The Oral Background of the Eddas and Sagas

The Árni Magnússon Institute for Icelandic Studies, University of Iceland

More than fifty years after the publication of The Singer of Tales the book and the ideas therein still serve as a fresh wind in the textually-oriented field of Old Norse Studies where very few have taken up the issue of how an oral background for medieval texts can possibly affect the modern appreciation of them. It should be noted however that a notoriously high percentage of the exemptions to that general rule are contributing to this volume.

The lack of interest in the oral background is most deplorable and it may perhaps be explained by the fact that the majority of medievalists naturally focus their attention on religious and learned texts as these are a considerable part of the medieval textual production in continental Europe. Such texts have a much stronger background in booklearning than the orally derived poetry and secular narratives that make up the bulk of the Old Norse material. The approach of medievalists who are analyzing learned products of clerics has influenced scholars of Old Norse, who have in recent decades shown that the texts from the Old Norse world also show clear influence of Latin learning, theology, and the Christian book culture of the time, thus making them a part of the European scene. We now appreciate that this is what should be expected from cultural products of the time as opposed to regarding them as more or less purely oral and secular products of a nationally-defined Icelandic culture—as was perhaps an acceptable approach in the 19th century. The more interesting, from the present point of view, and the more problematic aspect of this Europeanization of the culture is how the texts also reflect the indigenous oral culture by which they were ultimately inspired (Sigurðsson 2004:21–48).

When we take the oral approach, everything which has previously been accepted as the received wisdom must be reevaluated: we must both focus on each manuscript version of a given text and give it full credit in every respect as the unique product which it is (both looking at the age of the manuscript and every meaningful variation in its social and personal context, in the manner of New Philology—an exciting example of which is to be found in the Homer multitext project, described in this volume) but at the same time we have to read the manuscripts as the products of a continuous oral tradition that can go far back in time. Thus we are dealing with texts that are young and old at the same time, reflecting contemporary interests and learning as well as some memory of old. And we cannot be certain which is which.

We have acquired this knowledge and awareness from the study of living oral cultures rather than from academic speculation about how such a culture could have been, speculation necessarily involving inference from incomplete written evidence and ideas about memory from within the medieval church, as opposed to using concrete evidence about living oral cultures as a point of reference. And this new knowledge should alter our approach to the texts, that is if we lived in an ideal world of learning where scholars can change their minds and methods when new evidence is presented to them—as we are all told we should do in our scholarly quest for the truth (the veritas which we see in the Harvard emblem). But rather than meeting the challenge of this new knowledge many either choose to ignore it and continue with their old editing ways (McGann 1991:15), or at the best reinvent some literary terms to be able to claim that they actually allow for all this, but within the framework of written texts and written traditions.

I have tried in various publications to apply the learning derived from oral studies to the Old Norse corpus: the Eddic poetry, the mythology as it is presented in Snorri’s Edda, and to the secular family sagas. I have not found it fruitful to focus on the formulaic language as such but rather to apply the comparative approach and use the idea of variation in an oral and manuscript culture. The variation reflects the interests, knowledge and talent of all involved in a performance—and as a result: the problems associated with the textualization of a primarily oral text.

In the case of Eddic poetry the scholarship has been based on the idea that these poems, which all agree were written down from oral tradition in the 13th-century, can and should be dated to different centuries from the 9th century onwards, based more or less on circular arguments and differences in language, style, form, and content. Even though the difficulties, if not impossibilities, of dating are dawning on more and more scholars, many have been reluctant to take the full oral comparative approach which would suggest that all these differences could be much better explained with different individual preferences of the singers and their audiences, based on their sex, social status, geographical surroundings, and other factors. When applying the oral approach we can thus explain the differences in the poems in terms of oral variation as a result of the conditions around their performance and writing, rather than as the product of a Darwinian literary development of taste in static and fixed poetic texts that were composed by an individual at some point in time and then committed to the imperfect memory of preservers until the time of writing. If we read the written poems from the thirteenth century as the products of an oral culture (as opposed to seperate works of authors in different ages that were memorized verbatim), we see before us variation of the main heroic stories or themes based on different perspectives that can be attributed to the differing interests of women and men as I have demonstrated earlier (Gísli Sigurðsson 1990, 1998), some focusing on the heroines and their emotions and others thinking more about the male heroes and their physical activities in a male dominated warrior context, still others reflecting special interests in the lower social classes, even showing some geographical preferences and so on. From that oral perspective we can classify and understand the similarities and differences between the poems in light of a multivoiced variation that is likely to have been alive at the time when these texts were written—rather than as the product of many different but singlevoiced time periods, each with their stylistic and thematic preferences.

When it comes to the mythology as it was recorded by Snorri Sturluson (1178/9-1241) in his Edda in the early 13th-century, more than 200 years after the official acceptance of Christianity in Iceland, it has been a constant source of bafflement how all these seemingly pagan ideas could possibly be recorded so long after paganism as a religious system was abandoned in the north. It is an achievement in itself to have this learning recorded in book form by a professionally trained, secular insider in the culture, rather than by an outside field worker or a cleric; recorded, that is, by an individual who realizes the potential of the book which he is the first to entrust with his oral learning. Scholarship has shown, as might have been expected, that Snorri’s ideas are not purely pagan by any means but heavily influenced by ideas and book learning of his time, and his text is not strictly oral either but presented in a very bookish way. Given that, the most interesting part of this achievement excapes us if we do not add the thinking of the oral background to our toolkit. Snorri is not a folklore collector, rounding up stories from informants, and he is not a scholar using scattered and poorly preserved poetry to reconstruct a system out of fragmentary information or sources, as a modern scholar would do. Rather Snorri was brought up and professionally trained orally to become a politician, lawman, and scaldic poet, a profession which required the mythological background to compose and understand the images and metaphors used in scaldic potery (Gísli Sigurðsson 2004:6–17). Snorri reached the highest level in all of his professions, not just as a poet, story-teller, and writer. He was a chieftain (goði) and lawspeaker in 1215-18 and again in 1222-31. A son of a minor chieftain, Snorri was fostered by the most influential man in the country at the time, Jón Loftsson in Oddi who had direct contacts with the Royal House in Norway. In connection with Snorri’s climbing up the social ladder, he spent time in Norway and Sweden in 1218-20 and became a friend of King Hákon and Earl Skúli in Norway. When Snorri returned to Norway in 1237-39 he was drawn into a conflict between Hákon and Skúli, who was revolting against Hákon and was killed in 1240. Snorri had backed Skúli in this confrontation and left for Iceland without Hákon’s consent. This was interpreted as treason, so the King had his men in Iceland execute Snorri at his home in Reykholt on September 23rd, 1241. Besides the Edda, Snorri is regarded as the author of a seperate saga about St. Ólaf (Ólafs saga helga hin sérstaka), Heimskringla (assembled sagas about the Kings of Norway from a mythological beginning down almost to Snorri’s time), and often promoted as the author of Egils saga, possibly the first family saga about the tenth century poet, farmer, and viking Egill Skallagrímsson—in addition to several scaldic poems attributed to him. But how could Snorri and all the professional skaldic poets like him keep track of the background mythological information required for their exceptionally complicated verbal art form?

Again the idea of using field work in cultures who still keep their mythologies alive orally comes to our rescue. The stories and the systematic presentation of them may have been held together by the world itself, the earth and the dome above, which the stories propose to describe in mythological terms: Snorri’s illusion in the Gylfaginning (lit. Gylfi’s Illusion) of the Edda turns the sky into the abode of the gods in front of the recipient of the wisdom within the written text. The mythological phenomena which the text explains are constantly referred to as being in the sky (77 obvious references). For example, the gods’ dwellings are in the sky (á himni), the world tree stands above the sky (yfir himni) and Gylfi (who calls himself Gangleri) claims that the three instructors, Hár, Jafnhár and Þriðji, tell great tidings about the sky (mikil tíðindi … af himninum). In the end, when the lesson is over and the illusion thus vanishes, the pupil Gylfi/Gangleri is left alone in the open. He is surrounded by the world around him as we usually see it with our bare eyes, and with none of the mythological illusions visible any longer. Rather than to see a palace and a city as before, Gangleri is in the middle of a field, probably with the ordinary sun and the moon, stars, and planets above him where the story-telling magic had previously made him see the dwellings of the gods and the world tree above. Snorri’s Edda, as a presentation of his knowledge, was no doubt intended and written for an exclusive audience, that is a literary/poetic elite intent on seriously studying the art of skaldic poetry. The Edda thus becomes understandable as the written product of an oral culture that did not let go of its cosmological knowledge and terminolgy even though the religious aspect of it had to yield to Christianity. The mythology therein would then be similar to mythologies around the globe, which all have an astronomical aspect to them, as becomes apparent when they are studied as a part of a living culture. [1]

A short, almost in passing, explanatory note in the 19th century Icelandic folklore collection of Jón Árnason might contain the confirmation needed in order for us to be able to actually proove that people in Iceland spoke of real and visible phenomena in the sky with mythological vocabulary in the 19th century. In the chapter on nature legends in the first volume of Íslenzkar þjóðsögur og Æfintýri (1862) it says on pp. 658–59, under the heading “Airvisions and Moontales” (Loptsjónir og Túnglsögur):

In a footnote Jón himself refers to Chapter 12 of Gylfaginning in Snorri’s Edda and stanza 39 of the Eddic poem Grímnismál in order to explain these wolves. Snorri’s Edda says of the sun and the wolves:

If it were not for the chapter in the Þjóðsögur of Jón Árnason no one would ever contemplate the possibility that this episode in Snorri’s Edda was explaining sundogs that can sometimes be observed on each side of the sun-halo that surrounds the sun in such conditions. The same can be said about the moon-halo which Björn Jónsson (1994?:185-88) thinks is being referred to in the previous chapter in Snorri’s Edda, where the characters Bil and Hiuki are said to be “carrying between them on their shoulders a tub called Sæg; their carrying-pole was called Simul. Their father’s name is Vidfinn. These children go with Moon, as can be seen from earth” (Faulkes 1987:14). [4] Björn suggests that the carrying-pole should be understood as the light beam that stretches between the moondogs—and the tub itself must then be the moon—“as can be seen from earth,” which is a very straight-forward statement in the text itself encouraging this kind of interpretation. Before Björn, Richard Allen (1899:267) had come up with this interpretation and suggested a parallel in Jack and Jill (i.e. Hiuki and Bil), whom native speakers of English know from the nursery rhyme: “Jack and Jill went up the hill / to fetch a pail of water, / Jack fell down and broke his crown / and Jill came tumbling after.” This little verse, according to Björn (1994?:186), describes how the moon-halo dissolves—as seen from earth of course.

The vocabulary used by Jón Árnason about the sundogs on the halo around the sun can hardly be understood as anything but evidence for a continuous tradition from much earlier times about how myths are used to describe real phenomena in the sky. If that were not the case we would have to conclude that Jón Árnason or someone else before him had come up with this interpretation of the mythology—and thus preceded both Richard Allen and Björn Jónsson in putting forth the mythological explanation of celestial phenomena. Of these two possibilities the former is much more likely, to say the least. It is not easy to come up with any alternative explanation. So, for once, it looks as if we have something in our field that amounts to a scientific proof, so that we can state with full confidence that the mythology of the Edda has an astronomical dimension that has largely gone unnoticed.

A parallel development can be observed in how the classical pagan perception of the heavens has survived through the ages. It has not yet yielded to attempts to Christianize the sky above us. We still apply the names of classical gods to the planets and the constellations. The entire system is “written” in the sky and thus memorized through naming phenomena in front of us—in a manner reminiscent of the mnemotechnics discussed by Carlo Severi in this volume. As such, Snorri cannot be wrong about the tradition, because he is a living part of it. Snorri is an authentic source as is every other informant—which is not the same as claiming that what Snorri writes is the same as another individual would express, either in his own time or 100, 200 or 300 years earlier (Gísli Sigurðsson 2014).

When it comes to the sagas themselves we have, since the dawn of saga studies, been debating how oral and historical the sagas are, as opposed to authorial works of fiction—with every possibility in between as an argued option. These oppositions are of course no longer valid, as we have realized that orality and traditional continuity is no evidence for historicity, and oral story-tellers and poets can be equally creative and inventive as any author who operates in writing. In the process of the scholarly debate we have however been entangled in a web of hypothesis about written connections between sagas, debate about hypothetical dates of hypothetical written originals, required to fill in the generally assumed pattern of one written work influencing another. Rather than to go along with this ever more complicated fabrication of lost written works it comes as a fresh start to be able to allow for orality behind the written texts and thus begin our analysis afresh from the manuscripts as they are, allowing for circulating oral stories about the same characters and events as we read about in the written works. It should of course not be downplayed or forgotten that these written works are of course structured and composed utilizing the medium through which they are presented: the Book, with a beginning, middle, and end, and lots of intertwined stories woven in, in between.

Even though we acknowledge the general role of writing as an artform in the structuring of the sagas we can see that the written sagas are still operating with the esthetic principles of oral story-telling. Characters can be abruptly brought into the text, with their traditional background from outside that particular written text assumed; well known events can be referred to on the assumption that the audience is familiar with them, even though nothing is told of them in that particular saga; and so on (Sigurðsson 2004:123–250).

We can even see this oral allowance, if that is what we can call the possibility in an oral performance of referring to information outside the text, being used to its full potential in the highly literary construction of Njáls saga—to finally single out an actual saga (cf. Sigurðsson 2010) on the general advice to take the most simple example and then evaluate its complexity.

Njáls saga is preserved in numerous manuscripts, the oldest dated to around 1300, and it is generally assumed that the saga was first written shortly before that time. The saga is set in the Southern quarter of Iceland in the late tenth and early eleventh centuries, that is, in the period when the old paganism was being replaced by Christianity (which was officially accepted as the religion of the country at the legal assembly, the Alþingi, in the year 1000). Njáls saga is often claimed to be the very best saga, carefully constructed with every minor detail playing a role in its complicated web of meaning. The main characters are the friends Gunnar of Hlíðarendi in Fljótshlíð, physically strong but not very intellectual, and Njáll of Bergþórshvoll, a man of law and wisdom. Their wives, Hallgerður and Bergþóra, get into a trivial conflict over seating at a feast; the conflict escalates until they start having each other’s workmen (húskarlar) killed. Gunnar and Njáll fail to make peace between them and as a result the marriage of Gunnar and Hallgerður suffers—she even steals some cheese and thereby arouses Gunnar’s anger. Later Gunnar does not follow Njáll’s advice, is found guilty of murder, and sentenced to exile abroad. He does not leave the country, however, and his enemies gather their forces and eventually kill him at home—after a renowned defence. Skarphéðinn, son of Njáll, takes revenge but spares a certain Mörður Valgarðsson who had played an active role in the killing of Gunnar. The sons of Njáll get into a feud with Þráinn Sigfússon, Gunnar’s cousin and the son-in-law of Hallgerður. Skarphéðinn kills Þráinn and Njáll then offers to foster his son, Höskuldur. Höskuldur grows up as one of Njáll’s sons and Njáll arranges to set up him up as a chieftain, goði, while his sons receive no social status. Mörður plants lies that encourage hatred and envy between the sons of Njáll and Höskuldur, resulting in an undeserved killing of Höskuldur, much deplored by Njáll. Höskuldur’s widow, Hildigunnur, asks her cousin Flosi to take revenge. He gathers a force and burns down the farm at Bergþórshvoll, with Njáll and his entire family inside. Only one male escapes, Njáll’s son-in-law, Kári Sölmundarson from the Hebrides. After legal dealings at the Alþingi, in which Mörður Valgarðsson plays an active part, Kári takes up revenge and kills one burning-participant after another, a mission that even takes him out of the country. The revenge culminates when some of the “burners” fall in the Battle of Clontarf near Dublin in 1014. Finally Flosi and Kári go on a pilgrimage to Rome and eventually make peace between them when Hildigunnur marries Kári.

For many the idea of an oral background for this saga (as demonstrated by Lars Lönnroth [1976]) can still be waved away as a reasonable general theory but of no importance for a modern reading. It has long been appreciated that Njáls saga is an exceptionally well composed saga from a literary point of view, so well that many have agreed with Bååth (1885:159) that the author had such good command over his material that he wrote “den första linjen […] med blicken fäst på den sista” (‘the first line with his eyes fixed on the last’). Many have also seconded Lars Lönnroth’s observation that Njáls saga differs from other sagas in not providing a thematic prelude about the ancestors—as do Egils saga and others, in a well known medieval fashion.

If however we read the oral traditional background into the first two sentences in Njáls saga (“Mörður hét maður er kallaður var gígja. Hann var sonur Sighvats hins rauða.” ‘There was a man named Mord, whose nickname was gígja [a violin-like stringed instrument]. He was the son of Sighvat the Red’) we realize that previous emphasis on written textual links and failure to take the oral traditional background of Njáls saga into account has led scholars to overlook that this is one of the most thematic openings of any prose narrative, thematic in the sense that it evokes all the major themes of the saga through what Foley (1991, 6-8) called “traditional referentiality.” The opening includes a pointed reference to the oral traditional background, as we can deduce from an unrelated written source, The Book of Settlements, which provides the knowledgeable listener (as the written Book of Settlements would not have been widely available to the audience of Njáls saga for private reading) with the key themes of the saga told about the forefathers and mothers of the main hero of Njáls saga, Gunnar of Hlíðarendi: conflict of húskarlar (’male servants’) over seating that leads to a killing near a ferry of the son of Sighvat the red (rather than of Sighvat himself, as Njáls saga later indicates in passing without elaborating further), a lamenting woman craving for revenge by fire, and reconciliation (involving Mörður gígja, the first person mentioned in the Njals saga) by marriage of members of the feuding parties (the parents of Gunnar of Hlíðarendi). [5] By a pointed reference that immediately evokes what must have been a well known story about a serial killing in the family of Sighvat the red, a version of which has only been preserved for us to read in The Book of Settlements, the audience of Njáls saga is immediately tuned into all the major themes of the magnificent saga that is about to begin: a trivial matter of honor and seating (in a ferry—rather than at a feast as in the saga — involving húskarlar in both works) that leads to the killing of a significant character; outlawry (from Fljótshlíð in both stories); a woman goading male members of her family to take revenge; houseburning; and reconciliation (involving Mörður gígja rather than his namesake Valgarðsson in the saga) by marriage of members of the feuding parties (the parents of Gunnar of Hlíðarendi in The Book of Settlements rather than the parents of Flosi (i.e. son of Kári and Hildigunnur), father of Kolbeinn, the very last characters mentioned and introduced in Njáls saga).

The opening of Njáls saga can thus be read as the first of its many forebodings, preceding by several lines the more famous one about Hallgerður’s thief’s eyes when she is first introduced as a child—as is later developed when she steels the cheese as a married woman. [6] The foreboding effect of the allusion in the first line stretches all the way to the last line of the saga about the ultimate reconciliation with the marriage of Kári and Hildigunnur, whose grandson is Kolbeinn “er ágætastur maður hefir verið einnhver í þeirri ætt. Og lúkum vér þar Brennu-Njáls sögu” (‘who was the most distinguished man in that line. And here I end the saga of the burning of Njal’). To tie all this perfectly together and make sure that no one has missed the parallel with Gunnar and his parents—whose background story was evoked in the first line of the saga — it may be observed that the only other instance of the word ágætastur in Njáls saga is when Gunnar is advised towards the end of Chapter 70 to take precautions because of all his envious enemies “þar er þú þykir nú ágætastur maður um allt land” (‘since you are thought to be the most outstanding man in the whole land’)—thus grouping these two “products” of the reconciliations, Gunnar and Kolbeinn, together with the same attributive word. Even though many have agreed in general with the previously quoted conclusion by Bååth (1885:159) about the author of Njáls saga having had such good command over his material that he wrote “the first line with his eyes fixed on the last,” I do not think that anyone has taken it quite as literally as this pattern seems to suggest.

It does not do justice to the marvelous opening of Njáls saga, which brings to mind the entire “prelude recounting the settlement in Iceland of an ancestral clan” (Andersson 2006:183) that scholars have found missing in the written Njáls saga, to think of it as evidence of the author of Njáls saga having borrowed these themes and names from a lost version of The Book of Settlements in order to get some ideas for his fictive saga—as Baldur Hafstað has suggested (2001). Baldur is an adherent of the literary reading of the sagas and was the first to comment on these paralells. Rather, the opening should be read in view of the oral background, as a reference to the tradition from which Njáls saga is derived and with which its audience was familiar — to a greater or lesser degree, of course, depending on the individuals. This reference is a fine example of the poetics of oral tradition, carried on into the written world of Njáls saga, poetics that will remain obscure and unappreciated as long as scholars continue to read the sagas with only the methods of contemporary literary analysis and without fully incorporating the notion of the sagas’ oral traditional background.

To conclude: more than half a century after its publication, The Singer of Tales can still be regarded as the ultimate inspiration to new approaches and fresh readings of the medieval texts that take up most of our time in Old Norse Studies. The insights into oral poetry, storytelling, performances, reception, textualization, and cultural contexts that it has evoked are still far from being fully explored. The examples I have taken from the Eddic poems, Snorri’s Edda and Njáls saga should be sufficient to demonstrate that the oral approach leads to no trivial findings but generates results and ideas that redefine the field as we know it, if they are taken up and used by more scholars than just those of us who have added the oral aspect to our eye-sight.