On Wed, 17 Aug 2005 10:22:58 +0000, tgpedersen
<
tgpedersen@...> wrote:
>I read on in the article so I could understand Dybo's law and get to
>understand why some Russian verbs end-stress the 1sg and root-stress
>other persons and numbers.
That has nothing to do with Dybo's law.
Dybo's law is the shifting forward of the stress from a
non-acute syllable to the next one.
>This is what I found
>
>*vezó:
>*vezés^U
>*vezétU
>*vezémo
>*vezéte
>*vezó,tI
>
>->
>
>*vezó,
>*vezes^Ú
>*vezetÚ
>*vezemó
>*vezeté
>*vezo,tÍ
>
>etc
That is wrong. Vezti is a mobile verb, so Dybo's law does
not apply. An example of a verb where Dybo's law applies
would be an origially root-stressed verb, with a non-acute
root syllable, such as mogtí, mogóN < *mógoN. The other
persons are stressed on the root syllable (mòz^es^I,
mòz^etI, mòz^emU, mòz^ete, mògoNtI), because of Stang's law.
One can think that Dybo's law applied, but its effect was
undone by Stang's law (móz^es^I (Dybo)=> moz^és^I (Stang)=>
mòz^es^I), which seems to be the standard way of thinking
(though not mine).
The mobile verbs, such as <veztí>, should have had:
*vèzoN
*vèzes^I
*vèzetI
*vezemÚ
*vezeté
*vezóNtI (c.q. *vezoNtÍ, with stress polarization),
with the PIE distribution of mobile stress (extended to
thematic verbs by Pedersen's law). What is actually found
is:
vézoN
*vezes^Í > vezès^I
*vezetÍ > vezètI
*vezemÚ > vezèmU
vezeté
*vezoNtÍ > vezóNtI,
i.e. the 2/3 sg. also had end-stress. Undoubtedly, this is
due to polarization against the post-Stang's law b-paradigm:
jidóN
jídes^I
jídetI
jídemU
jídete
jídoNtI
>So this is where that alternating stress appears. All forms move the
>stress out to their final vowel or jer, but 1st sg. doesn't have
>one, so it can't.
The 2sg. also doesn't have one.
>But there is something I can't understand. Instead of moving stress,
>the 1st sg. lets its vowel grow a tail! It self-nasalises! Is this
>part of Dybo's plan?
No. As I said, the present paradigm of veztí has nothing to
do with Dybo's law.
>Because if 1st sg.'s vowel didn't self-nasalise there must have been
>a nasal ending which helped. And that must have been -mi or -mu,
>with a vowel
Obviously without a vowel: the ending is -oN.
The -N was added comparatively recently. It is absent from
PIE (*-o:) and Baltic (Lith. -ù < *-uó < *-o:), and it can't
have been present in Proto-Slavic, because *-o:N would have
given *-U (raising to u: before N gives *-u:N ; shortening
of long diphthongs gives *-uN ; denasalization of high
vowels gives *-u ; *-u gives -U). The -N must have been
added at least after the raising of back vowels before -N.
=======================
Miguel Carrasquer Vidal
mcv@...