Saxo on Runes and Icelanders

As is often the case for medieval historiography, Saxo Grammaticus’ great Gesta Danorum (c. 1208–1219) contains only a few references to its source material. Occasionally Saxo will mention another historian or allude to oral transmission, but the fullest account is found in two much-cited passages in the preface1. At the outset, Saxo refers to Danish songs preserved in stone:

Nor will I let it be unknown that the Danes of ancient times, having performed their laudable exploits, in the pursuit of glory and in imitation of the Roman style not only recounted these in a kind of poetry, but also took care to have the deeds of their ancestors, known from songs in their native tongue, carved on stones and rocks in the letters of their own language2.

He adds further down that they used cliffs instead of books, and that he has “adhered to their traces as if to ancient volumes and endeavored to produce a faithful translation, where poetry is represented by poetry3.” Clearly, this is an allusion to runic inscriptions. Saxo is saying that the Danes used to compose songs about their deeds, and on the basis of these inscribe their history in stone—at least to some extent in poetic form, to judge from his words on adapting “metra metris”. He maintains that his versions, even if they are adorned in Latin attire, will be close to the originals.

This is naturally a fanciful account and has long been recognized as such. Owing to these runic sources, Saxo asserts that his work provides “reliable information about antiquity (fidelem uetustatis notitiam)4”. To that end, exaggerating the importance of runes lends some authority to his history, mainly in their role as written, or rather carved, native sources.

Saxo goes on to praise the dedication of the Icelanders in preserving the historical lore of other nations; their industria5, their “diligence”, cannot go without mention. He acknowledges his use of Icelandic sources, commenting: “Their treasures of historical testimony I have perused with great interest, and a not inconsiderable part of the present work is founded on their narratives6.” The question of whether these works were mainly written or conveyed orally is a difficult one, since the reference to “treasures” (thesauros), though seemingly indicating physical objects, could just as well be a reference to oral compositions. The word is, however, later used to refer to Bede’s historical works: “in the most revered treasures of his volumes (sanctissimis suorum uoluminum thesauris)7”. Even so, we should not be too hasty in making any conventional assumptions on the nature of these sources. What were the circumstances around his reception of these and in what form did they reach him? Might an Icelander have written something down at his request or could Saxo himself have made notes on the basis of a recitation? The line between oral and written transmission is not always clear. Besides, it is doubtful how much of Saxo’s own account can be taken at face value; his assertions most likely disguise a greater complexity involved with collecting and using sources. Hence, I anticipate that a closer look at all the references to runes and Icelanders in the work can give us a clearer picture of both his professed and his actual use of sources. At least in the preface, Saxo seems to be downplaying the Icelandic contribution, especially when one compares with Sven Aggesen and Theodoricus—soon to be mentioned below. My aim is to explore whether or not this is the case in the rest of the work. Is the importance of runic inscriptions and native Danish sources exaggerated in comparison to the references to Icelanders?

Sven Aggesen and Theodoricus on Their Sources

Two other historians in Saxo’s time refer to Icelanders in a similar manner. Sven Aggesen, writing his Brevis historia regum Dacie (A Short History of the Danish Kings) perhaps in the late 1180s, alludes to Saxo’s ongoing work and was himself a source for parts of Saxo’s history8. In chapter one, Sven Aggesen gives a short account of the life of Skjöld whom he styles the first king of Denmark. He adds that “after him, the kings were first called Skjöldungar in Icelandic songs9.” Besides a brief mention of the raising of the larger Jelling stone, the only possible reference to runes by Sven Aggesen is found in the story of how King Vermund guided by the markings on a stone, had a sword dug up for his son Uffi10.

The second historian is Theodoricus Monachus, whose Historia de antiquitate regum Norwagiensium (History of the Ancient Norwegian Kings) it is less clear if Saxo had read. This work, finished in 1188 at the latest, would naturally have been of great interest to Saxo, who likely sought all the possible sources he could lay his hands on and there could have found further inspiration on how to write a national history. One candidate for the authorship of this work, Bishop Þórir of Hamar, died while passing through Denmark in the winter of 1196–1197 on his way home from a diplomatic mission to Rome11. Another suggestion is Þórir Guðmundsson, canon in Oslo and later archbishop of Nidaros, favored by the most recent editor Egil Kraggerud12. Whatever the case may be, in the prologue Theodoricus describes his aim to briefly sketch the history of Norwegian kings, “according to what I have been able to learn, by asking those with whom the memory of these [kings] is believed to be particularly strong, those whom we call Icelanders, who preserve these famous matters in their ancient songs13.” Towards the end of the prologue, he adds that not he, but his sources are the ones responsible for the validity of his account, since: “I have put to pen what I have heard, not myself seen (non uisa, sed audita conscripsimus)14.”

As such, all three writers acknowledge that Icelanders were particularly important for transmitting knowledge of the past, but whereas Sven Aggesen and Theodoricus placed great emphasis on poetry, Saxo speaks in more general terms of their devotion to history. For Sven Aggesen and Theodoricus the songs of the Icelanders guaranteed the truthfulness of the events described; this aspect is in turn lost with Saxo, the Icelandic role being replaced with runes. For him, the Icelanders are trustworthy in light of their learning and interest in the history of other nations. Perhaps he saw the magnitude of their historical writings as evidence enough of their scholarly attainments, without the need to mention poetry. By Saxo’s time, Icelandic industria had already engendered a substantial literature; the Icelanders were after all first, seeing as their literary production on historical matters had begun in the early twelfth century. This knowledge was to a great extent based on the skaldic tradition, but this vital detail is omitted by Saxo. It is, however, important to note that it was long until written sources incorporated Old Norse poetry to any large extent; the mostly lost history of Norwegian kings, Eiríkr Oddsson’s *Hryggjarstykki (“back-piece”, possibly a kind of wild duck), from around the mid-twelfth century is believed to have partly relied on skaldic poetry as sources, but without making any extensive use of citations. Likewise, Ari Þorgilsson includes only a quarter-stanza of poetry in his Íslendingabók (Book of the Icelanders). The same certainly held true for lost historical works in Latin, such as Sæmundr fróði’s history of Norwegian kings or the two lives of Olaf Tryggvason by Oddr Snorrason and Gunnlaugr Leifsson from the late twelfth century, as well as other works15, in which adapting skaldic poetry would have been an almost insurmountable challenge to the translator16. Even though Saxo is mostly taking on the simpler eddic genre, this tradition was by his time kept alive mainly by Icelanders.

Runic Sources

In the preface Saxo instead underlines the value of runic inscriptions, stating that the Danes preserved their history in songs carved on cliff and stone. If we for a moment disregard the questionable aspects of this statement, runes admittedly do have a few advantages. They are after all native Danish sources, more or less contemporary to their carving, a recurring feature in the Danish landscape, and as inscriptions could be seen as a valid equivalent to written histories. Saxo likely also believed that the poems on Danish heroes and kings were Danish in origin, however doubtful this might appear to us. Even if his sources indeed were Icelandic, describing how he had accessed older poetry through their mediation might have seemed an unnecessary detour.

As convenient an explanation as runic transmission of poetry may be, still, there is little to substantiate it. Although there are cases of poetry in runic inscriptions—and Saxo might even have drawn inspiration for one or two of these (I am thinking primarily of the short stanza by Hiarni to commemorate King Frothi, to be mentioned later)—it is clear that runic inscriptions were never meant to carry large quantities of text. Their potential to do so might, however, have seemed a reasonable assumption for someone less knowledgeable about runes, as has been previously noted in an article by Gottskálk Jensson17.

Runes appear in several other places in Gesta Danorum. Further down in the preface Saxo refers to Runamo, a rockside in Blekinge inscribed with “signs, meant to be read (factis ad legendum figuris)18”. King Valdemar I, he tells, sent men to interpret and copy the inscription, but “they were not able to make any interpretation of these (nihil ex eis interpretamenti comprehendere potuerunt)19”. It has in later times been determined that the cracks in the rock are in fact a natural formation20. All the same, in Book VII, Saxo claims that King Harald “Hilditan”, i.e., “Wartooth”, had the same rock inscribed in honor of his father Halfdan Biargramm21. There is a slight contradiction here. How could he have known what the inscription contained if the runes were unreadable? Though it may well be that local tradition had told of its background, this is never specified and no allusion to its meaning or history is made in the preface. Moreover, though it is not stated explicitly, most readers would get the impression that Saxo bases his account of King Halfdan at least in part on this inscription; it is, after all, the only supposed text mentioned in relation to this king22. This kind of ambiguity might give us some clue as to Saxo’s strategies in representing his sources; it is misleading, to be sure, but mainly by way of inference and omission. The contradictions may well be the result of a mixture of guesswork and forgetfulness. Another important advantage derives from the mention of these inscriptions; it establishes the kings of the Gesta Danorum in the real world, since these “runes” are depicted as a surviving contemporary account, still visible in the landscape and—in theory—still readable. The Runamo inscription, or rather, formation of cracks, is furthermore quite long, and would have been a comparatively abundant source, had it been genuine23.

Saxo mentions two other inscriptions made on stone. Book VI begins shortly after the death of Frothi III, the great king whose peaceful reign supposedly coincided with the lives of Christ and Augustus. As the king was believed to be without an heir, the leading men of the realm were uncertain as to who should succeed him:

They deemed that man most worthy of the scepter, who to increase Frothi’s glory would attach a song of praise to his mound and in a splendid inscription convey the fame of the dead king to posterity24.

A certain young man, called Hiarni, “highly skilled in poetry (admodum poesis peritus)25”, composed an excellent stanza that won him the kingship, by Saxo presented in an appropriate elegiac form:

Frothonem Dani, quem longum uiuere uellent,
    per sua defunctum rura tulere diu.
Principis hoc summi tumulatum cespite corpus
    aethere sub liquido nuda recondit humus.

The Danes long carried the dead Frothi, since they wished his life to continue, throughout their lands. The bare ground now holds the great prince’s body buried under turf beneath the clear sky26.

Attaching “a song of praise” and after the verses mentioning its “weaving together of letters” must mean that a stone was raised and inscribed in connection to the mound27. The fact that Saxo is mentioning the inscription, and in this case even transmitting a stanza which agrees with his description in the prose of Frothi being carried dead throughout the lands, is an invitation for us to believe that he bases his verses and the story of how the king’s corpse was carried through the realm, on a runestone extant from this time. Judith Jesch has pointed to the similar circumstances around the Karlevi stone, situated on Öland and commemorating a fallen Danish warrior28. This inscription was placed close to the burial mound and contains an Old Norse stanza in dróttkvætt, which likewise mentions the mound itself. A close parallel to the clear sky motif can also be identified in a now lost runestone from Aspa, Sörmland, dating to the first half of the eleventh century. Two brothers honor their father, “a stouthearted hersir (hærsi hugsniallan)”, as being “the best under heaven (Hann vas und hifni bæstr)29.” The words “best under the blue sky (bazt und blǫ́um himni)” occur also in Þjóðólfr ór Hvini’s Ynglingatal extolling King Rǫgnvaldr, to whom this poem was likely composed.30 His father Óláfr Geirstaðaálfr is commemorated in the preceding stanza, again with reference to a mound: “The war-daring army-king now lies covered by a mound in Geirstaðir31.” Though no extant text is close enough to suggest an actual model for Saxo’s stanza, the Latin translation at least allows one a glimpse of some traditional tropes of funerary, often runic, poetry in the Norse tradition.

A similar suggestion is found in a passage concerning Ragnar Lothbrok:

After the king of the Bjarmians had been killed and the king of the Finns put to flight, Ragnar had stones raised with inscriptions of his deeds and placed them high as an eternal monument of his victory.32

Once more, Saxo’s knowledge of this battle could—in the eyes of a credulous reader—be at least partly based on this monument.

Runes do not only appear on monumental stone inscriptions, but in smaller circumstances as well. There are four references to “signs” and “inscrip-tions” that are used for magical purposes. One such example is a piece of wood carved by the giantess Harthgrepa and placed by Hadding under the tongue of a dead man to make him speak33. As has been noted in the commentary by Paul Herrmann, in Hávamál 157 the speaker, who a few stanzas back is identified as Odin, boasts that he can carve and paint runes for a corpse: “so that this man may walk and talk to me (at sá gengr gumi / oc mælir við mic)34.” In another example from Saxo, Odin touches Rinda, a Ruthenian princess (though in Norse sources likely a giantess), with a piece of bark carved “with spells (carminibus)”, which renders her unconscious35. Further on, King Oller (a euhemerization of the god Ull) travels over the water on a bone carved with “fearful charms (diris carminibus)36”, this being, according to Karsten Friis-Jensen: “[c]learly a reference to ice-skating37.” Saxo’s claim to Ull’s use of a rune-inscribed bone cannot be corroborated in Old Norse sources, where he is commonly associated with skis, as described in Gylfaginning38. This makes it difficult to determine its exact origin; it could conceivably be a misunderstanding or a variation on the myth.

A possible fourth example of runes in a magic context can be found in Saxo’s version of the poem Bjarkamál, in a passage where Ruta tells her husband Biarki how to catch sight of Odin on the battlefield. He must “sanctify his eyes with the sign of victory (sacraturus uictrici lumina signo)”, and then look through her bended arms39. Though this is admittedly not a case of carving, the “sign of victory” has been seen in connection with the sigrúnar that appear in Sigrdrífumál 640. Here Sigrdrífa exhorts Sigurðr to carve “victory runes” on his sword, and twice utter the name of Týr. This parallel is far from convincing. Is Saxo saying that Biarki should make the sign of the Týr rune with his hand over his eyes? It is unclear, and even more difficult to know what the original wording was in the version of Bjarkamál which Saxo had at hand, these parts being lost, or if they are his own invention, perhaps after the fashion of some other poem or text that he has encountered.

Finally, two examples of runes being used in correspondence are present in the story of Amleth. King Fengi sends his nephew Amleth to the king of the Britons, along with two of his henchmen bearing a letter asking for the young man to be executed: “a letter carved in wood—for that was once a common type of writing material (literas ligno insculptas—nam id celebre quondam genus chartarum erat)41”. Amleth finds the letter, and “made sure to scrape away and replace the characters with new ones, changing the meaning of the command and directing the judgment toward his companions (curauit abradi nouisque figurarum apicibus substitutis damnationem suam in comites suos mutato mandati tenore conuertit)42”. A similar event takes place in the fourth book, when the queen of Scotland adjusts the letter Amleth brings from the British king asking for her hand43. Since she has no wish to marry an old man, she erases the letter and composes a new one, asking her to marry the messenger Amleth instead.

This motif of sending a letter telling the recipient to kill the messenger appears in the Second Book of Samuel, where David sends Uriah with a letter to Joab44. In a similar vein to Saxo’s story, the motif of changing a dangerous letter appears in several Old Norse texts, most famous of which is Guðrún’s letter in Atlamál in grœnlenzku and Vǫlsunga saga. Here, Guðrún tries to warn her brothers that Atli is planning to kill them, told in the poem, which is a fairly late composition, with the following words: “She took to carving runes, / Vingi distorted them45.” The saga gives a fuller account, telling how:

Gudrun carved runes, and she took a gold ring and tied wolf hair to it and handed it to the king’s messengers. On the king’s command they then set off. Before they came ashore, Vingi saw the runes and changed them, [to seem] as if Guðrún in the runes were eager for them to come meet him46.

Afterwards, Kostbera, the wife of Guðrún’s brother Hǫgni, sees that the runes have been altered and cautions him against going: “I interpret the runes that your sister carved: the bright one did not bid you to come at this time47.” The same utterance appears in Vǫlsunga saga as: “you cannot be very skilled at reading runes, if you believe that your sister has sent for you at this time48.” Her warning is ignored by Hǫgni. This device of altering a written message is thus most closely paralleled in these two texts, suggesting that the Old Norse tradition might have furnished Saxo with a possible source for his story—even if no exact counter-part is extant today.

Additional examples of runes are found on two pieces of equipment. It is mentioned that Amleth’s shield is painted with scenes from his wars, and when the Scottish queen has her men steal it, she interprets the various images based on “inscriptions added to it (affixis notulis)49”. Also in Book IV, Frithlef hears a short poem sung by swans, and presently a belt falls down from the sky, inscribed with letters providing an explanation of the song50.

Though these instances of runes used for magic and correspondence play an important part in the narrative, they do not have the same authoritative purpose as in the more monumental inscriptions. The mention of them does, however, underline the presence and role of runes in the Danish society of the past, in many instances seen through the lens of antiquarian constructs. It is also worth noting how many different uses of runes that Saxo is aware of–functions that they shared both in real life and in Icelandic sagas. There are many such inscriptions from the Viking Age as well as from Saxo’s own time, in wood and metal, and they likewise appear in the sagas. We have already seen literary examples from Atlamál in grœnlenzku and Vǫlsunga saga, and a few late runic letters appear in the archaeological record, especially through findings from Norway in Trondheim and Bryggen in Bergen. There are also early examples of rune sticks from Hedeby, at this time a Danish town, and Staraya Ladoga in Russia, but these do not seem to be letters, containing instead the fuþark, poetry and, possibly, incantations51. As an important external witness, Rimbert tells how Ansgar and his companions after devoting themselves to missionary activity in Birka with King Björn in the year 830 returned to Louis the Pious “with a letter shaped by the king himself in their own manner (cum litteris regia manu more ipsorum deformatis52)”.

It is clear from these ten examples that runes play a surprisingly large role in Saxo’s narrative. The use of runes for magic purposes or correspondence is likely taken from his sources or inspired by the way that runes are presented in similar stories. They could also in part be the result of what Saxo had seen with his own eyes. Runic letters such as those from Bergen and Trondheim might have crossed Saxo’s path, and the same could be true for lead amulets, which seem to have been plentiful in Denmark53. More importantly, the circles in which he found himself undeniably had antiquarian interests. A man of the greatest importance for Gesta Danorum, Saxo’s employer Archbishop Absalon, is mentioned on a runestone from Norra Åsum church, close to Kristianstad in Scania: “Christ, son of Mary, help those who built this church, Archbishop Absalon and Esbern Mule54.”

The many occurrences of runes in Gesta Danorum align with Saxo’s exaggeration in the preface of runic inscriptions as bearers of historical and poetic texts. He gives the impression of justifying his description of his sources by affirming the importance of runes for the ancient Danes55.

Saxo and the Icelanders

I made the claim that Icelanders, in contrast, are less visible than they perhaps deserve. In light of Saxo’s probable dependence on Icelandic sources, it is curious that they are only described as keepers of ancient lore, and not mentioned in relation to the poetry that appears in the work. Though in fact historical learning often depended on poetry, Saxo does not connect the two roles in relation to the Icelanders. He writes a passage concerning a man whom scholars have thought to be identical with a skald, Arnoldus Thylensis (“the Icelander”), who is mentioned in Book XIV as being part of the retinue of Absalon, at that time bishop of Roskilde, while fighting against the Slavic Wends in 1168. The case has been made that this is Arnaldr or Arnhallr Þorvaldsson, who appears in Skáldatal as one of the skalds who had composed poetry for Valdemar I56. The dating certainly agrees, as do the names. Provided that the identification is correct, one wonders why Saxo fails to mention this man’s profession. He describes him merely as “knowledgeable in antiquity, skilled at foreseeing the future and telling stories (Nec minus antiquitatis quam diuinationis peritus sollerti historiarum narratione callebat)57”. In addition to these abilities, Arnoldus is shown to be a virtuous and courageous man, being eager to follow Absalon into battle rather than remaining with the king. Assuredly, Saxo’s words are reminiscent of Theodoricus in suggesting that the Icelanders have their historical knowledge from their poetry, but it is remarkable that Arnoldus’ main occupation goes without mention, considering the likely identification with the skald.

Could it perhaps be that Arnaldr did not perform his poetry before Absalon or the other Danes present? There is certainly a distinct possibility that they would not have understood it. Already around 1150 in a stanza preserved in Knýtlinga saga the skald Einarr Skúlason complains of not receiving any gift from King Sven Grathe, saying that “[t]he Danish lord considers fiddles and flutes more valuable—that is not good enough (Danskr harri metr dýrra—dugir miðlung þat—fiðlur)58.” This disinterest in older traditions could likely be connected to the German influence at Sven Grathe’s court, which Saxo describes:

Since he found native customs to be less refined, even coarse and artless, he changed and substituted them for an elegance sought from neighboring peoples, and by spurning the Danish manner emulated the German one59.

Perhaps as a typological prefiguration in the legendary part of his history, the court of King Ingiald is in like manner disparaged for its German feebleness and dissolution60. A more direct reference to German minstrels also appears in the Gesta Danorum, when a Saxon cantor tries to warn Cnut Lavard against trusting his cousin, through a song on the treachery of Grimhild against her brothers61.

One wonders whether such lack of enthusiasm for Old Norse poetry prevailed with other Danish kings and nobles. Certainly, Skáldatal records several skalds for successive kings: two under Valdemar the Great, one under Cnut VI and four under Valdemar II62. Even so, the necessary knowledge of mythology and poetic traditions to understand or appreciate complex kennings was undoubtedly on the wane in Denmark. Would Arnaldr therefore have limited his performances to storytelling? Did he merely use the poems that he knew as sources? Then again, the easier eddic poems might have been largely intelligible, and some of them would assuredly have been a part of the skald’s repertoire. Had Arnaldr ever wanted to make use of his knowledge, it would likely have been through reciting heroic poetry on the ancient Danes. If so, then Saxo’s omission would be intentional.

It is unclear how long Arnoldus stayed in Absalon’s service. The events described by Saxo take place in 1168, which is probably around two decades before he began writing his history. Arnaldr is mentioned in Skáldatal as having been at the court of Valdemar I, but not of Cnut VI, who succeeded him to the throne in 118263. If he were to be considered as one of Saxo’s sources, it would likely have been a case of indirect transmission through Absalon.

Several Icelanders appear throughout the narrative, the most notable perhaps being Thorkil, who in Book VIII leads two expeditions north, first finding the corpse of the giant Geirrøth and later seeking out Utgartha-Loki, after which Thorkil converts to Christianity64. There are many similarities with Þorsteins þáttr Bœjarmagns, where the hero likewise meets Geirrøth, but takes on a role even closer to Thor in actually slaying the giant65. In the same book Saxo also tells the story of the Icelander Ræf who makes the Norwegian and Danish kings compete in generosity towards him66. He is paralleled by Refr or Gjafa-Refr in Gautreks saga, but similar events also take place in Auðunar þáttr vestfirska, which has an Icelandic protagonist67. Such stories where Icelanders play the main part have likely reached Saxo through Icelandic mediators68.

But to return to the question of poetry, there is one instance of an Icelander whose poetic gifts are mentioned, namely in Saxo’s prose rendition of Brávallaþula, a list of the warriors who took part in the battle of Bråvalla between Harald I (“Hilditan”) and Sigurd Ring. In one passage recording Harald’s champions, the text expounds on their various abilities, their equipment, intellect and skill in arms, but also on their poetic expertise: “they also knew how to compose poetry with great ease in their native tongue (poeses quoque patrio sermone contexere promptissime calluerunt)69”. Among these warriors is Blend, “a resident of far-off Thule (ultime Tyles incola)70”, i.e. Iceland. The original poem has been lost, but there is a later prose rendition from Iceland, which also testifies to their poetic gifts, in Sǫgubrot af nokkrum fornkonungum, an extant portion of the younger redaction of Skjǫldunga saga. In the corresponding passage, the name appears as Blœngr, yet there is no comparable mention of him or anyone else in the group being an Icelander71. Saxo’s reference to Iceland is, however, anachronistic, since the battle takes place before the settlement. Perhaps the author of the saga had a different redaction of the poem as his model than Saxo, or made the change independent of his source, effacing the incongruous origin of the warrior.

Although missing from Sǫgubrot, one could see Saxo’s words about Blend’s homeland and his poetic gifts as hinting at an Icelandic source for his version of the poem. Still, there might be some clues as to his direct model. Some scholars have argued for a written South-East Norwegian exemplar based on the epenthetic /a/ suggested by names such as Sambar for Sámr or Gotar for Gautr, but since this at the time is present only in runic inscriptions (which hint at the spoken language) and not in the manuscript tradition, oral transmission would be a more reasonable explanation. This, of course, does not rule out that the poem might have originated in Iceland72. There are, however, signs that this distinctive orthography is the result of Christiern Pedersen’s work on the editio princeps of Gesta Danorum, which I will expound upon in my upcoming dissertation.

Finally, I would also mention an exception to the otherwise prevailing emphasis on writing. Disregarding tradunt ueteres, proditum est (“the ancients tell”, “it is said”)73 and similar unspecified references, there is a noteworthy description that appears in connection with Saxo’s version of Bjarkamál, the longest poem that features in the Gesta Danorum. After its conclusion, Saxo makes a reference to oral tradition:

I have principally composed this series of exhortations in meter, because the same subject, arranged within the confines of a certain Danish poem, is recited from memory by many who are knowledgeable in antiquity74.

“Danish” could naturally be a more general reference to the language used also in other parts of the North, but it is notable that Saxo fails to mention any Icelander in the case of a poem which most certainly reached him through Icelandic informants. The intended meaning seems instead to be that the poem is still declaimed in Denmark in Saxo’s own time. Not unimportantly, this would make it a native source. Indeed, there are signs that Danish should be understood in this more limited sense in Gesta Danorum. Saxo at one point presents Norwegian and Danish as different languages, when recounting how Erik the Eloquent of Norway sent two of his men who were fluent in Danish to gather information upon first arriving in Denmark75. He apparently felt that the differences were great enough, as he saw them in his own day, to merit such a distinction.

Yet if Saxo had begun speaking of Icelandic mediation, this would have invited doubts as to whether the poetry was genuine or not. The literary production in Iceland was already considerable and Icelandic skalds still distinguished themselves as poets at the Nordic courts76. There is, moreover, good reason to believe that parts of Bjarkamál are twelfth-century additions. Of the seven fragments preserved of the Old Norse poem, three or more are doubtful, since they show some metrical peculiarities and a few kennings which are paralleled only by twelfth-century skalds, as noted first by Axel Olrik and recently by Margaret Clunies Ross in the latest Skaldic Poetry edition77. In my upcoming dissertation, I argue further that stanzas three to seven of Bjarkamál are younger additions, having traits that separate them from the first two. To judge from these fragments, it seems as if a large part of the poem was composed quite late, presumably by an Icelander. For his part, Saxo avoids any implied changeability of poetry that could stem from a complicated tradition, while focusing on its Danish origin78.

With Saxo’s narrow use of “Danish” in mind, one may also see abundant references to the poetry being native. He describes how the name of King Gram became synonymous with “prince” in “the oldest poems of the Danes (in uetustissimis Danorum carminibus)79”, and as seen previously Hiarni was learned in “Danish poetry (Danice [...] poesis)80”. Halfdan Biargramn was skilled in composing poetry “in the native manner (patrio more)81” and Starkath is called a “fluent poet of a Danish muse (Danice uates promptissime Muse)82”. Although Starkath is generally said to have been born in Norway in Icelandic sources, according to Gautreks saga in Agder83, Saxo assigns his birthplace to a region just east of Sweden84. Still, this perennial warrior is closely associated with the Danish court, having served under King Frothi IV, and stirring his son Ingiald from unmanly torpor to avenge his father.

The Advantage of Inscriptions

In light of Saxo’s minimizing of Icelandic mediation, particularly for the poetry, it is not at all surprising that many generations of scholars believed him to have used mainly Danish sources. Seeing through Saxo’s smokescreen, Axel Olrik identified similarities with many fornaldarsögur85, and much later, Bjarni Guðnason determined that Saxo must have used the lost *Hryggjarstykki (cited in later works) and a source common to Skjǫldunga saga86. Olrik as well as other scholars have also made comparisons with the few extant Icelandic sources for the poems87. Both Olrik and Friis-Jensen were nonetheless convinced that eddic poetry was still alive in Denmark by Saxo’s time, since alliterative verses are found inscribed on runestones88. These, however, are rare and somewhat uncertain cases, in light of their fragmentary nature. Even so, if such poetry were still to be found in Denmark, it would clearly have been dependent on the vigorous Icelandic tradition89.

From the examples treated in this article one would surmise that Saxo’s references to runes are meant to give authority to the poetry: the poems are thereby confirmed to be ancient, and moreover work to corroborate the surrounding prose. Since Saxo claims in the preface to have used runes as his sources, one may well imagine that the other runestones in his work were meant to be understood as giving the basis for some of his accounts. Another aspect is the emphasis on writing. The insistence on runic sources could stem from a desire to guarantee the authenticity of poems that in many cases could have been conveyed to Saxo orally. As mentioned, runic sources are not an absurd supposition; numerous Danish runic inscriptions from the Viking Age contain poetry, and there must have been more stones still standing in Saxo’s time, though none of those we know of are longer than a single stanza90. If Saxo ever were inspired by runic verses, the brevity of these must have permitted only a limited use. The impression of Saxo’s reliance on runes is lessened by an additional example. In his description of the death of King Gorm, Queen Thyra implies that their son Cnut had been killed, whereupon the king drops dead91. This is in line with Icelandic sagas such as Jómsvíkinga saga, but differs from that of the smaller Jelling stone, which reveals that Thyra died before Gorm92. Saxo is clearly aware of the monument, since he at one point refers to the erection of the larger Jelling stone93; in this, however, he seems to be dependent on the narrative of Sven Aggesen94. Even if he were ignorant of the content of the smaller stone, rather than seeking out runic sources, he settled for the Icelandic tradition of Gorm’s death.

Friis-Jensen makes a relevant comment on Saxo’s account of King Uffi. After describing the king’s youth, Saxo bemoans the fact that there are not more written records for his later deeds95. “Here as elsewhere”, writes Friis-Jensen, “Saxo is extremely unwilling to acknowledge in explicit terms the existence of an oral tradition96.” Similar laments over a lack of sources can be observed in other places in his work97. In a later article, Friis-Jensen concludes that written sources are described as “a rule rather than the exception98”. This disparaging of oral sources and the emphasis on inscriptions I would interpret as a sign of Saxo’s belief in the constancy of written texts. Runes are after all contemporary to their carving, and a welcome solution if a long oral tradition seems difficult to conceive. In that sense, one could see it as a kind of rationalization; runes should have been or even have the potential to be the true sources of this part of his history. The origin of Saxo’s belief in the written or inscribed text as more trustworthy might be his understanding of Roman inscriptions, as suggested by Friis-Jensen: “Saxo knew, at least from literary sources, that inscriptions were an important genre in Rome, and the literary sources in particular may easily convey the impression that verse inscriptions are dominant99.” Furthermore, during Saxo’s time as a student, possibly in northern France—be it Paris, Reims or Orléans—he would have encountered a plethora of Latin inscriptions, both ancient and medieval.

To Saxo, runes were therefore a useful explanation for the poetry that appears in the work. With this native source, Saxo underlined that the poems are not only composed by Danish heroes and kings, but also handed down in a Danish tradition. Of course, this is in a way a “patriotic fiction”, as Gottskálk Jensson calls it100, and, as has been shown, it has some glaring inconsistencies. One may ask whether this would not have been recognized by his contemporaries. This is far from certain. There were likely few who would doubt the statements of such an unquestionably learned man. Furthermore, Saxo was writing for the future—for future Danes and foreign scholars, who to his mind would likely have less prospect in criticizing his sources.

Ergo, while the poetry in Gesta Danorum gives color and variety to the narrative, this is not its only function. It provides authority to Saxo’s history. It showcases all kinds of meters, in an admirable Latin—the form that Saxo believed that the subject matter deserved. It thus works a little like an exhibition of high culture, showing Denmark at its best. The poetry being “originally Danish” also means that the first half of the history functions as an anthology of sorts, a representation of how varied and rich the poetic legacy of Denmark was, not to mention positively redolent of Vergil and Horace. Certainly, though many of the speakers of the poems are Danes—Gram and Hadding just to name two—there are notable poets of other nations as well, such as Erik the Eloquent and Starkath. Although foreigners, they use their gifts to serve the Danish kings, proving that all wit and verse inspired by a Danish muse can amend and elevate the Danish nation. The poetry sometimes exhorts the audience to manly virtues, to bravely fight for one’s country, or to modesty and restraint; sometimes it is merely used for entertainment. In any case, it is clear that the poetry is not only a parenthesis, but key to how the first half of Gesta Danorum should be understood. Through his emphasis on runes as a medium of transmission from ancient times, Saxo is able to present also the poetry that he includes as an essentially Danish witness to the past.