Íslenskt mál og almenn málfræði - 2020, Blaðsíða 323

justification must be present before an unmetrical line is corrected to make it
metrical. I state as a principle that any proposed correction must conform to the
metrical rules” (p. 168).
In the reconstructed text, this conservative principle is for the most part car-
ried out, and combined with a better understanding of the orthography and lan-
guage of the text in M, as well as the metre of Arbj, Þorgeir thus provides read-
ings that are (a) metrically superior to the previous ones, and (b) more in line
with M or, in cases where the manuscript is today mostly illegible, the best ear-
liest transcripts of Arbj from M (ÍB 169 4to in particular). Good examples of
both practices are styr- in st. 4 (not stýrir as most editors have taken it; pp. 34, 180,
216), Þat allr herr in st. 17 (not heri as previously held; pp. 193, 233f.), and the
obviously correct verse order “*bragar fótum, / ‘[ı]b˘˘a[utt]’ stiginn” in st. 14 (pp.
190, 229f.), which has been altered — against the metre — by most editors. To
Þorgeir’s completely new reading of the second part of st. 12, i.e. “syni […]gð /
sonar Hálfdanar / á Játvarðs / áttar *skeiði” (pp. 188, 225f.), I will return at the
end of this opposition. His suggestion that the duplication of hilmi at in the first
stanza of Arbj is intended as a kind of stammering (p. 211), revealing Egill’s anx-
iety, is also persuasive.
Metrics
In Part II, “The Poetic Form”, Þorgeir Sigurðsson gives a thorough presentation
of the metre of Arbj, kviðuháttr (ch. 5), and an examination of how the metre is
carried out in both even verses (ch. 6) and odd verses (ch. 7), which have a diffe -
rent number of metrical positions and also slightly different rhythmic patterns.
In this part of the thesis, Þorgeir brings many insights into the metre, particularly
on certain restrictions on the unstressed positions within both odd and even vers-
es. He demonstrates that in the oldest kviðuháttr poems up to c. 1000 the dips of
trochaic verses are never filled with other kinds of syllables than inflectional end-
ings, the particle of/um or, in rare instances, the conjunction ok and the infinitive
marker at. In later poems such as Nóregs konungatal (late twelfth c.), the practice
is quite different, as it allows for both prepositions and other function words in
this position (see particularly pp. 124–29, 151–55). This finding is important
both because it improves our understanding of how the metre functions, and
because it introduces a new dating criterion for kviðuháttr poetry, which may be
used, along with other formal criteria, to weaken the claim for inauthenticity in
the case of Ynglingatal (Krag 1991). As Þorgeir says, “[t]he discovery of this in -
terest ing phenomenon seems long overdue” (p. 144), and it is to his credit that he
was the one to discover it. Another important observation is the fact that the odd
verses in kviðuháttr, which consist of only three metrical positions, have a metri-
cal mapping different from even verses (and most other Old Norse metres as
well), in that a strong position (lift) can have a light structure (short syllable). It
Comments from the second opponent 323