Menfergir bar margar
margspakr — Niðar varga
lundr vann sókn á sandi —
sandmens í bý randir,
áðr fyr eljunfróðum
allr herr Skota þverri
lǫgðis seið af láði
lœbrautar varð flœja.
Margspakr menfergir bar margar randir í bý sandmens; lundr varga Niðar vann sókn á sandi, áðr allr herr varð flœja seið lǫgðis af láði lœbrautar fyr eljunfróðum þverri Skota.
The very wise ring-destroyer [GENEROUS MAN = Haraldr] bore many shields into the settlement by the shore-ring [SEA]; the tree of the wolves of Nidelven <river> [SHIPS > SEAFARER = Haraldr] made an attack upon the shore, before all the host had to flee from the incantation of the sword [BATTLE] out of the land of the pollack-path [SEA > ISLAND] before the mettle-wise destroyer of Scots [= Haraldr].
[8] lœbrautar ‘of the pollack-path [SEA]’: The first element of the cpd has been subject to various interpretations: (a) Lœ- as from lœr, lýr ‘pollack’; lœbraut ‘pollack-path’ is then a sea-kenning which combines with láð ‘land’ to form an island-kenning (Nj 1875-8, II, 386-7; ÓT 1892, 350; ÍF 26; von See 1977b, 68; Hofmann 1978-9, 70; Hkr 1991 and this edn). The word form lœ- seems to be unique. Elsewhere the word appears as lýr, often as a determinant in sea-kennings, as in lýbraut ‘pollack-path’ in Þuríðr Lv 1/5V (Heið 4), an exact parallel to the present kenning; see also LP: lýr. The form lœ- cannot be explained easily. It could be regarded as poetic licence and an attempt to adapt the word to the inf. flœja ‘flee’. Alternatively, lýbraut and flýja might originally have formed the aðalhending, then, since both flýja and flœja existed as derivatives of Gmc fliuhan (ANG §§98.2, 230.2, 488 Anm. 2, 513 Anm. 3) flœja could have replaced flýja, which in turn would have caused the change from lý- to lœ- in order to preserve the aðalhending. (b) Læ ‘harm, deceit, woe’. As læbraut ‘path of harm’, this cpd has been arranged into various kennings with lǫgðis (s)eiðs (see Meissner 74; Holtsmark 1927, 9-11; Fidjestøl 1982, 79-80). The complexity of these kennings makes each of them doubtful, however, and more importantly læ- could not form an exact aðalhending with flœja in the C10th.