Bibliotheca Arnamagnæana - 01.06.1999, Blaðsíða 282

262
Part Two
second full stress (which always carries alliteration). On this basis Heus-
ler was able to reformulate Bugge’s rule in a very economic way: “The
ending of the full line is either stumpfor voll, never klingend”3 4 - “stumpf’
being defined as a bar consisting either of a monosyllabic word or of a bi-
syllabic word whose first syllable is short (— or ^ x), “klingend” as a bi-
syllabic bar with only one stress (— x), and “voll” as a bar with more than
one stressed syllable (cf. Heusler 1889: 105-109 = 1969: 700-703). The
line gestr at gest hædinn is thus according to Heusler’s version of the rule
a normal example of “full” verse ending.
In this form the rule is practically without exceptions - at least if it is
admitted that it does not apply to a series of consecutive full lines, as
Bugge held (cf. pp. 260-61 n. 1 above).
If Heusler’s two-bar theory is rejected, the number of exceptions to
the rule will rise (cf. Sievers 1879: 355-56), but still there can be no
doubt as to its general validity.
Previously we have referred to Sophus Bugge’s historically important
observation that if the endings of the Ijodahdttr full lines were restored in
their Proto-Nordic form, a great number of them would violate Bugge’s
rule. As examples he mentioned the words gest, horskr, renn, sæ, sæll, ty,
vifs, all from Håvamål, which in the language of the “Middle Iron Age”
would have been gasti, horskaR, rinnip, saiwi, saliR, tTwa, wTbas. Hence
Bugge concluded that the Eddie poems in Ijodahdttr must have been
composed after the syncope (Bugge 1879: 144-45, cf. p. 96 n. 45 above).
This statement was subjected to doser serutiny by Erik Noreen, who
in an article published in 1921 examined all Eddie Ijodahdttr full-line
endings with respect to their probable Proto-Nordic form and Bugge’s
rule (Noreen 1921b). Beginning with Skirnismal he grouped the 93 full
lines of this poem in three categories according to their endings:
1) Bisyllabic word with short first syllable (4- x), examples: afi, vita,
daga, truask, jptunnf in addition to some reconstructed forms like
séim (> sém), séa (> sja) - 55 occurrences in all.
3 “Der Versausgang ist entweder stumpf oder voll, nicht klingend” (Heusler 1889: 139 =
1969: 726). Conceming the terminology, cf. Heusler 1925-29, vol. 1: 145-48.
4 Jgtunn is to be reconstructed as Proto-Nordic *etunaR, which would give the metrical
form J xx, which is a form not found in Ijodahdttr (cf. Noreen 1921b: 7). This type is also
extremely rare in Old Norse poetry in general (cf. Kuhn 1939: 180-84 = 1969: 486-89).