Rit (Vísindafélag Íslendinga) - 01.06.1970, Page 316

3X4
the gap between the assumed Proto-Indo-European formant
/-én/ and the Old Norse reflex, while at the same time taking
into account the runic evidence, it was necessary to posit for
the F-rune a phonological value /æ/ (with or without nasaliza-
tion) in these forms. Thus, as recently as 1968, Krause main-
tains that the F-rune ‘besaB zur Zeit der áltesten Inschriften
mindestens 3 phonologische Werte: 1. kurzes orales a unter
den verschiedensten Akzentverháltnissen . . . sowie in Diph-
thongen ... 2. langes œ, und zwar a) als Zwischenstufe zwi-
schen urgerm. -én und altnord. -e, -i, vor allem im Nom. Sg.
der maskulinen n-Stámme... b) möglicherweise als Zwischen-
stufe zwischen urgerm. é1 und altnord. á... 3. nasales q. vor s’
(Krause 1968:120).
The assumption that the F-rune could represent /æ/ is vul-
nerable to attack on several grounds. First of all, Nils Lid
(1952) found evidence in Lappish loanwords of the reflex of
an o-vowel in the nominative singular of masculine n-stems
and came to the conclusion that the Old Norse ending cannot
represent a regular development from the attested runic end-
ing, but rather must have been introduced from the io«-stems,
a conclusion in which Harry Andersen (1960) concurred. If
the Old Norse forms do not reflect the regular development
of the original an-stem formant, then, of course, there is no
longer any basis for assuming that the F-rune represented /æ/ in
these forms. Lid and Andersen assume a development from
PIE /-ön/, which supposedly became runic /a/, represented
by F, which in turn yielded 0 in Old Norse. This endingless
form was then replaced by the ending of the m«-stems, ON
-i, -e. While I agree that the Old Norse ending must come
from the tare-stems,2 it must be pointed out that it is impossible
2The arguments set forth by Makaev (1963:200-202) against Lid’s interpret-
ation are not convincing, principally because they place too much emphasis
on a parallelism between the n- and the r-stems. The development of the r-stems
themselves, however, is problematical. The form swestar (Opedal K76), like
Gothic -ar, may not actually be a reflex of PIE /-ér/, but can just as well be a
leveling from the oblique cases with PIE /-or/.