Re: [tied] jer / full vowel question

On 2007-01-29 23:42, Miguel Carrasquer Vidal wrote:

> On Mon, 29 Jan 2007 12:52:26 +0100, Piotr Gasiorowski

>> It seems that the front vowels are original and the back ones in the
>> instrumental are due to some kind of umlaut-like assimilation.
>
> I'm not so sure for the second person: cf. Skt. D túbhyam.

Well, this is an independent innovation. My guess is that <túbHya(m)>
got its /u/ from the nom. <t(u)vám>. Note Av. taibiia:, surely from *té-bHj-

> The pattern seems to be *téwe ~ *twé => *twé-bhy- (vs. *méme
> ~ *[m]mé => *[m]mé-g^hy- c.q. *méwe ~ *m[w]é =>
> *m[w]é-g^hy-), which gives Skt. táva ~ tvá- ~ túbhy- vs.
> máma ~ má- ~ máhy-, Slavic tébe ~ tob- vs. méne ~ *mon-,
> with different kinds of levellings, assimilations and
> dissimilations going on.

Yes, something along these lines. As for vowel reduction in pronouns,
there's also what looks like enclitic e-raising in Lat. mihi, tibi, sibi
< *mehei, *tebei, *sebei. The short form <mi:> is more likely a
contraction of <mihi(:)> than a reflex of the PIE enclitic dative,
otherwise one would also expect *<ti:> and *<si:>.

Piotr