> --- In cybalist@yahoogroups.com, "Sergejus Tarasovas"
> <S.Tarasovas@...> wrote:
>
>> But we still have acute, eg., in the infinitives: *píti, *býti,
>> *z^íti etc. Of course, *l is a sonorant and *t is not, so the
>> vocalization rules might depend on that. Is that what you mean?
>
> On a second thought, this is not a problem, if the circumflex is not
> a result of the contraction, but rather a metatony by Meillet's law
> (infinitive stands out of the paradigm, so Meillet's law doesn't
> operate on it).
The accentuation of *pil7, *pila, *pilo is definitely due to Meillet's
law, as is the accentuation of *byl7, *z^il7 etc. But the question is why
do we have the mobility here in BSl and not the fixed root stress? The
same goes for Slavic *pivo, *z^ir7 (this we have already mentioned),
*dar7, *z^iv7 etc. It is not very clear how did the Lithuanian paradigm 3
(mobile with the acute) come into being - if there is an acute in the root
why is there no Hirt's law (and thus acc. paradigm 1)?
Some of the examples like Lith. galva` (3) are explained like *golHweh2
where there is no Hirt's law because the *H is not immediately after the
vowel (ther is a *l in between), but how can you explain that PIE
*gWriHwéh2 is a. p. 1 in Lithuanian (a. p. a in Slavic) and that PIE
*gWih3wós is a. p. 3 in Lith. (a. p. c in Slavic)? Both are oxytona in
Vedic.
Mate