The Anonymous Verse in the Third Grammatical Treatise

Tarrin Wills

(University of Sydney)

The Third Grammatical Treatise (TGT) was composed around the middle of the thirteenth century by Óláfr Þórðarson, a nephew of Snorri Sturluson. It is divided into two parts: the first is a paraphrase of standard Latin grammatical treatises (specifically Priscian’s Institutiones Grammaticae I-II), supplemented with information about the runic alphabet. The second part is a much closer adaptation of the Ars Major of Donatus on the faults of speech (III, 1-6), illustrated with examples of Norse-Icelandic skaldic poetry. Of the poetic examples, a large amount of material is not found elsewhere and a large proportion of that is anonymous. The anonymous material consists of mostly very short fragments (usually two lines) on a variety of subjects. Despite the existence of six editions of the treatise and a number of scholarly commentaries on aspects of it spanning a century and a half, there has been no explicit attempt to identify, classify or contextualise the anonymous verses in TGT apart from one brief and one indirect attempt by Finnur Jónsson (1920-4 and Skj). This paper will begin to undertake this project. I will identify some sources, propose some reclassifications and assess Finnur’s attempts at grouping and determining the provenance of the anonymous material.

A number of these verses are interesting in themselves, but the general significance of the anonymous material in TGT lies in the information it can provide about its status and transmission. This issue was first explored by Gísli Sigurðsson at the Akureyri saga conference. A revised version of the article was published in 2000; it contains a chronology of the life of Óláfr and a detailed account of the poetry of known provenance. I therefore refer the reader there for some of the details not discussed in the present paper. Gísli concentrated on the verse of known provenance, concluding that Óláfr’s oral sources – and by extension literary knowledge in general – derived largely from either the royal courts of Scandinavia or from the author’s own neighbourhood in Iceland (2000, 112-13). The present paper will show that this conclusion can be extended to the anonymous material, but requires some refinement with respect to that material.

The significance for Óláfr of Norse poetry in general, and by extension that of the treatise, is made quite explicit in the opening chapter of the second section (I have used Finnur Jónsson’s edition of 1927 for the sake of providing a normalised text for these pre-prints; translations are my own unless otherwise noted):

Í þessi bók [Donatus, Ars Major I] má gerla skilja, at öll er ein listin skáldskapr sá, er rómverskir spekingar námu í Athenisborg á Griklandi ok sneru síðan í látínu-mál, ok sá ljóða-háttr eða skáldskapr, er Óðinn ok aðrir Ásíamenn fluttu norðr higat í norðrhálfu heimsins ok kendu mönnum á sína tungu þess konar list, svá sem þeir höfðu skipat ok numit í sjálfu Ásíalandi, þar sem mest var fegrð ok ríkdómr ok fróðleikr veraldarinnar. (TGT 1927, p. 39)
In this book it may be clearly understood that everything is the one art: the poetry which Roman orators learnt in Athens in Greece and then turned into the Latin language; and the song-metre or poetry which Óðinn and other men of Asia brought north into the northern half of the world, and taught men this kind of art in their own language, just as they had arranged and learnt it in Asia itself, where beauty and power and knowledge were the greatest in all the world.

Page 2

This is a clear attempt to position Norse-Icelandic poetry as a classical literature on a par with the literature of Greece and Rome. It is evidently influenced by Snorri’s account of the origin of the Æsir in the prologues to Snorra Edda (Faulkes 1982, 4-6) and Heimskringla (ÍF 26, 11-16). In those works Snorri equates classical deities with the Norse gods and accounts for their origin in the migration of the Æsir from Troy.

To further contextualise Óláfr’s project in writing TGT, I also refer to the preface to the Gesta Danorum of Saxo Grammaticus. Óláfr was resident in Denmark ca. 1240-1 at the court of Valdemar, whom he refers to as minn herra Valdimarr konungr ‘my lord King Valdemar’ (TGT 1927, 29) when introducing a runic formula attributed to him. An interest in runes at the court of Valdemar is already signalled in Saxo’s history:

Non ignotum volo, Danorum antiquiores conspicuis fortitudinis operibus editis gloriæ æmulatione suffuses Romani stili imitatione non solum rerum a se magnifice gestarum titulos exquisito contextus genere veluti poetico quodam opere perstrinxisse, verum etiam maiorum acta patrii sermonis carminibus vulgata linguæ suæ litteris saxis ac rupibus insculpenda curasse. … quibus scribendorum series subnixa non tam recenter conflata quam antiquitus edita cognoscatur (Olrik and Ræder, 4)
I should like it to be known that Danes of an older age, filled with a desire to echo the glory when notable braveries had been performed, alluded in the Roman manner to the splendour of their nobly-wrought achievements with choice compositions of a poetical nature; not only that, but they engraved the letters of their own language on rocks and stones to retell those feats of their own ancestors which had been made popular in the songs of their mother tongue… My chronicle, relying on these aids, should be recognised not as something freshly compiled but as the utterance of antiquity (Fisher, 5)

Saxo likewise posits the poetry and (runic) stone inscriptions of Scandinavia as a kind of alternative classical literature, although he represents it as parallel with that of Rome rather than sharing a common origin, as in Óláfr’s account. This view of the status of poetry and runes forms the background to both parts of Óláfr’s treatise. It is not then surprising that Óláfr uses runic letters in parallel with the Latin alphabet in the first section of TGT; it is also evident in his use of examples of native poetry to exemplify Donatus’s figures of speech in the second section. What is puzzling is that he makes such frequent reference to unattributed skaldic poetry. Known skalds, such as those cited by Snorri in Skáldskaparmál, would seem to serve the ideological purpose better.

Page 3

By comparing the distribution of verse types and attributions in related works, we can gain some information about how Óláfr approached and used his anonymous material. The various Norse-Icelandic treatises on poetry and poetics provide points of comparison on the use of verse to illustrate points of diction, metre, syntax and other grammatical features. The most important of these are Skáldskaparmál and Háttatal in Snorra Edda and the so-called Fourth Grammatical Treatise. The other grammatical treatises do not quote verse extensively, and the citations in Gylfaginning are of eddic verse and are meant to illustrate myths and cosmology rather than poetic or grammatical points.

In many ways TGT is comparable to Snorri’s Skáldskaparmál (Skm; ed. Faulkes 1998): it is an instructional work on poetic composition illustrated with verses by more or less well-known skalds from the ninth century to the twelfth century (or later, in the case of TGT). In their poetic sources, the works are comparable in that only a small proportion of stanzas are found elsewhere (Skm: 61/394 (15% not including þulur); TGT: 35/116 (30%)). Of these, the largest number are found in konungasögur (Skm: 39; TGT: 12) followed by verses found in Íslendingasögur (Skm: 14; TGT: 9, with verses from Egils saga being the best represented in each case). However, they differ considerably in the proportion of anonymous material: Skm has 59 anonymous stanzas (15%), of which Gróttasöngr accounts for almost half; only 16 stanzas in the whole work (4%) are of wholly unknown provenance (that is, anonymous and not attributed to a named poem). TGT has a far greater proportion of material of no known provenance, 51 stanzas and fragments (44%).

The so-called Fourth Grammatical Treatise (FoGT; also ed. in TGT 1884) provides a further point of comparison, although it was written later than TGT. Like TGT, FoGT is an adaptation of Latin grammatical material illustrated with native poetry. Of the 62 verses cited, 51 are anonymous (82%); 11 are found elsewhere (18%: 4 from the Y version of Laufás Edda, 2 from konungasögur and 1 from an Íslendingasaga). They are mostly dróttkvætt, but there are also 4 hrynhent, 2 runhent, 2 nýi háttr and 1 fornyrðislag stanzas. Björn M. Ólsen confidently attributes 33 stanzas to the unknown author of the treatise, and a further 12 with less confidence (TGT 1884, 239ff). He does not do the same, however, with the anonymous fragments in TGT.

In terms of distribution of metrical forms, the poetic material in TGT is comparable to Háttatal (Ht; ed. Faulkes 1999) in very broad terms: the majority are dróttkvætt verses, or sub-categories and variants thereof. The remainder represent a variety of metrical forms, including variants of dróttkvætt, kviðuháttr, hrynhent and tøglag and a number of stanzas in eddic metres. All but 3 of the 105 stanzas in Ht are the work of Snorri, although none of them are attributed to him in the text of Ht itself. Ht, it should be noted, is Óláfr’s most valuable identifiable poetic source, with seven fragmentary examples taken from it, all but one of which are attributed to Snorri.

There appears to be good evidence that the author of Ht and to a lesser extent FoGT composed the majority of verse citations in their treatises. However, this conclusion has not been extended by editors to the anonymous citations in TGT. It is quite possible that Óláfr composed at least some of the anonymous verse in his treatise under the influence of Ht. The provenance of these anonymous stanzas, however, has received very little attention.

Page 4

They differ from Snorri’s poetic material in Háttatal in that Snorri’s verses there are very consistent in their subject matter, whereas the material Óláfr uses is somewhat eclectic. However, the identification of Snorri as the author of the verses in Ht is not on internal grounds and there is no reason why we should find internal evidence of Óláfr’s authorship, if that was the case.

There is one stanza attributed to an Óláfr in TGT; Björn M. Ólsen identified this Óláfr with the author of the treatise (TGT 1884, 198), which has been accepted since (e.g. Skj BII, 110). The verse is as follows (TGT 1927, 60):

Stundum verðr þessi fígúra [amphibolia] svá, at eitt orð hefir fleiri merkingar, sem Óláfr kvað:
49. Kœnn njóti vel vænnar
vin minn konu sinnar –
víst erat dapr um drósir
drengr – ok eigi lengi.
Hér er óvíst, hvárt þessi sögn, eigi, er viðrorð neitiligt eða orð eiginligt.
Sometimes this figure [amphibologia] occurs when a word has more than one meaning, as Óláfr said: ‘May my wise friend enjoy his beautiful wife and have her [or and not] for a long time; the man certainly isn’t downcast concerning women.’ Here it is uncertain whether this word eigi is a negative adverb or an actual verb.

This stanza may provide an implicit argument for not attributing the anonymous stanzas to Óláfr, because, if this verse is attributed to the author of the treatise, then the verses which are not attributed to him are probably not his compositions

However, there are at least two reasons to doubt Björn M. Ólsen’s attribution: firstly, there is no reason why Óláfr would refer to himself in this way in his own treatise – Snorri does not do the same in Ht, for example, although the evidence for his authorship of the verses there is otherwise quite solid. Secondly, there is a fragment earlier in the treatise attributed to Óláfr Leggsson (a contemporary of Óláfr Þórðarson) which appears also to refer to a woman and which would make that Óláfr seem a much more obvious identification than the treatise’s author. No other poetry which treats such a private topic is attributed to Óláfr Þórðarson.

The evidence for dating the helmingr is equivocal. It was probably composed before the middle of the thirteenth century and therefore before the treatise, as kœnn (kænn) would have formed aðalhending with vænnar after that date, which would have been incorrect in an odd line. However, other stanzas in the treatise also have aðalhending in the odd lines. It is also clear from a discussion of diphthongs in ch. 4 (TGT 1927, 31; cf. Raschellà 2000) that Óláfr would have considered the vowels in the two words as distinct.


There are five unattributed fragments in the treatise which, in other sources, are attributed to known poets. It is likely that some of the anonymous stanzas were once the work of known skalds, but do not survive elsewhere. There is only one other anonymous fragment in TGT which is known from elsewhere, st. 47/1-2 of the eddic poem Grímnismál (TGT st. 82).

There is no easily identifiable organisation of any of the verses in TGT, let alone the anonymous ones. Finnur Jónsson in Skj makes the only attempt to group the anonymous fragments which are not attributed to a poem. Finnur considers all but one to be from the twelfth century.

Page 5

They are included under the headings: ‘Vers om ubestemmelige personer og begivenheder’ (Anon. twelfth century ‘Verses about indeterminable persons and events’; 34 sts.); ‘Andre religiøse vers og herhen hørende digtbrudstykker’ (Anon. twelfth century ‘Other religious verses and poetic fragments belonging here’; 4 sts.); ‘Vers om bestemte personer og begivenheder’ (Anon twelfth century ‘Verses about particular persons and events’; 2 sts.); ‘Vers, hentydende til sagn og lign.’ (Anon. twelfth century ‘Verses referring to legends, etc.’; 2 sts.); plus one verse ‘Om Tor’ (Anon tenth century ‘About Thor’) and one attributed to Einarr Skúlason’s *Øxarflokkr on apparently scant internal evidence: it refers to an axe (this fragment has been reclassified as anonymous for the new skaldic edition). Within the largest grouping, Finnur has made a rough attempt to group verses with the same metre and on related subjects by marking off each group with a line. No further description or justification is given. Finnur’s categories mean that any fragment with an identifiable reference is removed from a potential grouping with fragments with no identifiable points of reference – even when they share the same general subject matter and verse form.

There are a number of fragments in TGT which can be grouped according to subject matter and verse form (the number of the stanza in TGT 1927 is given in bold, followed by Finnur’s attribution in the anonymous twelfth century section of Skj; the editions of verses here and translations are my own):

• Ten dróttkvætt fragments about battle, one of which is identified:

C. 18
Herrr búask hvatt til snerru.
The army prepared themselves keenly for battle.
C. 10
Hér fregna nú hyggnir
hjörflaug brimis draugar.
The thoughtful trees of the sword [warriors] now experience spear-flight [battle].
C. 11
Svanr þyrr beint til benja
blóðs vindára róðri.
The swan of blood [raven] rushes by the rowing of wind-oars [wings > flight] straight to the wounds.
C. 16
Lofðungr gekk at Lakkar
– laut herr í gras – snerru.
The prince went to the onslaught of Hlökk [battle]; the army fell in the grass.
C. 23
Hár rauð hvassa geira –
hneig þjóð í gras – blóði.
The high one reddened the sharp spears in blood; people sank into the grass.
C. 22
Hringlestir gekk hraustan
herjum kunnr at gunni.
The ring-breaker [generous man], known to people [i.e. famous], went brave[ly] to battle.
C. 13
Rann hræddari hranna
hyrbrjótr frá styr ljótum.
The more afraid fire-breaker of the waves [gold > generous man] ran from the grim battle.
C. 14
Braut stökk bauga neytir
bleikr frá sverða leiki.
The enjoyer of rings [man] fled, pale, away from the play of swords [battle].
C. 17
Hermenn gátum hinnig
hugstinnan gram vinna.
We have found out that the single-minded lord there defeated the warriors.
B. 26
Ekl vasa ógn á Stiklar-
óblíð -stöðum síðan.
The not pleasant fighting was not lacking later at Stiklarstaðir.

All of these are dróttkvætt fragments describing a battle. All are couplets except for 23 (one line). 14 and 97 are quite similar; and 24 provides a contrasting motif with 27 and 95. 44 mentions Stiklarstaðir, the site of the battle in which Óláfr helgi died,

Page 6

to illustrate the figure tmesis, that is, the separation of parts of a compound. All six instances of the name ‘Stiklarstaðir’ in the skaldic corpus have tmesis (cf. LP, p. 537). It is therefore unlikely that Óláfr would have had to compose this couplet himself to illustrate this point. It is possible that some or all of the fragments belong to a poem about Stiklarstaðir, which Óláfr may have learnt during his time in Nidaros/Trondheim. If this is the case, it would support Gísli’s argument for the provenance of the verse in TGT.


• Five fragments of dróttkvætt verses which mention women:

C. 27
Hér liggja brot beggja,
brúðr, strykvinna súða.
Here lie pieces of both the painted ship-sides, woman.
C. 28
Því hefk heitit mey mætri,
mest nema hamlan bresti...
I have promised the excellent maiden that, unless the greatest obstacle fails...
C. 29
Vættik harms, nema hitta
höfuðgulls náim Fullu.
I expect sorrow, unless I manage to meet with the Fulla of head-gold [headband > woman].
C. 30
Band gaf oss með endum
Ilmr lýskála bála.
The Ilmr of the fire of the fish-hall [sea > gold > woman] gave me [lit. us] a band with ends.
C. 31
En skinnbjarta skortir –
skap kannask mér svanna –
dýrs hón hætt at hváru –
halmmein Njörun steina.
But the bright-skinned Njörun of stones [woman] lacks straw-harm [fire]; the temperament of the woman is known to me; she is in any case a dangerous animal.

Finnur likewise groups these fragments together in Skj. The first, however, addresses a woman, whereas the others refer to a woman. There is no internal or external information to determine the provenance of these fragments.

• Seven fragments from a kviðuháttr court poem about a Norwegian ruler. Five of these are grouped together by Finnur (Anon XII, C. 1-5), but there are only two other anonymous kviðuháttr fragments and both could belong here:

C. 1
Hlýð, hlýð, konungr,
hróðri þessum.
Listen, listen, king, to this praise poem.
C. 2
Ok stórhöggr
stillir Þrœnda...
And the people of Trøndelag’s hard-hitting leader...
C. 3
Í herská
hilmis ríki...
In the king’s war-torn country...
C. 4
Fór hvatráðr
hilmi at finna,
áðr siklingr
til sættar gekk.
The resourceful one went to meet the king, before the ruler undertook reconciliation.
C. 5
Konungr kappgjarn,
kostum betri
allri þjóð,
alframr konungr...
The energetic king, better in (good) qualities for all the people, the excellent king...
C. 20
Sterkum stilli
styrjar væni...
...for the strong ruler... expectation of battle...
C. 39
Þat hefk sagt,
es sjálfr vissak;
dulðr ferk hins,
es drengr þegir.
I have said that which I myself knew; I am ignorant of that, about which the man is silent.

Page 7

TGT verse 101 refers specifically to Þrœndir ‘people of Trøndelag’. Some of these at least are quite possibly a composition of Óláfr’s from his time at the court of King Hákon in Nidaros/Trondheim in the years c. 1237-40.

The remaining anonymous material comprises:

More information on the provenance or transmission of the anonymous verse can be gleaned from a number of fragments which are closely related to other known verses. These include the following stanza about Þórr:

…áðr djúphugaðr dræpi
dolga ramr með hamri
gegn á grœðis vagna
gagnsæll faðir Magna.
…before the deep-minded, mighty, reliable, victory-blessed father of Magni [=Þórr] struck the enemies of the sea of wagons [land > giants] with [his] hammer.

This helmingr is very similar to Þjóðólfr ór Hvini’s Haustlöng st. 6b, both in the opening words and in the kenning for giant:

…áðr djúphugaðr dræpi
dolg ballastan vallar
hirði-Týr meðal herða
herfangs ofan stöngu.
…before the deep-minded guarding-Týr of booty [=Loki] struck the very strong enemy of the field [giant] from above between the shoulders with a pole.

This type of giant-kenning, ‘enemy of the land’, is unique to these two stanzas. The close correspondence, but nevertheless variance between these two stanzas would suggest oral transmission. The verse is presumably quite old – Finnur dates it to the tenth century.

Also of interest is the following couplet:

Page 8


Sás af Íslandi
arði barði.
The one from Iceland ploughed [the sea] with the prow.

This is in tøglag (‘journey metre’), and it appears to conform to the type known in Ht and Háttalykill as ‘the Greenlandic metre’ (hinn grœnlenzku háttr). It is quite similar to the Södermanland inscription Sö65 (Djulefors), which contains a tøglag-like (hinn skammi háttr) helmingr:

han : austarla : arþi : barþi : auk : o : lakbarþi|lanti : [anlaþis
He ploughed with the prow in the east and died in the land of the Langobards.

It is unlikely that Óláfr knew this particular inscription, but the verse may have been conventional or had a certain currency, such as the runic verses about the ill-fated expedition to the east led by Ingvarr (e.g. Sö131, Sö173, Sö281, Sö320, Sö335, U439, U644, U654, U661, U778), which mostly follow a similar pattern. Sö65 (I, 50) also cites a couplet from a verse by Rögnvaldr jarl (Lv 31/3-4; Skj BI, 486: erjum úrgu barði, / út at Miklagarði) but the similarity seems coincidental. The inclusion of this verse appears to reflect the interest in runes at the Danish court at the start of the thirteenth century, which I have pointed out above. Given the correspondence of these two otherwise-unique fragments and Óláfr’s known interest in runes, it is quite possible that he drew on runic epigraphy in citing this stanza.

A less direct correspondence between the runic corpus and the verses in TGT is found in the following fragment:


Svanr þyrr beint til benja
blóðs vindára róðri.
The swan of blood [raven] rushes by the rowing of wind-oars [wings or feathers > flight].

This verse is said by Óláfr to contain a shortening of the syllable to correct the line length, namely, that vindára should be pronounced vindara ‘of the wind-eagle’. The word is ofljóst ‘punned’ according to Óláfr. There is no other parallel for this kenning in the corpus, apart from in an inscription from Lund (DR Tillæg 5; Lund bone-piece 4), the B-side of which reads:

arar × ara × æru × fiaþrar
the oars of the eagle are feathers

The pun here seems to be a play on ára ‘of the oars’ and ara ‘of the eagle’. There is also a visual play on the letter combination ararara. Once again, the connection between the anonymous stanza and the inscription is unclear, except that both share otherwise-unique features.

As tenuous as these connections between the verse in TGT and the runic corpus appear, they are nevertheless highly unusual. Only three other examples of close correspondence between the manuscript and runic corpora exist and none are from the East Norse area: Egill Skallagrímsson, Lausavísa 38/1-3 (Skj BI, 51) and N829; Haraldr harðráði, Lausavísa 7/1-3 (Skj BI, 329) and N B88; and Hallar-Steinn, Rekstefja 16/7 and N B57.


Two fragments should be mentioned here because they are omitted from Skj by Finnur.


Flugu hrafnar tveir
af Hnikars öxlum;
Huginn til hanga,
en á hræ Muninn.
Two ravens flew from Hnikar’s [Óðinn’s] shoulders; Huginn to the hanged and Muninn to the slain [lit. corpses].

Page 9

This fragment is presumably omitted because it looks very much like an Eddic poem; however it belongs to no known poem and other such verses are found in Skj.

There is also a fragment attributed to ‘Starkaðr (gamli W)’. A number of stanzas are also attributed to Starkaðr in Gautreks saga; these are considered inauthentic by Finnur; this however, does not explain the exclusion of the following from Skj, which at least has Óláfr’s authority for the attribution:

Þann hefi ek manna
menskra fundit
hrammastan at afli.
I have met the mightiest ring-distributor among humankind as regards strength.

This fragment is meant to illustrate the barbarism of adding aspiration (viðrlagning áblásningar): Hér er hrammastan sett fyrir rammastan at kveðandi haldiz í bálkarlagi (TGT 1927, 46) ‘Here hrammastan is placed instead of rammastan so that the alliteration is preserved in the bálkarlag metre’. The word rammr, however cannot be derived from an earlier *hrammr, which would make the alliteration an unusual back-formation. Björn M. Ólsen attempts to resolve this problem by normalising the couplet in light of its apparent Danish provenance, namely: ringreytanda / rammastan at afle (TGT 1884, 176; the sound change hr > r occurred in the ninth century in Denmark). Although this contradicts Óláfr, who would have had a good knowledge of Danish, it does provide an explanation for the unusual alliteration.

Sveinbjörn Egilsson (SnE 1880-7, 139) speculated that the subject of the verse is the Saxon champion Hama, whom Starkaðr fought according to Saxo (Book VI; Saxo, 156; Fisher 174). More likely, however, is Jón Sigurðsson’s suggestion in the same work that Gegathus (Geigaðr) is referred to here (SnE 1880-7, 294). Saxo states that Starkaðr mentioned Gegathus in a poem:

unde postmodum in quodam carmine non alias tristiorem sibi plagam incidisse perhibuit (Saxo 154)
Later in a song he [Starkaðr] told how he had never encountered, before or since, such a rigorous blow (Fisher, 173)

It is more than likely that Óláfr learnt this verse at the Danish court; it probably belongs to a tradition of poetry attributed to Starkaðr to which Saxo refers.


The fragmentary poems identified above support Gísli’s assertion that Óláfr would have gained much of his knowledge of skaldic poetry from the royal courts of Scandinavia: the fragments apparently in praise of a Norwegian ruler and those possibly about Stiklarstaðir would have most likely been learnt or composed in Trondheim and were likely to be the work of court poets, probably Icelanders.

However, some of the other anonymous fragments would quite likely have originated from non-Icelandic, and some cases East Norse sources. There is good reason to think that the Starkaðr fragment is Danish in origin; and the tøglag fragment (44) is modelled on a runic epigraph, the only surviving example of which is from Sweden. Óláfr’s sources therefore comprehended a broader base than Old Norse-Icelandic court poetry and its Icelandic domestic equivalents; it included East Norse sources and potentially runic ones. He also appears to have had oral sources beyond what Gísli discusses, namely, otherwise-unrecorded Eddic poetry (64) and pre-Christian skaldic poetry (85). While there is little evidence for runic sources beyond stanza 44 (the arði barði fragment), the use of runic sources would reflect both Óláfr’s stated project in the opening part of the second section of TGT, together with that of Saxo.

Page 10


DR: Jacobsen, Lis and Moltke, Erik. 1942. Danmarks Runeindskrifter. Ejnar Munksgaard, Copenhagen.
Faulkes, Anthony (ed.). 1982. Edda: Prologue and Gylfaginning. Viking Society, London.
Faulkes, Anthony (ed.). 1998. Edda: Skáldskaparmál. 2 vols. Viking Society, London.
Faulkes, Anthony (ed.). 1999. Edda: Háttatal. Viking Society, London.
Finnur Jónsson. 1920-4. Den oldnorske og oldislandske Literaturs Historie. Gad, Copenhagen.
Fisher: Davidson, Hilda Ellis (ed.) and Fisher, Peter (trans.). 1979. Saxo Grammaticus: The History of the Danes Books I-IX. D. S. Brewer, Cambridge.
Gísli Sigurðsson. 2000. ‘Óláfr Þórðarson hvítaskáld and oral poetry in the west of Iceland c. 1250: the evidence of references to poetry in The Third Grammatical Treatise’. In Clunies Ross, Margaret (ed.), Old Icelandic Literature and Society. Cambridge UP, Cambridge.
LP: Sveinbjörn Egilsson. 1931. Lexicon Poeticum Antiquæ Linguæ Septentrionalis. 2nd edition revised by Finnur Jónsson. Møller, Copenhagen.
Olrik, J. and Ræder, H. (eds). 1931. Saxonis Gesta Danorum. Levin & Munksgaard, Copenhagen.
N: Norges innskrifter med de yngre Runer; B: Bergen inscriptions courtesy of Runearkivet, Oslo.
Raschellà, Fabrizio. 2000. ‘Vowel change in thirteenth-century Icelandic: A first-hand witness’ in Michael Dallapiazza et. al. (eds), International Scandinavian and Medieval Studies in Memory of Gerd Wolfgang Weber. Edizioni Parnaso, Trieste.
Skj: Finnur Jónsson. 1912-15. Den norske-islandske Skjaldedigtning. Villadsen & Christensen, Copenhagen.
Sö: Brate, Erik and Wessén, Elias (eds). 1924-36. Södermanlands runinskrifter. Kungl. Vitterhets historie och antikvitetsakademien, Stockholm.
TGT 1884: Björn Magnússon Ólsen. 1884. Den tredje og fjærde grammatiske afhandling i Snorres Edda. STUAGNL 12. G. Knudtzon, Copenhagen.
TGT 1927: Finnur Jónsson. 1927. Óláfr Þórðarson: Málhljóða- og málskrúðsrit. Det Kongelige Dankse Videnskabernes Selskab Historisk-filologiske Meddelelser 13, 2. Bianco Luno, Copenhagen.
U: Wessén, Elias and Jansson, Sven B. F. (eds). 1940-53. Upplands runinskrifter. Kungl. Vitterhets historie och antikvitetsakademien, Stockholm.