The twenty-one stanzas that are thought to comprise Austrfararvísur ‘Verses on a Journey to the East’ (Sigv Austv) are customarily dated to c. 1019. They are all recorded in narratives of two eastward journeys in Snorri Sturluson’s Óláfs saga helga: in chs 53 and 75 of the Separate version (ÓH 1941, I, 134-6, 197-208) as well as in chs 71 and 91 of the Hkr version (ÓHHkr, ÍF 27, 92-4, 134-45). The first narrative (ch. 53/71) describes a visit by Sigvatr to Västergötland and to the Swedish royal court which would have taken place in 1018 or perhaps 1017 and which represented an attempt (ultimately unsuccessful) to make peace between King Óláfr Haraldsson of Norway and his namesake, King Óláfr Eiríksson of Sweden. The second describes a later visit to Skarar (Skara in Västergötland, south of Lake Vänern), which would be placed in 1019 or 1018. This was for the purpose, stated at the outset of ch. 75/91, of determining the sincerity of the good intentions of the powerful Rǫgnvaldr jarl Úlfsson towards Norway as tension between the two kings mounted. There are indications, however (e.g. in ch. 75/91), of a second purpose to the journey, pertaining to the projected marriage of Óláfr Haraldsson to Ingigerðr, daughter of the Swedish King Óláfr, of determining the strength of the opposition her father had come to harbour to the marriage. As discussed below, there is reason to think that Sigvatr was present only on the second of these journeys and that all the journey stanzas were occasioned by that.
In the version in Hkr, and in some mss of the Separate ÓH, shortly before the final stanza (st. 21), Snorri remarks (ch. 91, ÍF 27, 144): Ǫndurðan vetr fór Sigvatr skáld ok þeir þrír saman ór Borg ok austr um Markir ok svá til Gautlands ok fekk í þeiri ferð optliga illar viðrtekjur. Á einu kveldi kom hann til þriggja búanda, ok ráku hann allir út. Þá kvað Sigvatr skáld Austrfararvísur um ferð sína. ‘At the onset of winter, the poet Sigvatr went with two others from Borg (Sarpsborg) east through Markir (Marker) and so to Gautland (Västergötland), and on that journey he often got a cold reception. In one evening he approached three farmers, and they all turned him away. Then the poet Sigvatr composed Austrfararvísur about his journey’ (see Note to st. 21/1-4 on this passage). We thus have it on Snorri’s authority that at least some of the stanzas in this chapter belong to Austv. Identifying which of them do, however, is no simple task, since the poem is not a unified composition, like a drápa, but a ‘loose assemblage of verses’, as the term vísur suggests (Poole 1991, 6-7; Poole 2005c, 384; see also Kreutzer 1977, 88-9). The poem, also named Austrfararvísur in Fsk (ÍF 29, 179), is there called a flokkr, a skaldic poem without refrains (see ‘Reconstruction of skaldic poems’ in General Introduction for these terms.)
The style varies notably, ranging from stanzas in the manner of lausavísur, with present-tense verbs, deictic adverbs, a primary focus on the poet himself, and subjectivity in the form of comic irony and self-mockery, to those of high seriousness, combining panegyric with diplomacy. Schier (1965) is thus very likely right that the graphic descriptions and momentary impressions of some stanzas make it probable that at least some were composed by Sigvatr on his travels and later adjoined to the rest of the poem. Certainly the tone of some stanzas must have been intended to amuse Óláfr’s retainers, who are addressed in the first stanza (see Guðrún Nordal et al. 1992-2006, I, 218).
As a consequence of this diversity of style, and of Snorri’s vagueness about what actually constitutes Austv, there has been little scholarly agreement about which stanzas should be included, and in what order. The present edition retains the reconstruction adopted in Skj (and Skald), in which seventeen of the nineteen stanzas from ch. 75/91 appear as sts 1-8, 13-21 and four of the five stanzas from ch. 53/71 appear as sts 9-12. The first two stanzas from ch. 75/91 are classified as lausavísur (Sigv Lv 6 and 7 in the present edition) because Snorri tells us they were delivered before Sigvatr’s departure, although Finnur Jónsson (1932, 9) was himself uncertain whether they should be excluded, while Sahlgren (1927-8, I, 200-2, 205) argues for their inclusion, placing them last, in reverse order. Similarly, the first stanza from ch. 53/71 is unconnected with a journey and is classified as Lv 5.
Because Snorri assigns the four stanzas in ch. 53/71 (sts 9-12 in Skj and the present edition) to a different, earlier, journey, some scholars have excluded them from the poem. Ternström (1871), for example, in his edition regards the poem as comprising all and only the nineteen stanzas in ch. 75/91, in the order in which they appear there, thus including Lv 6-7 and excluding the stanzas from ch. 53/71, which he edits separately. And Schreiner (1927-9c, 1927-9d, supported by van Eeden 1943) has argued that Austv is in reality two poems, one (sts 9-12, 17-21) about the journey to Sweden through Eiðaskógr (Eidskogen forest), the other to Russia (see Note to st. 17/6 í gǫrðum, and cf. Sahlgren 1927-8, 203-4). Most scholars, however, have found it difficult to believe that the stanzas in chs 53/71 and 75/91 were not composed on the same journey, and indeed the singular farar, lit. ‘(of a) journey’, in the poem’s title supports this view. The four stanzas from ch. 53/71 were first included in Austv by Munch (1852-63, II, 548 n.), on the basis of his conclusion that the time frame for the events as described by Snorri is untenable, and Sigvatr could not have been a member of the first legation: see the summary by Ternström (1871, 38-9).
In addition to the different selection of stanzas included in the poem, there are different views about ordering; the following discussion uses the stanza numbers of the present edn. The stanzas comprising Austv in this edition and in Skj (and Skald) are reordered compared with ÓH, where they appear in the order 9-12, 2-8, 13-16, 1, 17-21, but several other arrangements have been proposed. Ternström, as noted above, retains the ÓH order. Sahlgren (1927-8, 204) would place Finnur Jónsson’s sts 6-8 before st. 4, and 9 and 10 after 1, adding Lv 7 and 6 at the end, as remarked above, while Toll (1924, 561-4), in addition to reordering most of Finnur’s stanzas, inserts Sigv ErfÓl 1 after st. 10, following the example of Weibull (1921). Bjarni Aðalbjarnarson (ÍF 27, xxxvi) recommends the order 1, 9-10, 2-8, 12-15, 11, 17, 20, 19, 18, 21, omitting 16. Stanzas 9 and 10 have been declared extraneous by Toll (1927-9, 508). (For a defence of the unity and order of the stanzas in Skj, see Patzig 1930b and Finnur Jónsson 1932.)
Ultimately, it is impossible to say with certainty how many stanzas should be assumed to comprise the poem, and how they should be ordered. This would be true even if there were no doubts about the accuracy of what Snorri has to say about the circumstances of composition. The uncertainties are aggravated, however, by the recognition that certain aspects of the stanzas are difficult to reconcile with what Snorri says. If the journey of 1019 was to Västergötland only, it makes little sense that Sigvatr should say that he made a trip east til Svíþjóðar ‘to the realm of the Swedes’ (st. 1/7), since Svíþjóð in Sigvatr’s day did not mean ‘Sweden’ in the modern sense but the area around Lake Mälaren, so that it did not include Götaland. Sigvatr’s reference to Rǫgnvaldr jarl as Óláfr Haraldsson’s most loyal friend á austrvega allt með grœnu salti ‘in the east all along the green brine’ (st. 21/7-8) also raises suspicions if, as Snorri says, he resided in Västergötland, since austrvega … salti appears to be an allusion to Eystrasalti ‘the Baltic Sea’ (as first argued by Guðbrandur Vigfússon, CPB II, 581). In addition, there are reasons to doubt that Sigvatr travelled the route indicated by Snorri, as detailed below. A further consideration is that in Fsk (ÍF 29, 179) it is maintained that Sigvatr served on a diplomatic mission to the Swedish king Óláfr concerning the marriage of one of his daughters to Óláfr Haraldsson, and about this journey he composed a flokkr which he called Austrfararvísur (see Notes to st. 21). It may thus be, as argued by Weibull (1921) and Sahlgren (1927-8, 206-7; cf. Beckman 1934), that Snorri, at a remove of more than two centuries from the events he describes, actually had little independent information about them, and he deduced much of the poem’s context (or all of it: Noreen 1922a, 69) from the poem itself – often incorrectly, as it appears so many saga writers did in such circumstances.
The possibility that Sigvatr did travel to Svíþjóð on the journey described in Austv suggests that there may have been just one journey, and thus that the stanzas in ÓH chs 75 and 53 and ÓHHkr chs 71 and 91 do indeed belong to a single poem. It might be objected that the journeys differ inasmuch as Sigvatr is on foot in ch. 75 (see st. 3), while he is on horseback in ch. 53 (see st. 11). But rather than disproving the unity of the stanzas, this discrepancy may in fact explain why Snorri assumed that there were two journeys rather than one, and thus why he put the stanzas into widely separated chapters of ÓH. And it is, after all, not implausible that Sigvatr’s company started on foot and acquired mounts when the opportunity arose, as maintained by Bjarni Aðalbjarnarson (ÍF 27, xxxi-xxxii). Although Snorri tells us explicitly that Sigvatr took part in the mission of 1018, it is only for the purpose of introducing the stanzas in ch. 53 of ÓH (and ÓHHkr ch. 71). Rather, if Sigvatr did take part, it is peculiar that no mention is made of any role he played in the negotiations either at Rǫgnvaldr jarl’s residence or at the royal court in Uppsala (see Finnur Jónsson 1932, 5-7).
Both the point of departure (Borg, i.e. Sarpsborg in Østfold, about 75 kilometres south-east of Oslo) and the time of year (the onset of winter) indicated by Snorri for Sigvatr’s journey have been contested, and as a result, the route of the travellers has been much disputed, as chronicled in the Notes below. Much speculation has been involved in discussion of the route, and it is thus tempting to prefer the simplicity of the account given in ÓH. But two circumstances in particular suggest that Snorri’s information may be unreliable, despite the fact that he had visited Gautland (Íslendinga saga, Stu 1906-11, I, 331). In st. 3 we are told that the men walked thirteen rastir through the forest from Eiðar, and in sts 8 and 14 the forest is specifically called Eiðaskógr. Although the precise modern equivalent of thirteen rastir cannot be determined, the distance is sufficient to indicate quite a sizable forest, one that cannot very plausibly be identified with any that Sigvatr is likely to have encountered on the route indicated by Snorri. This would rule out Edsveden in the area of Larv, which lies about 20 kilometres south-west of Skara; cf. Beckman (1923, 332; Beckman 1934, 214-15). There are many places in Norway and Sweden with the name Eid or Ed, but the association with Eiðar most naturally suggests that the forest referred to is Eidskogen, which lies along the present-day Norwegian border with Värmland, Sweden. This is the Eiðaskógr through which journeys are made in several sagas (see ÍF 27, xxxiii n. 2), including elsewhere in ÓH, and to this day it is an imposing feature of the Norwegian landscape. If this is the forest intended by Sigvatr, then his journey must have begun not in Sarpsborg but in some more northerly place. Thus, it has been proposed that the journey began in Oslo (Finnur Jónsson 1932, 17, 20; cf. Beckman 1934, 215-16), or in Sognefjorden, proceeding overland via Valdres, the party crossing the Glomma and thence through the Norwegian Eidskogen into Värmland; from there Sigvatr may have continued east towards the Baltic, or he may have proceeded down to Skara, either crossing Lake Vänern by boat or following its eastern shore (or western, Finnur Jónsson 1932, 20): see Sahlgren (1927-8, 180-6); Patzig (1930b, 90); von Friesen (1942, 224-6); Edqvist (1943); Turville-Petre (1976, 78). (Toll 1924, 553 would have the journey begin rather in Trondheimsfjorden.) Such a lengthy overland journey would probably have been better undertaken in the spring than in the autumn. The reference to sailing in spring in st. 10 has been taken as evidence (cf. Sahlgren 1927-8, 193-4), and the reference to autumn in st. 1/6 can be reconciled with this supposition; see the Note on that line. It should be noted that a journey through Eidskogen would also be part of the most direct route from Norway to Svíþjóð, if that is where Sigvatr was headed.
Passage through Eidskogen could also reasonably be thought of as a route to Skara in Västergötland, but not from Sarpsborg (though this is indeed the route advocated by Noreen 1922a; cf. Finnur Jónsson 1932, 14). And so if Sigvatr did depart from a more northerly place, the question arises whether Snorri’s information about the journey’s end is any more reliable than that about the beginning. There is no mention of Skarar (Skara) or of Gautland (Götaland) in the poem, only in Snorri’s prose, and in fact no contemporary record indicates that Rǫgnvaldr’s domain was Västergötland, as Snorri says. Accordingly, it has been proposed that the jarl lived in Svíþjóð itself (Weibull 1921, 118-20; a northerly point of departure would then be demanded by st. 13/7 norðan ‘from the north’) or in Russia (Sahlgren 1927-8, 173-80; Schreiner 1927-9c, 44-5; see the Notes to st. 17/6 í gǫrðum and st. 21; and cf. Finnur Jónsson 1932, 6 n. 2). Rǫgnvaldr is said to have lived in Russia after his kinswoman Ingigerðr Óláfsdóttir sœnska was married there to Jarizleifr Valdamarsson (Jaroslav, son of Vladimir; ÓH 1941, I, 211; ÍF 27, 147-8; ÍF 29, 26), though the nuptials should have occurred after Sigvatr’s journey to the east, according to Snorri’s chronology of events. Bjarni Aðalbjarnarson (ÍF 27, xxxiv) discounts both possibilities, arguing that wherever Rǫgnvaldr lived, he owed fealty to the king not of Sweden but of Denmark.
The details of Sigvatr’s route, how many journeys he undertook, what places he visited, and who exactly Rǫgnvaldr jarl was, are matters of immense interest, perhaps precisely because they cannot be resolved conclusively. And yet an accurate edition of the poem depends upon the answers to questions like these, since they influence decisions about what stanzas must be included and how they should be ordered. In the face of so many uncertainties, it has seemed best to retain the structure of the poem as conceived by Finnur Jónsson (Skj) and retained by Kock (Skald), not in the conviction that it is correct, but with the acknowledgement that the reasonable alternatives are too many and too inconclusive. Retaining this structure may also prove the course that occasions the least inconvenience to scholarship.
The chief witness to the poem is Snorri Sturluson’s Óláfs saga helga. One’s choice is between adopting the readings of the earlier, Separate version of the saga (ÓH) and those of the version incorporated into Hkr (ÓHHkr). Some (including Finnur Jónsson, Skj A) prefer the latter. In support of this preference it might be supposed that Snorri’s exemplar would have been accurate and that he might have improved ÓH when he compiled Hkr. Careful study of the two ms. groups, however, suggests that the Separate ÓH tradition is, on average, at least as reliable as, and sometimes superior to, the Hkr tradition, when particular attention is paid to the most conservative ÓH ms. (Holm2); and there are more medieval mss of the ÓH redaction surviving. The following mss are used in this edition: the A-class mss (in addition to Holm2) 325V, R686ˣ, 972ˣ, J2ˣ, 325VI, 75a, 73aˣ, 78aˣ; the B-class mss 61, 68; and the C-class mss Holm4, 75c, 325VII, Flat, Tóm. The poem in 761bˣ is copied from 61 and therefore not of independent value, but variants from known mss and rare conjectures added by Árni Magnússon are referred to occasionally in the Notes. Hkr is represented by Kˣ and by Bb, which is copied in the relevant chapters from a ms. akin to K (ÓH 1941, II, 1116). Stanzas 18/5-8 and 21/1-4 appear in Fsk combined as a single stanza; mss FskBˣ and FskAˣ are used for these. For this edition, Holm 2 is the base text, and Kˣ is the next most reliable ms.
Previous critical editions of Austv are relatively numerous. In addition to its appearance in the standard skaldic editions (Skj and Skald), in editions of Hkr, and in editions of this and other poems of Sigvatr by Ternström (1871), Konráð Gíslason (1892) and Jón Skaptason (1983), the poem’s colour and vivacity have made it popular with makers of anthologies, of which several are cited in the Notes.