Re: [tied] The "Mother" Problem

--- In cybalist@yahoogroups.com, "Mate Kapovic" <mkapovic@...>
wrote:

> I agree. If Vedic and Germanic show no difference and Greek does -
> it is easiest to assume that Greek is archaic and that IIr. and
> Germanic have innovated. The analogy of *méh2te:r and *ph2té:r is
> easy to explain, but the difference in Greek isn't if we start
> from *meh2té:r. No obvious reason there why the accent should
> retract. Also, full grade in *meh2te:r, as well as analogy with
> *bhréh2te:r point to the root stress. And we also know that *-te:r
> stems did have static inflection (there are ample evidence of *´-
> trs in IE languages, although not in *meh2te:r contrasting
> *ph2te:r) as well as having proterodynamic one (*-trós) like in
> *ph2té:r or *dHugh2té:r. Thus, we can conclude that it's at least
> a little bit hasty to reconstruct PIE *meh2te:r with the end
> stress on the basis of Vedic ma:tá: and OHG muoter as is often
> done.

The problem with both *méxte:r and *bhréxte:r is that they violate
one or both of two apparent earlier phonological processes in IE:
1. Syllables with an unstressed later-Ablauting vowel (which I
reconstruct as */a/ based on its centrality) reduce it to zero when
phonotactics allow (= zero-grade).
2. Later, syllables with an atonic Ablauting vowel (that is, /a/)
become /o/ and those with a tonic vowel (/á/) become /é/.

Given what we know if IE phonological processes, then, if the
attested "mother" and "brother" words had been formed while one of
those processes was at work, we should see *mxté:r and *bhrxté:r or
*méxto:r and *bhréxto:r, respectively. Since we apparently do not,
there are two possibilities:
1. The forms were different earlier and were changed due to some
sort of analogical pressure, or
2. The forms were coined only after the second phonological process
was no longer operating.

- Rob