to obtain a coherent grasp of its full meaning.

Hœnsa-Þóris saga employs parallels with and allusions to myths in order to highlight and bring out the significance of major events. By such means it passes stern judgment on the bullying and overbearing of chieftains and the imperfections of the law and a judicial system governed not by justice but by political influence and the machinations of individuals. It is perfectly reasonable to see connections here with conditions in the 13th century, when the saga was written. Faced with the anarchy of the Sturlung Age it would have been only natural for people to harbor dreams of a strong and centralized government, able to enforce justice and maintain peace free from the personal failings and interests of individual chieftains, under which, in order to assert their rights, people did not find themselves thrown on the mercy of fallible and self-interested powerbrokers. And since rules and laws are never enough to put curbs on human nature, in the end, the saga concludes, it is down to love to act as a final check on man’s injustice.

New Growths on Ancient Roots

The sagas of Icelanders are written literature, mostly preserved in manuscripts from the 14th and 15th centuries and telling of events that are supposed to have taken place in the 9th, 10th, and 11th centuries. Beyond this there is no certainty. But this is not the whole story, since the written sagas that we have show signs of drawing on an older oral tradition of stories and historical lore, though without anyone having any idea what part this tradition played in the actual writing of the sagas. By an oral tradition I mean a tradition of stories that has had no contact with writing such as would make it possible to speak of one recorded version of a story being more ‘correct’ than any other.

It has proved difficult to isolate any universally accepted method to demonstrate beyond doubt that sagas of Icelanders were based on an oral tradition. The sagas could, theoretically, be entirely the creation of imaginative clerics of the 13th century with a thorough background in Latin proverbs and theology, who wanted to direct the minds of their contemporaries to the blessings of Christianity and the determinism implicit in the heroic ideal. This is of course highly improbable, but imaginable nonetheless, just as it is also theoretically possible that men with fertile imaginations and an easy disregard for the truth got together in the 12th and 13th centuries to promote their own personal interests by concocting Landnámabók (‘Book of Settlements’) on the basis of existing place names, and that the characters they invented there were later used as the starting point for the sagas of Icelanders.

Because of this uncertainty surrounding the origins of the sagas and their links with oral tradition, many scholars have preferred to look no further than what has come down to us, the written literature, and confine their discussions to the texts and manuscripts themselves rather than indulge in unverifiable flights of fancy on the subject of origins and oral tales. But, as argued at length, this attitude provides a false sense of security and gives a misleading picture of the old Icelandic sagas.

As with all types of ancient literature, it is necessary to adopt a conscious position on the question of the origins of the sagas and their connections with oral tradition before it is possible to say anything at all about the texts themselves. Once we have dealt with the ‘hard facts’—the paleographic and linguistic forms of the manuscripts and the internal connections among related manuscripts—everything else that we say on the texts, such as regarding original forms, lost variants, and their relationship and preservation, is contingent on the position we take, whether consciously or unconsciously, on the insoluble ‘big’ question of the role of orality in the writing of the sagas as we know them.

There are two final points that need to be stressed. Firstly, the sagas of Icelanders are not direct records of oral tales. Oral tales are fluid and reshaped on each occasion by the particular performer and the particular audience. A very large part of their artistry is ‘performance art,’ something which is lost in recording (to say nothing of the physical impossibility of setting down on parchment or paper the words and sounds, let alone the gestures, of storytellers in the actual flow of performance). In order to record an oral story, the storyteller must dictate his story to a scribe word by word; but this is an unnatural method of delivery and would fail utterly to reproduce the way a story would be delivered in actual performance, at natural speed, in front of an audience. Accurate, verbatim records of stories from oral traditions were thus not possible before the advent of modern sound recording techniques. The second point is that oral tales are not necessarily ‘truer,’ more factually correct, than any other creative work of literature—though they can have roots in historical reality, as has, I hope, been adequately demonstrated in the preceding chapters.

Perhaps the main problem we face when discussing the links between the sagas of Icelanders and oral tradition is that we have no agreed methods on which we can rely to point the way forward. It is, for example, irrelevant that some sagas repeatedly make reference to oral tales, or say that one man said this but others say something different; references of this type could easily all be a stylistic device of the scribes, aimed at convincing their readers that their work was based on genuine sources and their accounts reliable. Neither does it tell us much that the sagas frequently make use of fixed formulas, stylistic techniques, and prestructured themes of kinds that we generally associate with oral narrative; such oral features can equally occur in purely written works, put in by authors who are making up their stories for themselves without any material basis in oral tradition.

Since the available methods cannot provide us with unequivocal answers, it is right to repeat just how little we really know, and how little we can know. But even so, we have to take on certain assumptions about how the sagas came into being if we are going to investigate them at all. We need to nail our colors to the mast on questions to which there are no black-and-white answers. It is not possible simply to ignore these questions, to leave them unresolved and imagine that by doing so they do not exist. Simply by asking particular questions of the manuscripts and texts we are adopting a position on the origins of the sagas, and there is thus no completely objective, scientific way of investigating them. All research depends on how we answer the question of origins, and the answers we give will always be based on personal opinions and theoretical attitudes. But even if the theory we adopt is widely accepted and enjoys broad scholarly support, it remains a theory, not a fact.

Repercussions

The role of the medievalist in literary research differs from that of the specialist in modern literature. Ancient texts demand to be studied within their proper cultural context. We therefore need to try to find out as much as we can about their backgrounds in order to understand how they may have been affected by their contemporary cultures and how they were viewed by their original audiences or readers. Where our knowledge ends, we need to make educated guesses, since the fact that we do not know something does not allow us to suppose that there is nothing there to be known. Even though the sagas of Icelanders have stood the test of time and remain the most widely read literature in the language, they arose out of a particular cultural environment and are bound to a particular time and particular conditions. They work on certain principles and assumptions and were written for people with certain types of knowledge and ways of thinking.

The answers we give on the questions of origins will, for example, determine how we go about interpreting a character in one saga who also appears in other sources; see, for instance, the discussion of the characters in the sagas of the east of Iceland in part II above. If we take the view that the sagas were grounded in an oral tradition, we have to assume that their audiences already possessed a certain amount of knowledge about the people who turn up in them. Each saga then becomes a link in the unrecorded, ‘immanent’ tradition as a whole, something we can now approach only at several removes by reading the sagas, Landnámabók, the myths, and other ancient writings whose material was plundered from the tradition as it existed at the time when these works were written. If, on the other hand, we imagine the sagas to be creative fictions produced by individual authors, we have to assume that all the material that is of any relevance to a saga and its characterization is available from the text itself; such an approach also brings with it the need to arrange the sagas into chronological order and entails the supposition that no knowledge was available until it had been written down and then came little by little into wider circulation.

In other words, whichever way we choose to view the sagas, we have to take account of a background of some sort. Studies like Óskar Halldórsson’s on Hrafnkels saga (1976) have demonstrated how much there is to be gained from applying the working hypothesis that the background to the sagas lies in an inheritance of oral tales. To understand the interaction between traditional tales and written literature we again need a working hypothesis, and here there is much to recommend the revised version of the ‘þáttr theory’ proposed by Carol Clover (1986), which appears to be broad enough and to rest on solid enough foundations to bring scholarship some way forward toward a valid reassessment of our attitudes toward the sagas.

Once we have decided to allow for the effects of a living oral tradition, it becomes possible to examine the learned influences in the sagas in light of what they may have added to this tradition and what their role was in the creation of the finished works of written literature. Viewed this way, the story tradition mainly provides the material and the learned influences the form. But the two are inextricably linked, making it necessary to interpret the sagas as a whole as the works of single individuals created against an assumption that their audiences would know other stories about the characters and settings mentioned in them.

However, if we assume oral origins, we can no longer think of an author in the modern mold, creating something out of nothing, or improvising around a brief grain of historicity culled from Landnámabók, or perhaps sitting down to construct a literary roman à clef in which contemporary events are shifted back in time and archaizing elements consciously introduced—though whatever view we take of saga origins we can readily admit that, even though stories may be set in the past, they will always bear the marks of the times in which they were created. Similarly, if we accept oral origins we have to discard the image of the saga writer so dear to the editors of Íslenzk fornrit, as the performer of a kind of literary ‘cut and paste’ job, collecting snippets of information from miscellaneous written sources and patching them together into a single work.

How we interpret the frequent references in the sagas to oral tales and performance will also depend on the theory of origins we ascribe to. Bookprose scholars, for obvious reasons, are inclined to view such references as adventitious insertions put in by authors with the aim of making their sagas seem more credible. But what is there to say that this was indeed the intention, or that the authors of the sagas in general thought in this way at all? If these writers were genuinely drawing their material from an oral tradition, it is only natural that they should present events as if there were always someone standing by to act as a witness; and if they say that accounts differ on what happened in their saga, they would only have been reflecting what their audiences knew to be true.

One of the most striking implications of a theory of oral origins for our overall view of the sagas and how they should be studied concerns their dating: if they were written against a living background of oral tales, the age of the written sagas becomes considerably less significant. These implications also extend to the methodology used in dating the sagas. The received methods are based largely on comparing supposedly literary relations (rittengsl)— correspondences of diction or material taken as evidence of borrowing between written sources. But such relations are seldom verifiable, and so, if information can also be transmitted through an oral tradition, it is only in rare instances that we will be able to say with any confidence that one written work took its material from another, and thus must have been written after it.

The writing of a saga did not suddenly produce something new for people to use and refer to, since the stories it contained existed before the time of writing and continued to exist after it. This is of great importance to manuscript research, which at present generally ignores the possibility that the oral tradition persisted alongside and influenced the written tradition. Copyists could have had access to oral tales as well as the texts they were copying, and used them in the versions they produced, and thus the interaction between the two traditions would have continued for some time after the original writing of the saga. During this period the saga would not have been the preserve of the manuscripts and their copyists but would have had a dual aspect and led a double life. [10]

Assuming the existence of an oral tradition behind a saga changes our whole attitude toward the significance of the details it gives on the genealogies and family connections of characters that are also mentioned in other sources, especially Landnámabók. What one saga says can no longer be viewed in isolation; we also need to take into account that people might have possessed a whole range of information on genealogy and events from other sources and used this to make sense of the alliances and enmities between the various characters. The fact that the preserved genealogies do not always concur is another matter, but not one that need create problems once we have freed ourselves from the perceived duty to search for historical veracity in works of creative literature.

Perhaps one way to come to grips with the concept of an oral tradition and what it entails is to consider the ancient mythology, in which each ‘episode,’ i.e. individual myth or incident, assumes of its audience a knowledge of the tradition as a whole—an understanding of why characters are as they are and what went before and what will come after. In all the stories we hear of Loki, for instance, we are aware of how he came to be among the gods, of how he was captured and bound for his misdeeds, and of how he will eventually escape and fight on the side of the forces of chaos at the end of time.

The myths must have formed a part of the same oral tradition as the sagas of Icelanders and we can assume that they were no less well known than the genealogies and tales of human feuding. We thus need to keep our eyes open as we read the sagas for allusions to and themes from the myths, as in the discussion of the mythological overlays in Hœnsa-Þóris saga earlier in this chapter. Allusions of this sort would hardly have been comprehensible if the myths had not already been well known to audiences from a common oral tradition; the myths did not need to rely on the vicissitudes of distribution of a small number of manuscripts to achieve general public currency.

Even though, by its very nature, an oral tradition is fluid and ever-changing, constantly adapting itself to the prevailing conditions, it is also a unity, continuous, unbroken, and integrated. Where such an unbroken tradition exists in a comparatively stable society, we can suppose that it contains real memories of past events and is able to give us a general picture of the past that can, to some extent at least, stand up to historical scrutiny. By assuming a tradition of this kind behind the written sagas it becomes possible to make reasoned comments on any historical truth that may have inspired the stories in the first place—though only on the principles of the tradition itself rather than with historical source criticism as our methodological tool.

Finally, it is extremely difficult to account for the special position of Iceland in the literature of the Middle Ages without appeal to a strong and flourishing tradition of oral storytelling. Specifically Icelandic conditions in the 12th and 13th centuries, and the powerful currents of Latin learning and continental European culture that were felt in Iceland at this time, do little to explain the literary outpouring through which the sagas of Icelanders came into being. In all main respects, the sagas are a perplexing and inexplicable phenomenon unless there was a domestic tradition of oral tales lying somewhere behind them.

To explain why the art of oral recitation attained a completely different and more sophisticated level in Iceland than in Norway, it is worth considering the Gaelic cultural influences which were brought to Iceland at the time of the settlements and which may have laid the grounds for a much more powerful culture of storytelling and oral poetry than developed elsewhere in Scandinavia (see Sigurðsson, G. 1988). Thus our attitude toward Gaelic influences on old Icelandic literature is determined by the position we take on orality behind the sagas—providing yet another example of the way in which the premises we start out with affect the questions we can ask and so the conclusions we can reach. By allowing for an oral tradition, the Gaelic contribution becomes a workable means for explaining what was special about Iceland. If, however, all we are interested in is hunting down verbal loans and literary relations there is no place for Gaelic influences in the Icelandic saga tradition—and as a result scholars of the stature of Jón Helgason and Sigurður Nordal saw no reason to interest themselves in them in any way; there was simply no place available for them within the framework they set themselves when conducting their research.

*

There is no avoiding the question of where the Icelandic sagas come from, since the way we answer this question shapes all our other research into them. But, however much significance we ascribe to a strong tradition of oral storytelling in the background, it is right always to remember just how remarkable a task it was to bring the diverse range of stories and other information together in an organized fashion and turn it into written works of literature. This is where the saga writer comes in, armed with the pan-European tradition of learning he could find reflected in Latin writings of the 12th and 13th centuries and carried forward by the domestic written tradition that began to develop in the first decades of the 12th. In other words, oral origins in no way detract from the importance of the learned and written traditions; the sagas of Icelanders are written literature, produced under the influence of the Latin schools and the common literate culture of medieval Europe, an influence which did not preclude but rather reinforced and augmented the domestic tradition. The learning acquired from abroad provided the technical tools that were needed for this domestic cultural heritage to find expression in written form. And although throughout this study we have concentrated our attention on the sagas as the outcome of a common oral tradition, it should never be forgotten that each saga is also an independent work of art, produced by a single writer who chose to treat precisely this material and present it in precisely this way in all probability because he wanted to tell his contemporaries a story of his own and interpret the world he saw around him for an audience of his own age, exactly as his predecessors had done for centuries before him when they told stories about the same people and the same events, each in their own way—and perhaps not always with the happy disregard for truth of which we are sometimes inclined to accuse them.

Footnotes

[ back ] 1. On the relations between the two sagas, both internal and with other sources, see Gering 1879:xxxiii-xxxix; Björn M. Ólsen 1939:243-9, 336-48; Vogt 1921:xxxii-xxxiii; van Hamel 1934; and Halldórsson, J. 1959:lix-lxvii. A review of the literature, including a detailed comparison of the various factors involved, can be found in an unpublished BA thesis by Ásdís Haraldsdóttir 1980. Ásdís herself comes down in favor of the view that oral accounts played a significant part in the composition of the two texts.

[ back ] 2. For an excellent discussion of similar approaches to saga research, with particular reference to Hávarðar saga Ísfirðings, see Thorsson 1990:38f.

[ back ] 3. Einar Ól. Sveinsson puts the differences between Landnámabók and Vatnsdœla saga down not to a rewriting of the saga but to adaptation on the part of the copyist: ‘sá maður sem skrifaði þá skinnbók, sem núverandi handrit eru komin frá, hafi lagað textann í hendi sér heldur í meira lagi, en án þess þó að mér þyki brýn ástæða að ætla, að hann hafi skapað með því nýja gerð sögunnar’ (‘the writer of the vellum from which the surviving manuscripts derive adapted the text he had before him with considerable freedom, though without there being, in my view, any compelling reason to suppose that he thereby created a new version of the saga’) (Sveinsson 1939:lv).

[ back ] 4. Íslendingadrápa is often believed to be from the 12th century and refers to several characters who are also known from the sagas. For opposing views on the age and source value of this poem, see Kristjánsson 1975:76-91 and Einarsson, B. 1989:127-31.

[ back ] 5. This view of the purpose behind the writing of Finnboga saga ramma echoes those of perhaps the three most important names of the ‘Icelandic school’ of saga studies, Björn M. Ólsen, Sigurður Nordal, and Einar Ól. Sveinsson. Ólsen (1937-9:340) writes: ‘missagnir þær, sem eru, sjeu að minsta kosti sumar hverjar sprottnar af því, að höfundur Finnboga sögu hafi af ásettu ráði vikið frá sögu Vatnsdælu, þar sem honum þótti hún bera Ingimundarsonum of vel eða Finnboga og fjelaga hans Bergi rakka of illa söguna, og breitt frásögninni sínum söguköppum í vil’ (‘at least some of the differences that exist arise from the author of Finnboga saga having deliberately deviated from the account in Vatnsdœla saga because he thought the latter presented the Ingimundarsons in too favorable a light or Finnbogi and his companion Bergr rakki too unfavorably and so altered his narrative to the advantage of his own heroes’); Nordal (1953:268): ‘En sammenligning med Vatnsdœla viser, at de to forfattere tager ivrigt parti for hovedpersonerne i hver sin saga, men Finnboga s. er øjensynlig den yngste og kan betragtes som en slags modskrift mod den anden. Finnboga s. er antagelig skrevet í Víðidalr i Húnavatnsþing og giver udtryk for jalousi mellem to bygder’ (‘A comparison with Vatnsdœla saga indicates that the two authors enthusiastically take the parts of the main characters in their respective sagas, but Finnboga saga is plainly the younger and can be viewed as a kind of refutation of the other. Finnboga saga ramma was presumably written in Víðidalr in Húnavatnsþing and gives expression to a rivalry between the two districts’), see also Nordal, S. (1920:129); and Sveinsson (1939:xiv), that Finnboga saga was written ‘til að rétta þann krók, sem Víðdælum var beygður í Vatnsdæla sögu’ (‘to put right the slights on the character of the people of Víðidalur inherent in Vatnsdœla saga’).

[ back ] 6. In later centuries rímur (late medieval rhymed narrative poems) were composed about Finnbogi including material that cannot be traced directly to the written version of the saga. This led Ólafur Halldórsson (1975) to suggest that there had once been an older version of Finnbogarímur, now lost. Some of this material was also known in the Faeroe Islands, where the story was relocated and adapted to fit local conditions: see Poulsen 1963. Both these facts seem to support the idea that Finnbogi was a well-known character in oral tradition and that stories about him enjoyed wide currency, though whether this goes back to oral narratives or written sources is impossible to determine.

[ back ] 7. There is probably a deliberate contrast here between the related words ofláti, n. literally, someone who behaves too big, and stórlátr, adj. behaving in a big, i.e. generous, manner.

[ back ] 8. Going under a jarðarmen (lit. ‘necklace,’ ‘halter of earth’) is mentioned in several sagas as a ritual means of solemn confirmation. In this case it clearly denotes submission or abasement (cf. ‘going under the yoke’), though it also appears in other ceremonial contexts, e.g. the formalization of blood-brothership in Gísla saga and Fóstbrœðra saga. The jarðarmen appears to have been a strip of turf cut from the ground but with both ends still attached that the participant had to walk under, in this case needing to stoop to do so. Exactly what ‘swine-bow’ (Icel. svínbeygja) means is unclear, but it patently implies gross humiliation, with either sodomistic or bestial connotations.

[ back ] 9. Another marked tendency among recent commentators has been to examine the saga for signs of some kind of extenuation for the villain Hœnsa-Þórir, particularly through close reading of the text and investigation of the legal complications. Examples include Mundt (1973) and Berger (1976), who study the legal position of the characters as presented in the saga; Miller (1986), who assesses Blund-Ketill’s impounding of Hœnsa-Þórir’s hay in light of ideas about reciprocal gifts and the rights and obligations thereby bestowed (see also Durrenberger et al. 1988; Miller 1990:93-101; Þorláksson 1992); and Baumgartner (1987), whose detailed critique of the saga centers on the social standing of the characters and Hœnsa-Þórir’s aspirations to acquire status in keeping with his acquired wealth. Baumgartner’s reading is based solely on the written text, but his highly exacting approach adds much to our understanding of Hœnsa-Þórir’s situation and motivation.

[ back ] 10. See Slotkin 1979 on similar considerations with regard to the ancient Irish sagas.

About

Fellowships

The Library

Programs

The Medieval Icelandic Saga and Oral Tradition: A Discourse on Method

8. Implications for Saga Research

The Feud between Finnbogi the Strong and the Men of Hof

Conflicting openings to the feud

The winter wedding in Vatnsdalur

The duel from conflicting perspectives

The end of the affair

Independent chronologies

Conclusions

Mythological Overlays in Hœnsa-Þóris saga

‘Discrepancies’ between Ari’s Íslendingabók and Hœnsa-Þóris saga

Source value, literary relations, the part of the ‘author,’ and social comment

Hœnsa-Þóris saga in interaction with oral tradition

The killing of Helgi and the death of Baldr

The root of evil and dual structure: Hœnsa-Þóris saga and Vǫluspá

Conclusions

New Growths on Ancient Roots

Repercussions

*

Footnotes

The Center for Hellenic Studies	Contact Us
	Phone: 202-745-4400
3100 Whitehaven St NW	Fax: 202-797-8650
Washington, DC 20008	outreach@chs.harvard.edu