--- In
cybalist@yahoogroups.com, "Rob" <magwich78@...> wrote:
> Again, what caused the distinction between unaccented *-tor-s and
> accented *-tér-s? It seems like the former could not have been
added
> during the stage when the zero-grade was created.
My understanding is that *-tors was the immediate and quite regular
product of unaccented *-ters. After the change of (the prestage of)
unaccented *-e- to (the prestage of) unaccented *-o-, the alternants
*-térs and *´-tors developed into *-té:r and *´-to:r respectively,
again in fully regular fashion. I assume that the lengthening
caused by the nominative sibilant preceded the reduction of short
unaccented vowels all the way to zero, since the lengthening saved
the reduced vowel from complete loss. I do not see how this could
be handled differently without unnecessary complications. In my
view the o-timbre points more to a parameter of tone than to one of
strength, but that does not exclude that accented and unaccented
vowels differed in both respects.
Jens