--- In
cybalist@yahoogroups.com, "Thomas Olander" <olander@...>
wrote:
> As to the reduction of *-mas to -mus to -ms, note that Stang got
convinced of it after
> having read Kazlauskas (Stang 1975: 49). So did I.
Unfortunately, I didn't. Driven by a gut feeling (which turned out to
be correct) that your theory stems from the abovementioned
Kazlauskas' article, I re-read it last night in the hope something
important probably escaped me during the first reading (some years
ago), if only again to find his argumentation to be very weak.
Indeed, he suggests <u> in the D.pl. <-mus> of Old Lithuanian texts
to render a kind of labialized schwa (not [u] proper*) developed from
an unstressed *a after a nasal. He doesn't comment on the fact that
despite the dialectal and orthographical variety (or rather
orthographical chaos) of Old Lithuanian texts the sound in question
was uniformely rendered with <u> and never, say, with <o> (which
often rendered the short *lax* closed <o.> in the texts of Z^emaitian
stock), although an o-type orthogram seems to do better for a kind of
schwa (cf. Old English <eo> [e&]). It's also noteworthy that Dauks^a
usually renders the schwa derived from an *unstressed -a after a
labial* with an apostrophe (<dwiem' dienom'> < *'dviema die'noma),
but consistently writes <-mus> for the D.pl. desinence.
The only example of progressive labialization he adduces (Lith. dial.
[nuove:] < nùave:) is not quite to the point, since here we are
dealing with two labials (at least one of them being vocalic), and
the outcome is still [o], not an [u]-like sound.
His explanation why the unstressed *a of D.du. -ma (akima) wasn't
subject to the same reduction/labialization (allegedly because -ma <
PIE *-ma, while *-mas < PIE *-mos) is untenable.
--------------
* According to Bu:ga (_Rinktiniai ras^tai_, II, p.24), in the end of
the 19th c. -mus (not [-m&s] or the like) was still used by the older
generation in the East Auks^taitian dialect of Lé:nas (near Ukmerge:).
Sergei