Íslenskt mál og almenn málfræði - 2020, Qupperneq 327

one reads Nú erumk auðskœf as Nu’rumk auðskœf with elision, shortening
and resolution in the first position (p. 231). There are parallels to this in
dróttkvætt poetry, e.g., Sigv Ber 3.3 Nú eru þegnar frið fegnir (> Nu’ru).
(2) In st. 18, where the first couplet is partly illegible, Þorgeir reconstructs it
as En Hróalds / *á hǫfuðbaðmi, where á is supposed to be the verb mean-
ing ‘owns, has’ (pp. 194, 235). The verb cannot be seen on the page in M
(ÍB 169 4to reads i, alternatively a), and its existence is highly unlikely for
syntactic reasons: the placement in-between the two constituents of the
subject (Hróalds hǫfuðbaðmi, i.e. Arinbjǫrn) in a dip at the beginning of
the even verse seems to be without parallels in the corpus of kviðuháttr
poetry.
(3) Another unlikely analysis from the point of view of skaldic syntax is
given for the first two verses of st. 11, which Þorgeir proposes should be
divided Arinbjǫrn er / oss einn um hóf (pp. 187, 225). This would put the
relative particle in a marked position (at the end of the odd verse) that
would normally require stress, which is not possible for a particle. This is
moreover unnecessary, since Arinbjǫrn would fill the verse alone (with a
short first position), and er oss einn um hóf could be analysed as a type-B
verse with resolved first position (pace Sigurður Nordal (ed.) 1933, p. 262).
Leaving Arinbjǫrn to fill the first verse alone adds rhetorical emphasis to
his name, which is mentioned for the first time at this place in the poem.
(A similar, rather creative proposal involving the placement of the relative
particle is made in the notes to st. 22 on pp. 242f.)
(4) Within Part I, the discussion of Óláfr Þórðarson’s use of Arbj st. 15 in
3GT (pp. 81ff.) is not entirely satisfactory, in that Þorgeir does not pay
enough attention to Óláfr’s notion of hljóðsgreinir (accents). In the context
of 3GT, Óláfr states that the words tvenn and þrenn, which are the normal
forms of these words in the contemporary Icelandic that Óláfr generally
takes as his point of departure, undergo an “aftekning stafs” ‹-nn› > ‹-n›
“fyrir fegrðar sakir, þvíat þá þykkir betr hljóða þessar samstǫfur í kviðu-
hætti, at þær hafi umbeygiliga hljóðs-grein heldr en hvassa”. The key to
understanding the actual form in which the words are used in the Arbj
stanza is the supposed change of hljóðsgrein, since umbeygilig hljóðsgrein in
3GT is attached to long vowels only (cf. Myrvoll and Skomedal 2010, pp.
80, 84). The aftekning stafs must thus result in the forms tvén and þrén,
which, additionally, may be given a plausible etymology (see Myrvoll and
Skomedal 2010, p. 85). Hence, this example of Óláfr cannot be drawn
into a discussion of the rhythmic peculiarities of the odd verses of kviðu -
háttr, as Þorgeir does (p. 83); tvenn (or þrenn) and tvén (or þrén) are met-
rically equal. In his reconstructed text, Þorgeir prints tvenn ok þrenn (p.
191), ignoring the evidence of 3GT altogether.
Comments from the second opponent 327