From: Mate Kapovic
----- Original Message -----
From: "willemvermeer" <wrvermeer@...>
Sent: Tuesday, January 18, 2005 1:31 PM
Subject: [tied] Re: Various loose thoughts
> --- In firstname.lastname@example.org, "Thomas Olander" <olander@...>
>> I agree that the <u> of <-mus> probably was an [u], not a schwa.
>> That is also more or less Stang's conclusion (1975:49). The central
>> point to me is that OLith. -mus reflects PBSl. *-mas.
> Note that there is also a Slavic angle. Final mU can reflect *mus or
> *mum or *mom, but much less easily *mos or *mons or *moH(n)(s) or *muH
> (n)(s) or anything along those lines. Unless you want to separate
> Baltic from Slavic that limits the possibilities.
I don't see a problem. If we accept OLith. -mus as being a reflection of
BSl. *-mas, then we can just presume that BSl. *-as yields regularly -U in
BSl. This can only solve the ancient question of which is regular *nebhos >
nebo or *wlkwos > vlUkU. Scholars which posite BSl. *-mus believe that
*nebhos > nebo is the regular development and that *wlkwos > vlUkU is
irregular (via some analogy which can be quite simple). Others believe that
PIE *-os yielded OCS -7 regulary and that -o in s-stems is secondary (via
*nebho(d) or similar). I like the analogy theory better in o-stems than in
s-stems but if Slavic -mU in dat. pl. is to be derived from *-mos this would
be in accord with the other solution.