Torsten:
> What I was referring to was your idea (as I have
> understood it?) of having two schwa's, [...]
No, I don't think that at all. As of August 25 2005,
my thinking is that the available unaccented vowels
in IE just before Schwa Diffusion were *e, *a and
*&.
Unaccented *e [E] was in suffixes like 1pp durative
*-mes, plural *-es and 2pp durative *-te. It remains
*e without any paradigmatic alternation with *o.
(Keep in mind, both **-&s and **-as would have
yielded **-os because voiced *-s lengthened *& to
*&. followed by merger to *a. Vowel Shift then shifts
*a to *o. So there is only one choice for the above
mentioned suffixes.)
Unaccented *a at this stage of the language was
contained in roots like *aktwa: "eight" and other
stems caused by a-Epenthesis right after Syncope.
By way of Vowel Shift, *a can only become *o (without
any paradigmatic alternation with *e).
Unaccented *& is the only source I offer for thematic
*e/*o ablaut and the only thing that is needed. It
is the only vowel that responds to the presence or
absence of voice in the following segment... or rather
the only vowel that shows any **traceable**
anticipatory conditioning caused by voicing.
This appears to be most logically efficient solution
and no more vowels are necessary.
> not that of keeping *so and *to separate (which
> I want to do too).
We see a straightforward counterpart to IE *to- in
Uralic as well, so it's a done-deal that the two
demonstratives are seperate entities that Jens has
unfortunately chosen to lump together.
> But now I learn from Decsay that Proto-Finno-Ugric
> had 'vowel harmonic alternating pairs' (vel sim.!)
> *ke/*ko "who" *te/*to "that". Things are getting
> complicated!
Yes, but Finno-Ugric isn't Uralic. It's just plain
ol' vowel harmony. I believe your term is referring
simply to back/front differentiation of the *a~รค
sort. That's all. That's not necessarily the state
of affairs in Uralic, depending on the style of
reconstruction you choose. In fact, the non-initial
vowels, in my view, can be reduced to only two vowels
in pre-Uralic, without vowel harmony present at all.
= gLeN
__________________________________________________
Do You Yahoo!?
Tired of spam? Yahoo! Mail has the best spam protection around
http://mail.yahoo.com