Rit (Vísindafélag Íslendinga) - 01.06.1970, Page 82

8o
of avoiding a problem. Why, we must ask, do some dialects
have the same predisposition?
(4) Concerning West, East, and North Germanic (urnordi-
ska) in the first centuries a.d. one may emphasize that there
are several western-northern innovations (e.g. é> a), but there
are also some very old eastern-northern innovations (e.g.
ww > ggw, jj>ggj, ddj). The language that had both these
types of innovation may be called urnordiska. Of course we
cannot expect to find the peculiarities of this language in the
oldest Nordic runic inscriptions, which are few, short, and
very often unreadable.
(5) A few details:
The vowel ú has not been diphthongized in Vásterbotten
(contrary to what the map shows).
The vowel balance area was considerably larger than shown
on the map (as shown, for instance, by Erik Neuman).
The word kyrka[n) is seldom, if ever, pronouncedgyr(t)ga even
in the northern Swedish dialects. There are better examples,
such as batgen ‘the brook’.
The ‘younger M-umlaut’ in northern Sweden is not an um-
laut at all. It is an instance of tilljdmning, cf. the opposition
Vásterb. tölu (< talug ‘talkative’) vs. backu (< bakkug ‘hilly’).
The age of the Scandinavian grave accent is problematic.
The only thing we really know is that it must have existed
when the svarabhakti vowel developed in words like ko'mmer.
Hreinn Benediktsson: The ‘problem of beginnings’ is a difficult
one owing to the scarcity of direct sources. Haugen says (in
the Preþrints) that evidence, both linguistic and historical,
‘can be presented to suggest that the earliest runic inscriptions
embody a language without peculiarly Scandinavian features’;
further, ‘there are no forms in the early runes [viz., down to
about 500 a.d.] which could not also be ancestral to West
Germanic.’
It is true, of course, that in the earliest period we find,
for instance, the retention of w- (e.g. w[u)lþuþewaR, Tors-
bjærg), and that in the following period we still find, for
j