Edited here are the eight full stanzas in praise of Eiríkr jarl Hákonarson which are generally agreed to constitute the surviving fragments of Bandadrápa ‘Drápa of the gods’ (Edáð Banddr); the klofastef ‘split refrain’ is printed as st. 9. Eiríkr ruled Norway from c. 1000 to c. 1014 (see ‘Ruler biographies’ in Introduction to this volume), but the poem is concerned especially with his viking exploits during the reign of Óláfr Tryggvason (c. 995-c. 1000). The fullest attestation of the poem is accordingly in the saga of Óláfr Tryggvason in Hkr (ÓTHkr), with all eight stanzas in two clusters, while the expanded ÓT has three stanzas, Fsk one and SnE two fragments (see details below). The title of the poem, attested in Fsk, Hkr and ÓT, is no doubt prompted by the repetition of the thematically important word banda, gen. pl. of bǫnd (nom. pl.) ‘deities’ (lit. ‘bonds, fetters’; LH I, 555; see Note to st. 9/1). If so, it underlines the poem’s importance as one of the last overtly pagan eulogies. An alternative tentatively suggested by Guðbrandur Vigfússon (CVC: band II. 4) is that the reference is to the klofastef ‘split refrain’ which binds the stanzas together; conceivably there is a double meaning in the title.
In the source texts, all the stanzas are specifically ascribed to Eyjólfr and/or linked to the title Bandadrápa (which Hkr names in connection with six of its citations and implies in the remaining two). Their coverage of events, the second-person address to Eiríkr in sts 6 and 8, and the reference to his divinely-favoured rule in the stef (st. 9) place the composition of the poem at some point in his reign in Norway (Skj suggests c. 1010). The actions included, in the order found in Hkr (and more fragmentarily in ÓT), are: Eiríkr’s victory over Tíðenda-Skopti (‘News-Skopti’, sts 1-2); his residence in Denmark and assumption of some power in Norway (st. 3); his campaigning in Gotland (st. 4), the Baltic (st. 5) and Russia (st. 6); his capture of four Danish warships (off Estonia according to Hkr, st. 7); and his attack on Gautland (Götaland, st. 8) (cf. LH I, 554-5; de Vries 1964-7, I, 182-3). Fsk (ÍF 29, 164-5) contains a numbered list of battles fought by Eiríkr and allegedly described in Banddr (ÍF 29, 165); it differs slightly from Hkr and ÓT as to order and includes some additional events that do not figure in the extant stanzas (Fidjestøl 1982, 111-13). The list mentions the following actions: (1) against Skopti (corresponding to sts 1 and 2 in Hkr and this edition); (2) in Gautland (Götaland; no corresponding stanza at this point in Hkr, unless Gautland is an error for Gotland, in which case st. 4); (3) against the Jómsvíkingar (no corresponding stanza in Hkr); (4) against four Danish warships in Eyrarsund (Øresund) (Hkr st. 7); (5) at Stauri (uncertain location; Hkr st. 5); (6) in Garðaríki (Russia) including Aldeigjuborg (Staraya Ladoga) (Hkr st. 6); (7) in Eystrasalt (Baltic Sea) (no corresponding stanza in Hkr, unless st. 7); (8) in eastern Gautland (Götaland; Hkr st. 8); (9) against Óláfr Tryggvason (no corresponding stanza in Hkr); and (10) off Yggjustaðir (Ystad; no corresponding stanza in Hkr). There are reasons for distrusting some details of the narrative in Hkr (ÍF 26, 340, n.; cf. Ohlmarks 1958, 501-3), and st. 7 presents particular problems (see Note to st. 7/3). Fidjestøl (1982, 113-14, 171) therefore prefers the sequence of stanzas implied by Fsk: 1, 2, 3, 4, 7, 5, 6, 8, supporting this by inferences drawn from the distribution of the stef lines (see Note to st. 9 [All]). But it is likely that Fsk itself contains errors, notably the placement of action (2) in Gautland, not Gotland, and any inferences from the stef are unsafe, since we cannot know how many iterations of the stef the poem contained. The earlier critique of Hkr by Ohlmarks (loc. cit.) depends upon an unworkable interpretation of st. 3, which he places first in the poem (see Note to st. 3/1-4 and cf. CPB II, 51, 570), and an unsubstantiated reconstruction of the stefjubálkr ‘refrain section’. This edition therefore follows the stanza order in Hkr (with Skj), as the only one of which we have direct attestation, while acknowledging that it may well not be original.
The fragments do not include any trace of an upphaf ‘beginning, proemium’. Stanza 1, narrative in style but lacking elements from the stef ‘refrain’, probably belongs near the beginning of the main body of the poem, but its use of the adjective ungr ‘young’ (l. 3) without specified noun subject implies continuity from a previous, lost, stanza. Stanza 1 is tightly linked onwards to st. 2/8, where the first line of the stef, as extant, makes its first appearance. The other stanzas also evidently belong to the narrative section. Stanza 9 is strictly speaking not a stanza but rather the constituent lines of the klofastef ‘split refrain’, of which the first three lines (one or two only in mss U and C respectively) are brought together for the purposes of Skm (SnE 1998, I, 207) and the final two are added by editors. Perhaps for this reason it is omitted from Skald (see further Notes to st. 9).
The poem is characterised by complex kennings, often inverted, and an alternation of second- and third-person discourse (see, e.g., Note to st. 2/3). Verbal parallels link Banddr with the poetry of Eilífr Goðrúnarson (EilIII), Einarr skálaglamm (Eskál), Glúmr Geirason (Glúmr), Kormákr Ǫgmundarson (KormǪV) and Tindr Hallkelsson (Tindr; cf. de Vries 1964-7, I, 183).
The following mss are used in this edition: the Hkr mss Kˣ, F, J1ˣ, J2ˣ (sts 1-8) and 325VIII 1 (sts 1-3 only); the ÓT mss 61, 53, 54, Bb, Flat (sts 4-6); the Fsk mss FskBˣ, FskAˣ (st. 1); the SnE mss R, Tˣ, A (for sts 1/1-4 and 9/1-3), plus U (for sts 1/1-4 and 9/1), W (for st. 1/1-4), C (for st. 9/1-2). Stanza 1/1-4 in LaufE (1979, 395) is copied from W and therefore not used in this edition. The normalised text in this edition is based on Kˣ but with variants adopted from other witnesses where necessary. Editions of Banddr in the main skaldic compendia and in editions of the prose sources are cited for each stanza; the poem is also edited and translated, with notes, by Ohlmarks (1958) in a highly speculative treatment.