From: tgpedersen
Message: 45999
Date: 2006-09-07
> > Here's another take:Gothic 3pl sokjand, 3sg sokeiT shows the PIE alternation must have
> > The *-je/o- extension appears just before the endings. It is a
> > type of thematic ending.
> > Suppose (Schmalstieg) the original verbal (mi-)inflection was
> > the semithematic one, ie the one where those endings that have
> > -o- in the thematic inflection (before voiced) are there and
> > those that have -e- in the thematic (before unvoiced) aren't
> > there. Such a version of a verb with the *-jo-/je- suffix
> > would look like *-jo-/-i- instead. Suppose further that the
> > thematic inflection was produced from the semithematic one by
> > placing -e-'s before the unvoiced endings. We'd now have
> > *-jó-/-íe-, it being so late that the latter (corresponding
> > to your *-ih-) didn't go -> *-jé-, and therefore later could
> > go to Slavic *i, like *ei and *i: did.
> >
>
>
> I never got an answer to what *-íe- would produce in Slavic, so
> I made up one myself:
> Stang postulates (Slavonic Accentuation, p. 109) that the reason
> for the accentuation pattern in his a.p. b with neo-acute stressed
> vowel of the root in the forms outside of 1sg, which has end-stress,
> is that stress was retracted from the ending in those forms because
> their vowel was circumflex in Proto-Slavic. He doesn't give any
> reasons as to how this situation has come about.
>
> I think my Proto-Slavic *-íe- ended up as PSlav. *-i^-, with
> circumflex because it is a contraction of two vowels.
>
> Note that my proposal places *-íe- in those forms in which Stang
> claims the vowels had circumflex, except for 1pl, but that joined
> the e-forms early, and 3pl.
>
> That takes care of Stang's a.p. b, as far as I can see. Any other
> problems you guys want solved? ;-)