--- In
cybalist@yahoogroups.com, "P&G" <G.and.P@...> wrote:
>
> >Nothing that I've read deals with the possibility of
> >analogical restoration of full grade forms as an explanation for the
> >triple representation of schwa in Greek.
>
> Try Szemerenyi "Introduction to Indo-European Linguistics", pages
128-129 in
> my edition, sections 6.4.7.2f in all editions.
>
> Szemerenyi is notoriously anti-layrngeal, but fair in his
presentation of
> data.
Okay, I reread that section. Szemerenyi does point out that many forms
can be explained as analogical forms based on the long grades. He also
points out that some can't. He also points out that there are
zero-grade forms in which Greek has a where e or o would be expected
by laryngeal theory. He still favors an analogical explanation, but
doesn't explain what model the analogy would be based on.
Both theories look to be in trouble. Perhaps, then, we should reject
the idea of a monogenic origin. Perhaps Greek represents an original
distribution of vowels and the rest of the languages merged them into
schwa (obviously accent would be the most likely candidate for the
causal factor).