On 2006-08-09 14:54, Piotr Gasiorowski wrote:
> The development of PIE *-o:n in Slavic is
> a difficult problem; the discussion of the details could fill a whole
> new thread.
One recurring question is whether the raising of *o: to *u: (> *y) is
caused by the following nasal (in which case *-y would be the regular
reflex of PIE *-o:n). As an alternative, a secondarily sigmatic nom.sg.
*-o:n-s is posited, in which case the proposed development is the same
as in the acc.pl. of thematic masculines. Both theories have their
adherents. I wouldn't lay my life down for either, but it seems to me
more economic to accept the raising of *-o: before word-final sonorants
(also *r, which would account e.g. for *c^etyre instead of +c^etvore as
influenced by the lost neuter *c^ety < *ket(w)u:(r) < *kWetwo:r) --
unless somebody comes up with convincing counterexamples.
Piotr