On Wed, 21 Apr 2004 01:06:30 +0200, Miguel Carrasquer
<
mcv@...> wrote:
>On Tue, 20 Apr 2004 21:36:32 +0000, elmeras2000
><jer@...> wrote:
>>I see that entirely differently. The prospects of reaching agreement
>>on this are slim.
>
>I fear so.
>
>> Even if I surrender, it cannot be used to support
>>Glen's theory, for it just shows /o/ before voice, not a case of non-
>>alternating /o/.
>
>I don't know if it supports Glen's theory or not, I haven't
>been paying much attention to what it is.
>
>If we have G. -osyo in the pronouns that have nom. -os, acc.
>-om, n. -od, then that should mean that the thematic vowel
>stood before a voiced segment. I think that voiced segment
>was simply the vowel *e of *-esyo, so *kWé-esyo becomes
>*kWó-esyo, and then after zero grade *kWó-syo.
I forgot: my impression was that the discussion was mainly
about the second vowel in the desinence *-esyo/*-osyo, about
which there can be no doubt that it was always *o.
I can think of a few other cases where we have final *-o in
PIE:
1) the middle endings *-to, *-nto/*-ro.
2) the demonstrative masc. nom. sg. *so
3) perhaps the Hittite allative ending -a, -(i)ya.
4) the dative of the sg. p.p. *meg^hyo, *twebhyo, *swebhyo
(Skt. máhya(m), túbhya(m); regularized (*-yo > *-oy) in
Latin mihi:, tibi:, sibi:, OCS mInê, tebê, sebê).
5) perhaps the pre-form of the dat.pl. *-bhios if analyzed
as *-bhi-o + *-s(W) [but more likely **-bhi-a-asW >
*-bhia:sW > *bhios].
As long as I don't understand the structure of those dative
forms of the personal pronouns, I prefer to explain the *-o
of *-(eo)syo in a way similar to the demonstrative *so, i.e.
as dissimilation of *<s...s>, as proposed by Jens. The
masc. nom. form of the genitive adjective *-(eo)syo-s became
generalized, and lost it's final *-s by dissimilation.
=======================
Miguel Carrasquer Vidal
mcv@...