Íslenskt mál og almenn málfræði - 01.01.1985, Síða 106

104
Jörundur Hilmarsson
This proposal conceming the formation and relationship of Gmc.
*sunþra-, *sunþa- and *sunnö (for *sunþö) has the following
advantages:
a) the long assumed, but never quite satisfactorily explained, con-
nection between these words becomes clear;
b) the proposed word-formation follows a well-known Indo-Euro-
pean pattem;
c) an equivalent of Gmc. *sunnö (for *sunþö) is found in Toch. B
swáhco/swanciye, A swáhcem ‘(sun)beam, ray’ as will now be demon-
strated.9
It was indicated above that the sequence VHnC could be realized
either as VHnC or was restructured as VHnC in the individual Indo-
European languages. Schindler (1969:149 ftn. 39 (hesitantly)) and
Mayrhofer (1982:187) suggest that a development of the latter type
might be termed European (Schindler) or West-Indo-European (Mayr-
hofer). However, due to the effect of Osthoffs Law, at least as far as
Germanic is concemed, it cannot be determined which development
was preferred, as both would lead to the same result as shown above.
As for Tocharian, it seems certain that here the Indo-European se-
quence VHnC could be restructured as VHnC. Only thus can the vocal-
ism of Toch. B yente ‘wind’ (matching Toch. A want) be explained: this
word must derive from Pre- or Proto-Toch. *wénto- < I.-E.
*H2ueH ,nto-.
However, there are indices that Tocharian utilized both variants,
VHnC and VHnC, for the latter is necessary to explain the East Toch-
arian stem form wánt- (beside want-), securely attested in 21 a6, 384 b3
(inst.sg. wántyo), 191 a6 (wáhcyo) and 218 b2 (obl.pl. wántás), i.e. I.-E.
*H2ueH,nto-.> Pre- or Proto-Toch. *wento- > *wento- > *w‘ántœ- >
wánt-.
Van Windekens’ explanation (1976:544) of the dichotomy A wánt- :
want, B yente as reflecting an earlier apophonic «/-formation is difficult
to accept in view of the thematic flexion of the Tocharian words,
matching that of their Germanic, Latin and Indo-Iranian counterparts,
9 Already Pisani (1942-43:249) suggested that this Tocharian word might be related to
Gmc. *sunnð. However, he did not attempt a definition of that relationship.
Van Windekens’ explanation (1976:447) of this Tocharian word is not convincing (to
Skt. vamSá- ‘bamboo cane’).