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

  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.

2. Óláfr Þórðarson Hvítaskáld and the Oral Poetic Tradition in the West of Iceland c. 1250: The evidence of the verse citations in The Third Grammatical Treatise

Collections, Anthologies, and the Literary Corpus

As pointed out in the Introduction (p. 6–17), there is good reason to reject the often expressed idea that Snorri Sturluson had to ‘construct’ the passages of prose narrative in his Edda out of the diverse assortment of skaldic verses that had come down to him through tradition. It seems altogether more probable that he would have known the stories anyway, presumably from having heard them as part of an oral story tradition (which would not always have been fully consistent with the poems as we know them). The part of oral tradition in the writing of the sagas has long been the subject of scholarly debate, but little thought has hitherto been given to the extent of the corpus of skaldic verse preserved in oral tradition and how much people might generally have known of the poetry of particular skalds. Which skalds did people need to know to be au fait with the art of poetry, as it were? Nowadays we can turn to anthologies as a guide to which authors are considered ‘important.’ An Icelandic student, for instance, who has worked through the selection in Sigurður Nordal’s Íslenzk lestrarbók (‘An Icelandic reader’) (1924), will be reasonably well up on the main sagas, poems, and authors that people need to be familiar with for the purposes of general knowledge quizzes and the like. The widely used teaching book, E. V. Gordon’s An Introduction to Old Norse (1927), fulfills a similar function for students outside Iceland taking their first steps in Old Icelandic literature.

It is possible to squeeze out indirect evidence about the extent of the common knowledge of poetry in the 13th century by looking at the nephew of Snorri Sturluson and elder brother of the saga writer Sturla Þórðarson, Óláfr Þórðarson hvítaskáld (‘White-Skald’) (1210/12?-1259). Óláfr is known to us as a poet, scholar, and teacher. He was a consecrated subdeacon and served twice as lawspeaker, 1248-50 and 1253. He traveled abroad with Snorri in 1237 and spent time in Norway, Sweden, and Denmark. Later he probably founded a school at Stafaholt in Borgarfjörður in western Iceland. He was a highly educated man, fully conversant with the language of the skalds, and thus provides an excellent test case for investigating the ideas and interests of literary enthusiasts of his times. The most important work that can be ascribed to him with certainty is his Málskrúðsfrœði (‘Ars Rhetorica’), often known as The Third Grammatical Treatise, preserved in the Codex Wormianus manuscript of Snorri’s Edda. [1] In this work Óláfr quotes from a large number of Icelandic poems and poets, and we can suppose that the poets Óláfr quotes are the ones that would have sprung most naturally to the mind of an educated member of the literary circle in western Iceland in the middle years of the 13th century. Óláfr’s examples thus make it possible to draw up an image of the heritage of poems and poets that a lover of poetry in the 13th century might refer to with familiarity, and expect his readers to be familiar with through their first names alone. This allows us to ask certain questions: Did Óláfr’s perspective cover the entire country? Did he know verses by his counterparts in the east? What poets did he know from the past? We can suppose that Óláfr himself knew considerably more than he happens to quote in his treatise, but even so his choice of examples provides an indication of what he felt he might expect of his readers, of what he could be confident was part of their common knowledge. [2]

Óláfr’s life provides a striking example of the family conflicts and party politics surrounding the chieftains of the Sturlung Age. He himself is mentioned frequently in the most important record of his times, Sturlunga saga. His first known home was at Bjarnarhöfn, west of modern day Stykkishólmur on the Snæfellsnes peninsula. As a teenager in 1226 we hear of him at a midwinter feast at Snorri Sturluson’s manor at Reykjaholt—in a winter of volcanic eruptions and sand drifts in which Snorri lost a hundred head of cattle at Svignaskarð. In summer 1227, at the family’s ancestral home of Hvammur, Óláfr acts as peacemaker between his father Þórðr and his irascible cousin Sturla Sighvatsson. From 1234 we hear of him taking an active part (at times with his brother Sturla, the saga writer) in clashes with another cousin, Órœkja, son of Snorri Sturluson, and in 1236, on Snorri’s advice, he moves from Hvammur to Borg in the Mýrar district (the home of the poet Egill Skallagrímsson in the 10th century and closer to Snorri himself). In 1237, on the death of his father on April 10 at the age of 82, Óláfr comes into his inheritance and that winter composes a drápa (a formal poem of praise in dróttkvætt meter) to Bishop (Saint) Þorlákr; nothing of this poem survives, but the next summer saw the institution of the most important festival in the Icelandic Church calendar, Þorláksmessa, June 20.

Two years later, in the winter after the Battle of Örlygsstaðir, 1238-9, Snorri, Órœkja and Þorleifr are with Earl Skúli at Niðarós and Þórðr kakali with King Hákon at Bergen. The next spring Snorri’s party wishes to depart for Iceland but the king refuses them permission to leave. Óláfr hvítaskáld is also said to have been with the earl at this time, together with the Norwegian Arnfinnr Þjófsson, who is cited as the source for the news that Skúli has made Snorri an earl (probably of the island of Fólskn (modern Norwegian: Storfosna) off Niðarós: see Pálsson, H. 1992:165-8). A little later, when Earl Skúli sails off to face Hákon, Óláfr composes a poem about his ship Langafrjádagr (‘Good Friday’). The Battle of Láka (modern Norwegian: Låka) took place on March 6, 1240; the king suffered heavy losses and Earl Skúli moved south to Oslo and had himself proclaimed king. Óláfr Þórðarson is next heard of again at Niðarós, but now with the king, composing a poem about the battle in which he announces it his wish that the king be restored to his rightful position. This change of allegiance on Óláfr’s part is puzzling; Björn M. Ólsen (1884:xxxiv) suggested that Óláfr had gone to Sweden in 1239 to perform a poem in honor of King Eiríkr and, on his return to Niðarós, found Earl Skúli already gone and King Hákon newly arrived (Skúli left on 18 February 1240 and Hákon arrived shortly afterwards), leaving Óláfr little option but to go over to the king’s side.

After his return to Iceland nothing is heard of Óláfr until 1248, when Þórðr kakali has him appointed lawspeaker. Björn M. Ólsen came to the conclusion that Óláfr was not in Iceland by 1242, since he is not associated with the arrest of Sturla Þórðarson and Órœkja Snorrason at the Hvítá bridge in that year, but that he must have returned with an account of his journeys before 1245, the year when Aron and Brynjólfr are boasting to King Hákon about his reception at the court of King Valdimarr.

Back in Iceland, aged just over 30, Óláfr found conditions much changed from when he had left over five years earlier: Sturla Sighvatsson, who had been all powerful at the time of Óláfr’s departure for Norway, had died with his father in 1238 at the Battle of Örlygsstaðir; Snorri had been killed by agents of King Hákon in 1241; and Órœkja Snorrason and Sturla, Óláfr’s brother, were now a spent force. Óláfr relinquished his position as lawspeaker to his brother Sturla in 1251 but was reinstated the next year. The same year he received Þorgils skarði at Stafaholt when Þorgils arrived from Norway bearing Hákon’s demand for a share in Snorri’s estate. Óláfr supported the king’s claim at a meeting at Höfðahólar late in August, thereby setting himself in opposition to his brother Sturla and second cousin Þorleifr Þórðarson of Garðar. In December the same year, Þorgils turned up at Stafaholt with a large band of supporters and in considerable fear. The same evening, Sturla Þórðarson and Hrafn Oddsson sprang an attack and captured Þorgils, who prepared himself for death. Óláfr was incensed, declaring the attack a sacrilege and saying that God and St. Nicholas, to whom the church was dedicated, would have revenge. Subsequently Óláfr persuaded Þorgils to agree to break faith with Gizurr Þorvaldsson (on the grounds that they all had scores to settle with him following Snorri’s killing) and to accept terms from Sturla and Hrafn. That same night Þorgils went back on the oath coerced from him and rode north.

Table 2-1: Chronology of the Life of Óláfr Þórðarson hvítaskáld (based on Sturlunga saga, Hákonar saga, and Knýtlinga saga)

1226 present at a midwinter feast at his uncle Snorri’s at Reykjaholt in the west.
1227 (summer) promotes peace at Hvammur between his father Þórðr and his cousin Sturla Sighvatsson when Sturla arrives in militant mood.
1234 start of ongoing struggles between Óláfr (and his brother Sturla) and Órœkja, son of Snorri Sturluson.
1236 on Snorri’s advice, moves from his father’s estate at Hvammur south to Borg in Mýrar.
1237 (spring) inherits from his father and composes a drápa about Bishop/Saint Þorlákr. Present at the Battle of Bær. During the summer, he flees to Norway with his uncle Snorri, and at the same time as Þórðr kakali, to escape the fury of Sturla Sighvatsson.
1239 with Earl Skúli when Snorri declares his intention to return to Iceland. Composes a poem about Skúli’s ship.
1240 after the Battle of Láka (modern Norwegian: Låka), Óláfr is with King Hákon. He has probably spent some of the interim in Sweden and composed poetry to King Eiríkr.
?1240–1 with King Valdimarr I of Denmark, held in high esteem.
1248 Þórðr kakali has Óláfr appointed lawspeaker. Indirect evidence suggests he returned to Iceland between 1242 and 1245 as he is not present when Sturla and Órœkja are arrested at the Hvítá bridge in 1242 but by 1245 has already told his kinsmen about his time in Denmark.
1250 Sturla Þórðarson replaces Óláfr as lawspeaker.
1253 Óláfr appointed lawspeaker for a second term but resigns the same year on the grounds of ill health.

Óláfr is often believed to be the author of Knýtlinga saga, which traces the history of the kings of Denmark on the model of Snorri’s Heimskringla (see Guðnason 1982:clxxix-clxxxiv and references there). Knýtlinga saga refers to Óláfr’s stay at the court of Valdimarr Valdimarsson (d. 1241) ‘er einhverr hefir verit ágætastr konungr hingat á Norðrlǫnd. Með honum var Óláfr Þórðarson ok nam at honum marga frœði, ok hafði hann margar ágætligar frásagnir frá honum’ (‘who was one of the finest kings yet seen in Scandinavia. With him was Óláfr Þórðarson, who obtained a great deal of information from him and had many excellent stories from him’) (Knýtlinga saga, ch. 127). Óláfr himself refers to Valdimarr in The Third Grammatical Treatise in connection with the runic characters (Ólsen 1884:45): ‘Þessa stafi ok þeirra merkingar compileraði minn herra Valldimarr konungr með skjótu orðtæki á þessa lund’ (‘My lord King Valdimarr compiled these letters and their meanings in brief as follows’). Óláfr acquired some knowledge of Danish and German, probably while in Denmark, as he refers to these languages in the Treatise. Laxdœla saga has also been ascribed to Óláfr (Hallberg 1963), though the arguments in support of this have been disputed (Allén 1971, Thorsson 1994).

Óláfr’s poetic remains are meager: we have three single occasional verses; one verse from a poem on King Hákon the Old; twelve verses in hrynhenda meter from a poem about King Hákon and Earl Skúli; two verses from a drápa about Aron Hjǫrleifsson; and two fragments from a poem on Thomas à Becket (the English saint, 1117-70, murdered in Canterbury Cathedral). Nothing survives of the poems he is known to have composed on Earl Knútr Hákonarson, King Eiríkr of Sweden, King Valdimarr the Old of Denmark, and Saint Þorlákr. Thus, in addition to his clerical learning, Óláfr was trained in poetics and a writer of prose. He may well have gotten some of his education from his uncle Snorri Sturluson and thus been ideally placed to absorb the poetic tradition cultivated by the educated and literate elite of the Dalir and Borgarfjörður regions of western Iceland in the 13th century.

Scholarly Neglect of Óláfr’s Poetic Examples

Óláfr probably wrote his Third Grammatical Treatise shortly after his return to Iceland from his time abroad. In support of this, Björn M. Ólsen pointed to the warmth with which Óláfr refers to King Valdimarr and a certain animosity toward King Hákon that appears in the glosses to one of the verses. The treatise is divided into two sections, on grammar and rhetoric, and is based largely on the works of Priscian and Donatus. Óláfr’s task was thus to find or compose examples of verse in Icelandic to exemplify the main concepts of Latin poetic style that he introduces. These examples demonstrate the independence of the domestic poetic tradition vis-à-vis the Latin learning that the work attempts to expound to its readers (see Tranter 2000). The treatise also reveals that Óláfr had a thorough knowledge of Norse runes, since one of his purposes is to show his readers that the Roman alphabet is better suited to the sound system of Icelandic than runes—which he takes for granted that his readers will also be familiar with.

Scholarly attention to the Treatise has been directed first and foremost at Óláfr’s command of the Latin tradition of learning, and in particular at his sources and how he used and interpreted them (Ólsen 1884; Collings 1967; Raschellà 1983; Albano 1985-6; Tómasson 1992:529-32). The section on grammar—‘Málfrœðinnar grundvǫllr’ (‘Basics of grammar’), which follows Priscian—has been studied by Micillo (1993); Kristján Árnason (1993) has discussed pitch in Old Icelandic in light of Óláfr’s evidence; Björn M. Ólsen (1884) and Raschellà (1994) have looked into the background of the chapter on runes; and Tómasson (1993) has analyzed the prefaces to the four grammatical treatises in the Codex Wormianus.

The examples of skaldic poetry cited by Óláfr have, however, received less scholarly attention, [6] except perhaps one verse that he ascribes to Kormakr and another ascribed to Bjǫrn Hítdœlakappi in the saga that bears his name (see p. 102, note 9). A certain amount of scholarly fun has been had with Óláfr’s example of the trope ofljóst, [7] and there is a well-known verse found only in the Treatise that is generally given as the final stanza of Egill Skallagrímsson’s lament for his friend Arinbjǫrn, Arinbjarnarkviða (‘Vas’k árvakr’). [8] It seems fair to assume, however, that Óláfr’s choice of examples reveals his knowledge of and taste in the poetry that was known to him and his contemporaries, and which he also expected of his readers. As Finnur Jónsson (1923:924) observed, without further comment, the poetry cited comes from all ages of Icelandic history, by both Icelandic and Norwegian skalds. The examples serve as evidence of systematic thinking on the art of poetry at the oral stage, prior to the arrival of Latin learning (Raschellà 1983:293, 298), but their potential source value has nevertheless escaped scholars, as typified by the following comment from a recent literary history (Tómasson 1992:531), reflecting the conventional view: ‘Einstaka vísur eru og hvergi til annars staðar’ (‘A few of the verses are not found elsewhere’). As is made clear below, this is simply not true: a simple count of the lines reveals that in fact over two thirds of Óláfr’s examples are not found in other sources, going by the comparative evidence cited in the footnotes in Finnur Jónsson’s 1927 edition of the Treatise and in his standard edition of the extant skaldic corpus Skjaldedigtningen IA (esp. pp. 590-602; see Jónsson, F. 1912-15).

Despite the lack of interest among scholars, Óláfr’s indigenous verse examples merit special investigation. As stated previously, one may suppose that, when Óláfr cites a stanza or verse fragment by a particular poet without further specification, he is taking it for granted that it will be reasonably familiar to his audience or readership. This assumes, of course, that Óláfr was writing for people who were interested in this poetry and had a fair knowledge of it, up to the level, say, one might expect of beginners in formal book-learning; but in this regard it may be said that the whole treatise bears the marks of having been written for readers already possessed of a sound knowledge of both runes and skaldic verse. By looking into Óláfr’s examples and taking account of their provenance, we may therefore be able to draw up some kind of picture of the condition and strength of the skaldic tradition in the mid-13th century.

From this we should be able to see which poems and poets were so familiar to him that he turned to them automatically when looking for Icelandic examples to illustrate his Latin stylistic concepts—in the confident belief that the poets he refers to with familiarity would also be familiar to his potential audience.

The Provenance of Óláfr’s Citations

In the Treatise, Óláfr names 34 poets and presents 123 examples of poetry, 354 lines in all. Most of the examples are only fragments of stanzas, but seven stanzas are quoted in full, of which one is repeated.

Figure 2-1: The verse examples in The Third Grammatical Treatise

The question that now arises is: Where did Óláfr get these examples from? There are at least four possibilities:

  1. from written sources
  2. from oral tradition
  3. Óláfr composed or reworded them himself
  4. translated or adapted from the Latin exemplar

Examples known from other written sources

If Óláfr composed or translated some of his examples himself, it seems most likely that they would be among the 51 examples that are both a) unknown from other sources, and b) presented without a named author. Since there is no way of getting any further information on the origins of the examples in this group there is very little we can say about them.

With the other examples things are rather different, and I will start by considering the forms used to introduce the 33 verses that are known from other written sources.

Table 2-2: Introductory formulas to verses cited in The Third Grammatical Treatise that are also known from other sources

sem Snorri kvaða (‘as Snorri said’) 2 lines (from Háttatal 83)
sem hér (‘as here’) 1 line (from a verse by Þormóðr Kolbrúnarskáld, in Fóstbrœðra saga but not in Heimskringla)
sem Egill kvað (‘as Egill said’) 8 lines (Arinbjarnarkviða 16)
sem kvað Hárekr í Þjóttu (‘as Hárekr of Þjótta said’) 2 lines (in Óláfs saga helga, which names the poet Hárekr Eyvindarson)
sem Glúmr kvað (‘as Glúmr said’) 4 lines (Glúmr Geirason, from Gráfeldardrápa, in Heimskringla)
sem hér (‘as here’) 2 lines (ascribed to Bjǫrn Hítdœlakappi in Bjarnar saga Hítdœlakappa)  
sem hér (‘as here’) 2 lines (by Einarr skálaglamm, from Vellekla, in Heimskringla)
sem Eyvindr kvað (‘as Eyvindr said’) 2 lines (Eyvindr skáldaspillir Finnsson, in Heimskringla)
sem Arnórr kvað (‘as Arnórr said’) 2 lines (Arnórr jarlaskáld Þórðarson, from Hrynhenda, to Magnús, king of Norway, in Morkinskinna)
sem kvað Halldórr skvaldri (‘as Halldórr skvaldri said’) 4 lines (? from Útfarardrápa)
sem Sighvatr kvað (‘as Sighvatr said’) 1 line (Sighvatr Þórðarson, from Nesjavísur, in Fagrskinna)
sem Snorri kvað (‘as Snorri said’) 2 lines (from Háttatal 28)
sem Bjǫrn kvað (‘as Bjǫrn said’) 2 lines (in Kormaks saga, where it is attributed to Hólmgǫngu-Bersi)
sem Sighvatr kvað (‘as Sighvatr said’) 1 line (Sighvatr Þórðarson, in Heimskringla)
sem Snorri kvað (‘as Snorri said’) 3 lines (from Háttatal 15-16)  
sem hér (‘as here’) 4 lines (Snorri Sturluson, from Háttatal 40)
sem Snorri kvað (‘as Snorri said’) 4 lines (from Háttatal 73)
sem Hallfreðr kvað (‘as Hallfreðr said’) 2 lines (Hallfreðr vandræðaskáld Óttarsson, from Heimskringla, not in Hallfreðar saga)
sem kveðit er í Grímnismálum (‘as is said in Grímnismál’) 2 lines (from Grímnismál 47)
sem Máni kvað (‘as Máni said’) 4 lines (in Skáldskaparmál in Snorri’s Edda)
sem Eyvindr kvað (‘as Eyvindr said’) 4 lines (Eyvindr skáldaspillir Finnsson, from Háleygjatal, in Heimskringla)
sem Markús kvað (‘as Markús said’) 2 lines (Markús Skeggjason, in Snorri’s Edda)
sem Ormr Steinþórsson kvað (‘as Ormr Steinþórsson said’) 2 lines (in Snorri’s Edda)
sem Snorri kvað (‘as Snorri said’) 2 lines (from Háttatal 5)
enn sem Snorri kvað (‘again as Snorri said’) 2 lines (from Háttatal 5)
sem hér (‘as here’) 4 lines (by Þórðr Kolbeinsson, in Heimskringla)
sem hér er kveðit (‘as is said here’) 4 lines (also in Snorri’s Edda)
sem hér (‘as here’) 1 line (by Earl Gilli, in Njáls saga, about Brian Boru and the Battle of Clontarf, 1014)
sem Sighvatr kvað (‘as Sighvatr said’) 4 lines (Sighvatr Þórðarson, from Bersǫglisvísur, in Heimskringla)
sem Einarr kvað (‘as Einarr said’) 4 lines (Einarr Skúlason, from Geisli 1)
sem hér (‘as here’) 3 lines (from Gátur Gestumblinda)
sem Egill kvað (‘as Egill said’) 4 lines (also in Egils saga)
sem Snorri kvað (‘as Snorri said’) 8 lines (in Hákonar saga)

a. In general Old Norse usage, the verb kveða means simply ‘say’. However, it is the etymological origin of several words referring to poems and poetry (kvæði, kviða, kveðskapr) and therefore has a special sense of ‘compose (a poem)’, without distinction between oral and written poetry.

Table 2-3: Works containing stanzas also quoted in The Third Grammatical Treatise

Written texts Poems preserved as independent entities (written and/or oral) Sources uncertain
Snorri’s Edda
*Older version of Morkinskina
Egils saga
Snorri Sturluson’s verses in Hákonar saga
*Brjáns saga
Gátur Gestumblinda
Bjarnar saga Hítdœlakappa (one verse unascribed by Óláfr but attributed to Bjǫrn in the saga)
Fóstbrœðra saga (one verse unascribed by Óláfr but attributed to Þormóðr Kolbrúnarskáld in the saga)
Kormáks saga (one verse attributed by Óláfr to Bjǫrn but to Hólmgǫngu-Bersi in the saga)

The only well-known poet from the sagas of Icelanders that Óláfr refers to by name is Egill Skallagrímsson. This fact might lend support to the frequently expressed view that Egils saga was the first, or one of the first, of the sagas of Icelanders to be written (see Hafstað 1990, Kristjánsson 1990, Ólason, V. 1991). Three of Óláfr’s examples are also found in other sagas of Icelanders; of these, two are unascribed in the Treatise (the verses found also in Bjarnar saga Hítdœlakappa and Fóstbrœðra saga), while the other is ascribed to different people by Óláfr and in Kormaks saga (by Óláfr to Bjǫrn and by the saga to Bersi, though it should be noted that these names are synonymous). The lack of attribution for the examples also found in Bjarnar saga and Fóstbrœðra saga may be explained either as a) Óláfr not having known who the authors were (which might be due to his having learned the verses from somewhere other than the written sagas), or b) his considering the verses so well known that he did not need to specify the authors (as in the examples from Vellekla, Háttatal, Heimskringla, *Brjáns saga, and Gátur Gestumblinda, each of which is quoted without ascription). The second of these possibilities would imply that Óláfr knew more skalds from the sagas of Icelanders than just Egill, whether this knowledge came from books or from oral tradition. There is no way of deciding which explanation is closer to the truth, but the examples from Kormaks saga, Fóstbrœðra saga, and Bjarnar saga Hítdœlakappa are of potential significance for attempts to determine the ages of these sagas, e.g. to question the arguments put forward by Jónas Kristjánsson (1972:292-310) and Bjarni Guðnason (1994) that the last two are rather younger than was formerly thought. Alternatively, we might use the uncertainties surrounding these verses as reason to reconsider the view that the sagas had no existence before the time they achieved written form (see Kellogg 1994 and below, p. 242).

Examples not known from other written sources

At this point it is worth considering the formulas Óláfr uses to introduce the 90 verses and fragments that are not known from other written sources:

Table 2-4: Introductory formulas to verses quoted in The Third Grammatical Treatise that are not known from other sourcesa

sem kvað Auðunn illskælda (‘as Auðunn illskælda said’) 4 lines. One other verse is preserved in a þáttr about the poets of King Haraldr hárfagri in Óláfs saga Tryggvasonar. FJ considers Auðunn to be Norwegian, from the 9th century.
sem Arnórr kvað (‘as Arnórr said’) 1 line. Arnórr jarlaskáld Þórðarson. Second half of 11th century. A large body of verse survives, mainly to earls and kings Magnús the Good and Haraldr harðráði. From Hítarnes í Hnappadalssýsla in the west of Iceland. There is a þáttr about him in Morkinskinna.
sem hér (‘as here’) 2 lines with mythological content (e.g. the kenning ‘Brynhildar bróðir,’ ‘Brynhildr’s brother’). Author not specified.
sem kvað Eilífr Guðrúnarson (‘as Eilífr Guðrúnarson said’) 4 lines. FJ attributes this fragment to Eilífr kúlnasveinn, a 12th-century Icelander and author of Kristsdrápa. Eilífr Guðrúnarson is also said to be Icelandic (though this is uncertain). Known from Snorri’s Edda. He was alive around the year 1000 and composed the mythological poem Þórsdrápa; there is also one fragment of Christian verse.
sem hér (‘as here’) 2 lines
sem Einarr kvað (‘as Einarr said’) 2 lines (the fragment about ‘Máni’s wife’: see note 7, p. 99). Possibly Einarr Skúlason, a prolific poet, fl. mid-12th century. If this is correct, he was from Borg in Mýrar, western Iceland. There is a þáttr about him in Morkinskinna. Óláfr quotes elsewhere from his famous religious poem Geisli.
sem Skraut-Oddr kvað (‘as Skraut-Oddr said’) 4 lines. Skraut-Oddr is mentioned only here (another stanza later). FJ considers him Icelandic and from the 11th century.
sem Starkaðr gamli kvað (‘as Starkaðr the Old said’) 4 lines. A famous prehistorical hero from the legendary sagas.
sem hér (‘as here’) 2 lines
sem hér (‘as here’) 1 line
sem Óláfr Leggsson kvað (‘as Óláfr Leggsson said’) 1 line (mythological material). Icelandic poet of the 13th century, nicknamed ‘svartaskáld.’ He composed to King Hákon, Earl Skúli, and Christ. From the Lundarmenn family of Borgarfjörður, western Iceland. Involved in the killing of Jón murti, son of Snorri Sturluson, in 1231.
sem hér (‘as here’) 1 line (mythological material)
sem Einarr kvað (‘as Einarr said’) 2 lines. Einarr Skúlason (see above)
sem hér er kveðit (‘as is said here’) 2 lines
ok sem þetta (‘and as this’) 2 lines
sem hér er kveðit (‘as is said here’) 2 lines
sem Þorleifr jarlsskáld kvað (‘as Þorleifr jarlsskáld said’) 4 lines. Þorleifr Rauðfeldarson jarlsskáld (‘earl’s poet’). 10th century, from Brekka in Svarfaðardalur, northern Iceland. There is a þáttr about him in Flateyjarbók. The verse is about Hákon (?Earl of Hlaðir); Heimskringla has another verse by him about Hákon. Spent his last years at Mýrdalur, southern Iceland.
sem Snorri kvað (‘as Snorri said’) 2 lines. The only known piece of religious verse by Snorri Sturluson.
sem hér (‘as here’) 2 lines
sem í Hafliðamálum (‘as in Hafliðamál’) 3 lines. FJ considers this to be from the 12th century (1121), possibly about Hafliði Másson (see chapter 1 above, passim).
sem hér er kveðit (‘as is said here’) 2 lines
sem Þjóðólfr kvað (‘as Þjóðólfr said’) 4 lines. Þjóðólfr Arnórsson. Family from Svarfaðardalur in the north. Went abroad young and worked for kings Magnús the Good and Haraldr harðráði. Died with the latter at Stamford Bridge in 1066. Much of his poetry survives. This verse was perhaps composed on the death of King Magnús in 1047. Mentioned in Heimskringla, Sneglu-Halla þáttr and Brands þáttr ǫrva.
sem hér er kveðit (‘as is said here’) 2 lines
sem hér (‘as here’) 5 lines
sem Kolbeinn kvað (‘as Kolbeinn said’) 4 lines. ?Kolbeinn Tumason of the Ásbirningar clan from the north, died 1208.
sem hér er kveðit (‘as is said here’) 4 lines
sem Arnórr kvað í Magnúsdrápu (‘as Arnórr said in Magnúsdrápa’) 2 lines. Arnórr jarlaskáld (see above). The fragment is found only here and is perhaps from the introduction to a drápa to King Magnús.
sem Guðbrandr kvað í Svǫlu (‘as Guðbrandr of Svala said’) 2 lines. FJ believes this to be the Guðbrandr mentioned in Hrafns saga (c. 1200), who composes a verse quoted there in confirmation that Hrafn has received a certain Loftr Markússon, who says he has been sent to him from Mýrar in the west by Sighvatr Sturluson.
sem Sneglu-Halli kvað (‘as Sneglu-Halli said’) 4 lines. Sneglu-Halli, fl. mid-11th century, from the north of Iceland. He spent time at the court of Haraldr harðráði. There is a (comic) þáttr about him in Morkinskinna, Hulda, and Hrokkinskinna.
sem hér er kveðit (‘as is said here’) 4 lines
ok sem hér er kveðit (‘and as is said here’) 2 lines
sem Þjóðólfr kvað (‘as Þjóðólfr said’) 2 lines. Þjóðólfr Arnórsson (see above). Heimskringla links this fragment with King Haraldr harðráði.
sem Guðlaugr kvað (‘as Guðlaugr said’) 4 lines. We know of no poet called Guðlaugr. FJ considers him Icelandic, 12th century. Several men called Guðlaugr are mentioned in Sturlunga saga.
sem hér er kveðit (‘as is said here’) 2 lines
sem hér er kveðit (‘as is said here’) 2 lines
sem Óláfr kvað (‘as Óláfr said’) 4 lines. ?Óláfr hvítaskáld himself.
sem Hallar-Steinn kvað (‘as Hallar-Steinn said’) 2 lines. Hallar-Steinn Herdísarson, c. 1200. The name may refer to Höll in Þverárhlíð in Borgarfjörður, western Iceland. Composed Rekstefja, 35 stanzas about King Óláfr Tryggvason.
sem Egill kvað (‘as Egill said’) 2 lines. Egill Skallagrímsson. The only fragment of this type by Egill. Meaning obscure.
sem hér (‘as here’) 2 lines. ?From Íslendingadrápa.
sem Bragi hinn gamli kvað (‘as Bragi the Old said’) 2 lines. Bragi gamli Boddason. Possibly 9th century, Norwegian.
sem Kormakr kvað (‘as Kormakr said’) 2 lines. Not in Kormaks saga! Kormakr was from Melur in Miðfjörður in the northwest. 10th century. His saga contains a large number of occasional verses ascribed to him, and there are fragments of his Sigurðardrápa in Snorri’s Edda and Heimskringla.
sem Arnórr kvað (‘as Arnórr said’) 2 lines. Arnórr jarlaskáld (see above). ?From the introduction to Magnúsdrápa.
sem hér (‘as here’) 1 line (previously quoted)
sem fyrr er ritat (‘as was written previously’) 1 line (previously quoted)
sem fyrr er ritat (‘as was written previously’) 1 line (previously quoted)
sem í Bjúgum vísum (‘as in Bjúgar vísur’) 1 line. FJ assigns this to the 12th century.
sem Sighvatr kvað (‘as Sighvatr said’) 2 lines. Sighvatr Þórðarson. Possibly from a memorial ode to King Óláfr Haraldsson (St. Olaf). Sighvatr was from Apavatn in Grímsnes in southern Iceland. There is a þáttr about him in Snorri’s Óláfs saga helga.
í þessum orðum (‘in these words’) 1 line (previously quoted)
sem hér er kveðit (‘as is said here’) 4 lines. Apparent Latin influence. ?By Óláfr himself.
sem hér er kveðit (‘as is said here’) 4 lines. Probable influence from the Paternoster.
sem hér (‘as here’) 4 lines. ?From a praise poem by Óláfr himself.
sem hér er kveðit (‘as is said here’) 1 line
sem Glúmr kvað (‘as Glúmr said’) 2 lines. Glúmr Geirason. Fragment from a poem on King Eiríkr Blood-axe. Icelandic poet of the 10th century. Lived at Mývatn in the north and Króksfjörður in the northwest. Composed a poem on the death of Eiríkr Blood-axe and Gráfeldardrápa about King Haraldr gráfeldr, cited in Heimskringla.
sem hér (‘as here’) 4 lines. ?From a praise poem by Óláfr himself.
sem hér (‘as here’) 4 lines
sem hér (‘as here’) 2 lines (in ms. K, not W).
sem hér er kveðit (‘as is said here’) 2 lines
sem hér (‘as here’) 2 lines
sem hér (‘as here’) 4 lines
sem hér er kveðit (‘as is said here’) 4 lines
sem hér er kveðit (‘as is said here’) 2 lines
sem Egill kvað (‘as Egill said’) 8 lines. Egill Skallagrímsson. ?From Arinbjarnarkviða. The two stanzas in the Treatise thought to come from Arinbjarnarkviða (this one and the ‘final stanza’ below) could once have stood in the Mǫðruvallabók ms. where the text is now illegible or on a lost sheet.
sem hér (‘as here’) 8 lines
sem hér (‘as here’) 4 lines. Resembles Haustlǫng by Þjóðólfr of Hvinir (see below).
sem hér (‘as here’) 4 lines
sem Skraut-Oddr kvað (‘as Skraut-Oddr said’) 2 lines. For Skraut-Oddr, see above.
sem hér (‘as here’) 1 line
sem Þjóðólfr kvað (‘as Þjóðólfr said’) 4 lines. FJ ascribes this to Þjóðólfr of Hvinir (10th-century Norwegian) from a perceived likeness to his Ynglingatal. Óláfr does not distinguish this Þjóðólfr from Þjóðólfr Arnórsson, from whom he has already quoted (see above), so it seems reasonable to assume that it is the latter Þjóðólfr that is intended here.
sem Markús kvað (‘as Markús said’) 2 lines. Probably Markús Skeggjason, lawspeaker from 1084 to his death in 1107. From the south of Iceland. Composed a drápa in hrynhenda meter to Eiríkr the Good Sveinsson, king of Denmark, which he sent by a representative to the king and fragments of which are preserved in Knýtlinga saga.
sem hér (‘as here’) lacuna in the manuscript
sem Þjóðólfr kvað (‘as Þjóðólfr said’) 4 lines. Presumably Þjóðólfr Arnórsson (see above).
sem hér (‘as here’) 2 lines
sem hér (‘as here’) 2 lines
sem hér (‘as here’) 2 lines
sem hér (‘as here’) 4 lines
sem hér (‘as here’) 2 lines
sem hér (‘as here’) 2 lines. Reminiscent of Egill’s Arinbjarnarkviða.
sem Leiðólfr kvað (‘as Leiðólfr said’) 2 lines. The name is rare. A certain Leiðólfr is mentioned in Njáls saga. This is the only piece of verse ascribed to anyone of this name, and nothing is known about him.
sem hér er kveðit (‘as is said here’) 2 lines
sem hér er kveðit (‘as is said here’) 4 lines
sem Egill kvað (‘as Egill said’) 8 lines. Egill Skallagrímsson. ?Final stanza of Arinbjarnarkviða.
sem hér (‘as here’) 2 lines
sem hér (‘as here’) 2 lines
sem Sveinn kvað (‘as Sveinn said’) 4 lines. The fragment resembles a verse by Úlfr Uggason (late 10th century, Icelandic). FJ identifies this Sveinn with a Sveinn who composed Norðrsetudrápa, quoted in Snorri’s Edda and assigned by FJ to the 11th century.
sem í Kúgadrápu (‘as in Kúgadrápa’) 2 lines. A certain Kúgi appears in Orkneyinga saga.
sem Sveinn kvað (‘as Sveinn said’) 1 line. For Sveinn, see above.
sem hér er kveðit (‘as is said here’) 4 lines
sem Egill kvað (‘as Egill said’) 8 lines. Egill Skallagrímsson. This stanza has already been quoted and is thought to be from Arinbjarnarkviða.
sem Nikulás ábóti kvað (‘as Abbot Nikulás said’) 8 lines. Probably the same man as Nikulás Bergsson, abbot of Munkaþverá, died 1159, to whom is ascribed Leiðarvísir, an itinerary for pilgrims to Rome and the Holy Land, and who composed a drápa to the Apostle John. The stanza here is thought to be about Christ.
sem hér er kveðit (‘as is said here’) 4 lines

a. FJ refers to the standard edition of the skaldic corpus, Finnur Jónsson Den Norsk-Islandske Skjaldedigtning (1912-5).

In the introductory formulas there are references to three otherwise unknown poems, but in each case no author is given:

Table 2-5: Poems named in The Third Grammatical Treatise that are not known from other sources

Hafliðamál (which FJ thinks could be about Hafliði Másson) Bjúgar vísur Kúgadrápa (A character called Kúgi appears in Orkneyinga saga.)

The poets named in the introductory formulas to the verse examples are as follows (in the order they appear in the Treatise):

Table 2-6: Authors of verses quoted in The Third Grammatical Treatise but not known from other sources, listed in the order in which they appear in the Treatise

Auðunn illskælda
Arnórr jarlaskáld
Eilífr Guðrúnarson
Einarr (Skúlason)
Starkaðr gamli
Óláfr Leggsson
Einarr (Skúlason) (2nd appearance)
Þorleifr jarlsskáld
Snorri Sturluson
Þjóðólfr (Arnórsson)
Kolbeinn (Tumason)
Arnórr jarlaskáld (2nd appearance)
Guðbrandr of Svala
Þjóðólfr Arnórsson (2nd appearance)
Óláfr (?hvítaskáld)
Hallar-Steinn (Herdísarson)
Egill (Skallagrímsson)
Bragi inn gamli
Arnórr jarlaskáld (3rd appearance)
Sighvatr Þórðarson
Glúmr (Geirason)
Egill (Skallagrímsson) (2nd appearance)
Skraut-Oddr (2nd appearance)
Þjóðólfr (Arnórsson) (3rd appearance)
Markús (Skeggjason)
Þjóðólfr (Arnórsson) (4th appearance)
Egill (Skallagrímsson) (3rd appearance)
Sveinn (2nd appearance)
Egill (Skallagrímsson) (4th appearance)
Abbot Nikulás

The verses Óláfr attributes to the twenty-six poets in the table above are not known from other written sources. There is thus a fair probability that he learned their poems by some way other than working his way through them on vellum. This list might therefore include poets who were sufficiently prominent in Óláfr’s mind that their works found their way into his collection of examples from oral knowledge. They would equally be poets who he might have assumed would be familiar to people with an interest in poetry in the west of Iceland at the time he was writing his Treatise (though this of course does not mean that all the actual examples would have been familiar to everyone). By trying to work out where these poets came from, we can perhaps build up a map of the literary horizons of Óláfr Þórðarson as he worked in his home at Stafaholt. As can been seen from the table of introductory formulas, it is not in all cases clear which poet Óláfr is referring to when he gives only a single name; however, with the aid of Finnur Jónsson’s Skjaldedigtningen (1912-5), and with an eye to the kinds of the formulas used to introduce the verses that are known from other sources, we can in most cases make informed guesses as to the identity of the particular poets.

The twenty-six poets named by Óláfr can be divided into two groups according to how much we know about them. The first group comprises nine poets whose origins are uncertain though some of them do appear elsewhere in written sources:

Table 2-7: Poets of uncertain origin, cited as authors of otherwise unknown verses in The Third Grammatical Treatise

  • Auðunn illskælda (mentioned in Óláfs saga Tryggvasonar as a poet of King Haraldr hárfagri)
  • Eilífr Guðrúnarson (fl. around 1000; known from Snorri’s Edda)
  • Skraut-Oddr (mentioned only in the Treatise; two fragments)
  • Starkaðr the Old
  • Guðbrandr of Svala (possibly from the circle around Sighvatr Sturluson and Hrafn Sveinbjarnarson around 1200)
  • Guðlaugr (?Icelandic, 12th century)
  • Bragi the Old
  • Leiðólfr
  • Sveinn (possibly author of Norðrsetudrápa, cited in Snorri’s Edda, ?11th century)

These nine skalds of unknown geographical origins fall into three groups according to where our knowledge of them (such as it is) comes from:

  1. Generally known prehistorical and (semi)legendary characters: Starkaðr the Old and Bragi the Old. Both are among the best-known figures in Scandinavian prehistory and would hardly have required further explanation.
  2. Poets known from Snorri’s scriptorium in the writing of Heimskringla and the Edda: Auðunn illskælda, Eilífr Guðrúnarson, and Sveinn.
  3. Poets known only from Óláfr: Guðbrandr (unless he is the same man as appears in Hrafns saga), Skraut-Oddr, Guðlaugr, and Leiðólfr.

About the following poets more is known (with a certain amount of guesswork in cases where a name is shared by more than one known skald):

Table 2-8: Authors of verses in The Third Grammatical Treatise that are not found in other sources, about whom we have reliable historical information

  • Arnórr jarlaskáld (‘earls’ poet’) Þórðarson from Hítarnes in Hnappadalssýsla, western Iceland. Second half of 11th century. There is a þáttr about him in Morkinskinna.
  • Einarr Skúlason from Borg in Mýrar, western Iceland. 12th century. There is a þáttr about him in Morkinskinna.
  • Óláfr Leggsson of the Lundarmenn clan from Borgarfjörður, western Iceland. Involved in the killing of Jón murti, son of Snorri Sturluson, in 1231.
  • Þorleifr Rauðfeldarson jarlsskáld (‘earl’s poet’) from Brekka in Svarfaðardalur, northern Iceland. 10th century. Known from Heimskringla and elsewhere. Composed poem to Hákon, earl of Hlaðir.
  • Snorri Sturluson, from Borgarfjörður, western Icelandic. 13th century.
  • Þjóðólfr Arnórsson, from Svarfaðardalur in the north. Died 1066. Composed for kings Magnús the Good and Haraldr harðráði. Mentioned in Heimskringla, Sneglu-Halla þáttr, and Brands þáttr ǫrva.
  • Kolbeinn Tumason, of the Ásbirningar clan from Skagafjörður in the north. Died 1208.
  • Sneglu-Halli, from the north of Iceland. Spent time at the court of Haraldr harðráði in the mid-11th century. Main character of a þáttr in Morkinskinna and elsewhere.
  • ?Óláfr Þórðarson hvítaskáld, author of the Treatise.
  • Hallar-Steinn Herdísarson, c. 1200. Possibly from Höll in Þverárhlíð, Borgarfjörður, western Iceland. Composed a poem about King Óláfr Tryggvason.
  • Egill Skallagrímsson from Borg in Mýrar, western Iceland. 10th century.
  • Kormakr Ǫgmundarson from Melur in Miðfjörður in the northwest. 10th century. Poetry preserved in Snorri’s Edda, Heimskringla, and Kormaks saga.
  • Sighvatr Þórðarson from Apavatn in Grímsnes in southern Iceland. 11th century. A þáttr about him is included in Snorri’s Óláfs saga helga.
  • Glúmr Geirason, from Mývatn in the north and Króksfjörður in the northwest. 10th century. Composed to kings Eiríkr Blood-axe and Haraldr gráfeldr. Known from Heimskringla.
  • Markús Skeggjason, lawspeaker 1084–1107. From the south of Iceland. Composed a drápa in hrynhenda meter to Eiríkr the Good Sveinsson, king of Denmark, known from Knýtlinga saga.
  • Nikulás Bergsson, abbot of Munkaþverá, died 1159. Probable author of a guide in Icelandic for pilgrims to Rome and the Holy Land.

Again we may ask: Where did Óláfr get his knowledge of the poets in the list above? To help us answer this, the poets can be divided into four groups:

  1. Generally known figures from Óláfr’s immediate cultural environment: Óláfr Leggsson, Snorri Sturluson, Óláfr Þórðarson himself, and Egill Skallagrímsson (Óláfr’s examples suggest a cultivation of Egill’s poetry among members of his and Snorri’s family.)
  2. Poets known to Snorri in the writing of Heimskringla and the Edda: Þorleifr jarlsskáld, Þjóðólfr Arnórsson, Hallar-Steinn, Kormakr, Sighvatr Þórðarson, and Glúmr Geirason.
  3. Poets known from þættir in the postulated older version of Morkinskinna: Arnórr jarlaskáld, Einarr Skúlason, and Sneglu-Halli (An interesting attempt has recently been made to associate Sneglu-Halli with Snorri: see Pálsson, H. 1992.)
  4. Nationally known figures of the 13th century: Kolbeinn Tumason, Abbot Nikulás Bergsson, and Markús Skeggjason (Markús’s Hrynhenda is quoted in Knýtlinga saga; this meter was also used by Óláfr himself and Arnórr jarlaskáld.)

Óláfr could not have obtained his examples of the verses of these poets from the texts mentioned previously in the form that we know them. Most of these poets are, however, fairly well-known figures, and so it is interesting to note where in Iceland they came from. The majority are familiar as court poets from the works of Snorri Sturluson and Morkinskinna, but the striking fact emerges that those who are not known from these sources all come from the west of Iceland.

Table 2-9: Two main groups of identifiable poets with verses quoted in The Third Grammatical Treatise that are not found in other sources

Poets from the west of Iceland, including wellknown court poets Older court poets and itinerant poets, chiefly from the north
Egill Skallagrímsson (10th century) Kormakr (10th century)
Arnórr jarlaskáld (11th century) Glúmr Geirason (10th century)
Einarr Skúlason (12th century) Þorleifr jarlsskáld (10th century)
Hallar-Steinn (c. 1200) Þjóðólfr Arnórsson (11th century)
Óláfr Leggsson (13th century) Sighvatr Þórðarson (11th century)
Snorri Sturluson (13th century) Sneglu-Halli (11th century)
Óláfr hvítaskáld (13th century)  

Other than Sighvatr Þórðarson, all the court poets who were not from the west of Iceland came from the north: Kormakr was from Miðfjörður in the northwest, an area in close contact with the Dalir region, as evidenced by Sturla’s quick journey to Miðfjörður prior to the attack by the Vatnsfirðingar clan on Sauðafell in 1229; Glúmr Geirason came from near lake Mývatn; Þorleifr and Þjóðólfr were from Svarfaðardalur; and Sneglu-Halli is said to have been of northern extraction. Óláfr’s knowledge of the skalds from this part of the country (all of whom except Kormakr figure to a greater or lesser extent in Morkinskinna) may perhaps be explained by the older version of Morkinskinna having been written in Eyjafjörður in the mid north, a region to which the tentacles of the Sturlungar extended after Sighvatr Sturluson set up home at Grund in Eyjafjörður in 1215 (see Andersson 1993, 1994; Andersson and Gade 2000:66-83). The fact that the north fell within the sphere of influence of one of his close kinsmen may account for the considerable interest and knowledge Óláfr reveals about the poets from this part of the country.

Outside these two main groups fall three prominent figures from Icelandic history and culture: lawspeaker Markús Skeggjason (who composed his verse in Iceland but sent it abroad to the king of Denmark), Abbot Nikulás, and Kolbeinn Tumason. All three, however, would have been well known throughout Iceland at the time when Óláfr was writing.

One particularly interesting feature of Óláfr’s list is the names that do not appear, in particular those of a number of poets who are well known from the sagas of Icelanders. What these ‘absentees’ have in common, however, is that their reputations were largely Icelandic, rather than having been established at the courts of Scandinavia. This might indicate that the common poetic tradition in the country was centered not at the Alþingi, the annual general assembly held at Þingvellir in southwest Iceland, but at the royal courts of the neighboring countries where the skalds plied their trade, notably Norway. For instance, Óláfr makes no mention of Gísli Súrsson and other poets whose lives form the subjects of sagas and whom we would unhesitatingly include in any general history of skaldic verse. It is also notable that, unlike Snorri in the Edda, Óláfr does not cite a single verse by the Borgarfjörður poet and saga hero Gunnlaugr ormstunga (‘Snake-tongue’). The feud between Gunnlaugr and SkáldHrafn over Helga the Fair was certainly well known in the first half of the 13th century since it is twice referred to in Egils saga. According to Gunnlaugs saga, Gunnlaugr was a prominent poet of the first order, but his absence from Óláfr’s Treatise perhaps suggests that the writer of the saga late in the 13th century inflated his hero’s poetic reputation and even ‘forged’ some of the verses attributed to him in the saga himself. This might explain Óláfr’s silence on this ‘well-known’ poet from his own part of the country.


An analysis of the verse examples in The Third Grammatical Treatise leads us to the general conclusion that, while collecting them, Óláfr made use of such written sources as were available to him at the time, in particular the works of his uncle Snorri Sturluson. Óláfr also cites a number of otherwise unknown stanzas by poets quoted in these sources, which might suggest that Snorri and Óláfr knew more verse than Snorri chose to incorporate into his writings. This in turn might point to a livelier poetic tradition than appears solely from the written works, which again argues against the claim that writers like Snorri were in the habit of composing verses themselves and attributing them to the characters in their sagas. Some of the verses by well-known poets (that is poets whose poetry is known in older books) appear to have been composed to women and deal with love, which might explain why they were not committed to writing in the works where these poets play a prominent role — whereas such verses could be cited out of context in the Treatise. Taking all the evidence together, we may conclude that Óláfr’s knowledge of verse from oral tradition was restricted to certain categories:

  1. Generally known Old Norse skalds of the early period
  2. Court poets and itinerant poets, chiefly from the north of Iceland, also familiar to Óláfr from the works of Snorri Sturluson and Morkinskinna
  3. Poets from the west of Iceland, either men who had achieved a reputation at the courts of Scandinavia in earlier times or contemporaries of Óláfr
  4. National figures from Óláfr’s own times

Figure 2-2: Categories of verses and poets referred to in The Third Grammatical Treatise

In light of these well-defined categories, it is unlikely that the absence of poets from other parts of the country can be explained through none of them having composed any verses that Óláfr considered suitable for exemplifying the stylistic concepts he was trying to expound in the Treatise. The findings may rather be taken as a general indication that the knowledge among saga writers of stories and poems from regions other than their own was severely restricted; writers in one region would probably never have heard the stories and poems from another, and would have had little opportunity to meet them in books before the advent of saga compilations such as Mǫðruvallabók and Vatnshyrna in the 14th century—and even then it would have been only the select few who had access to the comparatively limited number of vellum manuscripts in circulation. The most assiduous recorders of ancient poetry in Iceland were Snorri and his kinsmen in the west, and so it is hardly surprising that poets from their areas of influence predominate in the surviving sources. Even so, there were and are poems known from other parts of the country, and so it is interesting that Óláfr’s examples, which we may suppose he had acquired from oral tradition, should be so confined to his immediate area. This means, for instance, that we can say nothing for sure about what he might have known from other parts of the country.

This local character of Óláfr’s poetic knowledge also gives us reason to doubt that the knowledge of stories of Icelandic events extended to all parts of the country in the preliterate period, i.e. that so far as the oral historiographical tradition went the country constituted a single cultural unit. Studies of folktales in more recent times indicate that they are often restricted to particular localities and little known outside them; this was also true in Iceland up to the time of the awakening and cultivation of national consciousness in the 19th century, when the country’s folktales were first collected as a whole and made available in a single book, Jón Árnason’s Þjóðsögur. Before the days of national identity, which goes back only to the Romantic movement, we may suppose that people’s cultural horizons did not stretch far beyond their immediate neighborhoods, where people knew the local customs and had the landscape and placenames constantly before their eyes to keep alive the memory of the main characters and events that formed the substance of their stories.


[ back ] 1. AM 242 folio. Strictly speaking, the title Málskrúðsfrœði applies only to the part of the work dealing with poetic style. The treatise is also preserved in full in AM 748 Ib, 4to and in fragmentary form in AM 757a 4to and AM 757b 4to. The main editions are in Edda Snorra Sturlusonar (1848-87), vol. II, pp. 397-427 (with translation into Latin) and Ólsen 1884, Jónsson, F. 1927, and Krömmelbein 1998.

[ back ] 2. The task of selecting examples—essentially the same task as that faced by Óláfr hvítaskáld—is expressed as follows in the introduction to a modern Icelandic literary handbook: ‘Helsta nýbreytni bókarinnar er því fólgin í því að bókmenntafræðileg hugtök skáldsögunnar eru skýrð á íslensku og skýringardæmi valin úr íslenskum bókmenntum. Í því sambandi vil ég taka fram að ég hef leitast við að velja dæmi úr bókum sem ætla má að séu vel kunnar flestum íslenskum áhugamönnum um bókmenntir þegar þess hefur verið kostur, en fremur forðast að seilast til sjaldgæfra bóka.’ (‘The chief innovation of this book is that the literary concepts applicable to the novel are explained in Icelandic and the illustrative examples taken from Icelandic literature. In this connection I wish to say that I have, whenever possible, sought to choose examples from books I have reason to believe are well known to most Icelanders with an interest in literature and tried to avoid delving into books that are rare or obscure.’) (Njarðvík 1975:9-10). It is not unlikely that Óláfr’s choice of examples in his handbook was guided by similar considerations.

[ back ] 3. Þorleifr was one of Snorri’s closest allies. His father Þórðr was the brother of Guðný Bǫðvarsdóttir, the wife of Hvamm-Sturla, mother of Snorri, and grandmother of Óláfr.

[ back ] 4. If Ólsen’s conjecture that Óláfr spent the summer of 1239 in Sweden is correct, and since he was certainly back in Norway in March 1240 and Valdimarr died in 1241, the only period Óláfr could have spent with Valdimarr in Denmark is the winter of 1240-1.

[ back ] 5. Þórðr of Hítarnes was the half brother of Aron Hjǫrleifsson’s father and was with Þorgils during the attack on Stafaholt. Óláfr composed a poem (Aronsdrápa) in honor of Aron Hjǫrleifsson; Arons saga says that Óláfr and Aron were friends and that the poem was composed on the occasion of Aron’s journey abroad in 1227.

[ back ] 6. Nordal, G. (2001), published after the Icelandic version of this book had gone to press, provides a detailed survey of the learned poetic tradition in the 12th and 13th centuries, making extensive use of Ólafr’s Treatise (see esp. pp. 83-6) and drawing on an earlier version of the present chapter (Sigurðsson, G. 2000).

[ back ] 7. Ofljóst, see note 11, p. 8. Óláfr’s example is: ‘Víst erumk hermð á hesti / hefir fljóð ef vill góðan’ (Ólsen 1884:66-7). Óláfr construes the first line as representing ‘legg ek á jó reiði þokka’ (‘lay I on a horse a sense of anger’). From this, he juxtaposes ‘jó’ (horse) and ‘reiði’ (anger) to produce the woman’s name ‘Jóreiðr’ (dat. Jóreiði), and then takes in the last word in the next line ‘góðan’ (good) to complete the syntax ‘legg ek a Jóreiði þokka góðan’ (‘I lay on Jóreiðr good desire’). From what remains of the second line, ‘hefir flióð ef vill’ (‘has a woman if wishes’), he constructs ‘konu má ná’ (‘can get a woman’), and then puts ‘má’ and ‘ná’ together to form the man’s name ‘Máni’ (gen. Mána), i.e. ‘konu Mána’ (‘Máni’s wife’)! In other words, the verse is a (well) concealed declaration of love to Jóreiðr, the wife of Máni. This explanation of Óláfr’s demonstrates, if nothing else, how far off the mark many modern conjectures about the meanings of skaldic stanzas must be. On this example, see Snædal, M. 1993:216-7.

[ back ] 8. The stanza that is generally presented as the penultimate stanza of Arinbjarnarkviða is also preserved only by Óláfr but has not received the same attention as the ‘final stanza.’ These two verses do not stand together in The Third Grammatical Treatise.

[ back ] 9. Bjarni Einarsson 1961:57-9 discusses the verse found in both the Treatise and Bjarnar saga Hítdœlakappa (and agrees with Konráð Gíslason that it is probably by Bjǫrn Hítdœlakappi, a matter on which nothing will be said here) and the verse that Óláfr ascribes to Kormakr but which is not in his saga: see below for the verses that are not known from other sources. Einarsson takes the view that Óláfr may have come across this verse in Kormaks saga but misremembered it, and it is thus a corrupt version of his own compilation that he presents as Kormakr’s in the Treatise. In view of the number of verses Óláfr is happy to present without named authors, it is difficult to see why he should attribute this particular verse to a named poet if there was no tradition that this was indeed who it was by.

[ back ] 10. See Andersson 1993, 1994; also Andersson and Gade 2000:66-83.