Re: IE Thematic Vowel Rule

--- In cybalist@yahoogroups.com, "tgpedersen" <tgpedersen@...>
wrote:

> >It may put things into perspective to remember that the
> > thematic vowel rule is still directly observable in Modern Greek
> and
> > in part also in some Slavic languages,
>
> On the subject, Russian has, in what looks to me like athematic
> verbs either
>
> berú
> berës^
> berët
> (-ë- = -yó-)
>
> or
>
> mogú
> móz^es^
> móz^et

Sure, and other Slavic languages have similar patterns reflecting
the e/o interchange. What are you trying to say by quoting these
forms?

>
> I got a couple of pages into an article by you on Balto-Slavic
> Stress. I gathered this much:
>
> PIE mobile stress stays.
> PIE end-stress becomes mobile.
>
> Have I got it right, and, if yes, how does it apply here?

It applies to the fate of nouns as explained by Illich-
Svitych. "Stays" however does not mean that it stays for long. Read
on, and you'll meet Dybo's law and Stang's law, and soon very many
accents are in new places.

> >and indirectly in the
> > distribution of umlaut in the Albanian verb. These pieces of
> evidence
> > still match each other quite well, so one would think their

common

> > source is of relatively recent date. Now, the source, whatever

it

> was,
> > is definitely older than the IE protolanguage, so its effects

have

> > been lingering on for at least five millennia. Then the
> > ultimate "real" source of the thematic vowel rule does not have

to

> be
> > something created just yesterday either.
> >
>
> All true. You didn't take a position on whether the thematic stem
> and the person and number endings were independent words recently?

I don't understand the question. What does "recently" mean? Any
analysis of the collocations of stem + flexive as consisting of two
distinct words each goes beyond the time frame relevant for the
disentanglement of the ablaut rules. The IE ablaut rules (e > o >
zero and all that) apply to forms in which the flexives are already
fully fused with the stems. This is especially relevant for the
strong forms which act as if there are no vowels in the final part,
while the flexives of the weak forms act as if they contain a vowel
each. Some have taken the extra step of assigning vowels to all of
them even at this stage, i.e. one vowel to the strong forms and two
to the weak ones. I find this excessive, and I find it next to
inconceivable that the decline of the vocalism does not pass through
a phase with an interplay of one and zero vowels. If the strong-form
flexives once had vowels also, as may seeem credible, there has been
an earlier reduction before the stage of, say, nom. *-z, gen. *-os,
was reached in the prehistory of PIE. It is in that case the
*result* of that process that has been subjected to the working of
the ablaut rules we can specify. It has been my experience that such
a phase lends itself very well indeed to the specification of
automatic phonological rules.

> Now if it's the way I believe it happened, then the thematic
> inflection is a static-stress-ification of an old semi-thematic
> inflection. That proposal, since it implies a modification of
> an "existing model", doesn't need thematic stem and person and
> number endings to have been independent words recently. It also
> explains why it was introduced in both verbs and nouns; the common
> purpose was stress-regularisation, not a morphological or semantic
> one.

In my opinion it did not happen the way you believe. The thematic
inflections are not semi-, but fully thematic. They do not always
have anything in common of a functional nature, what they all have
in common is the mere fact that the stem end in a vowel. That can be
explained as the effect of an old juncture phenomenon.

Jens