Yes, I think this *basically* tallies with my reasoning, in so far as we
both invoke levelling. Kazlauskas adds some tonal points of which I do not
see the relevance. There are two things strange with the forms ran~koms,
ran~komis, ran~kose, z^iemomìs, z^iemosè (at this point I leave out the
dat.pl. in -óms because it has no vowel today and thus may confuse the
issue): 1. Saussure's Law ought to have shifted the accent from ran~k-
onto the following -ó- because the latter has acute accent (earlier
length). 2. The original form of the stem-forming -o- being IE *-aH2-,
this ought to have retracted the accent from the endings in persuance of
Hirt's Law which should prevent forms in -omìs, -osè. Neither rule is seen
to work, but both can easily have been overcome by analogical processes:
Immobile and mobile words are opposed to each other as still seen e.g. in
the gen.sg. ran~kos : z^iemo~s (it used to be more widespread, but
Saussure's Law has overgrown some of the evidence, most notably the
nom.sg. where *ran~ka: has become rankà and so no loger shows the neat
opposition to z^iemà it once must have had). Where the two sets *are*
opposed, it is by final initial accent in immobiles and final accent in
mobiles. That caused ran~komis to withstand change to -omìs (or to restore
the original type at a later point; Stang notes such restoration in the
illative ran~kosna as opposed to dialectal survivals of the reflex of
*rankósna, Slav.Acc. 63f), and it caused z^iemomìs to reintroduce
polarization of the mobility on the pattern of other stem-types like
-imìs, -umìs (and perhaps -ai~s).
The resulting tone of -óms (by reduction of -ómus, this perhaps and
perhaps not from older *-omùs) does not speak against retraction. If
retraction is old enough the newly accented vowel should keep its tone,
and if it is too young is should get circumflex, but that could easily
have been replaced analogically by the -ó- of the allied cases. Notably,
the change from -a~mus to -áms does not demand the parallel change from
-ómus to -oms to also undergo "metatony". The notation -áms just means
that the a is long and that, in the sequence -am- which is now
anteconsonantal, it is the a that has the tonal prominence. That would not
be expected to work any differently in -ómus > -óms, there being no reason
why the tonal prominence should suddenly shift to the /m/ which belonged
to the originally following syllable, and so there is no "-om~s".
I do not buy the tonal considerations and so cannot see from -óms that
it has retained a significantly old (how old is "senokas"?) place of the
accent. I would get from a presumed *-omùs via attested -ómus to -óms much
as one gets from vandenès via vandènes to present-day vanden~s, though I
do not know a good formulation of a rule that would enable us to predict
the syncope. At the end of the day -oms, -omis, -ose vs. -óms, -omìs, -osè
are as polarized as they can get, and that is what I take to be the main
factor in their history.
Jens
On Mon, 7 Oct 2002, Sergejus Tarasovas wrote:
> --- In cybalist@..., Jens Elmegaard Rasmussen <jer@...> wrote:
>
> > Oh, right, sorry. I should have starred it and checked
> attestations. The
> > accent type *is* with polarized mobility, so -ómus will be a stage
> between
> > *-omùs and -óms. (And, sure, it's Zink-).
> >
>
> J. Kazlauskas, while stating on p. 37 of his _LietuviuN kalbos
> istorine: gramatika_ that "Kirc^io alternacija kilnojamoje
> paradigmoje vyko ne tik tarp z^odz^io s^aknies (ar kamieno) ir
> paskutinio skiemens, bet kai kuriuose linksniuose kirtiN gale:jo
> ture:ti ir pries^paskutinis skiemuo, plg., pvz., dgs. naud. _galvóms_
> < _galvómus_", on p. 167 speculates, that "Aku:tine: priegaide: [in
> _galvóms_ etc. -- ST] rodo, jog dabartinis kirtis yra senokas...
> Tac^iau vargu ar ta aku:tine: priegaide: gali rodyti, jog
> daugiskaitos naudininkas is^ seno ture:jo pries^paskutinio skiemens
> kirtiN. ... Galima spe:ti, kad ... daugiskaitos naudininkas ...
> ture:jo galinio skiemens kirtiN. Daugiskaitos naudininke kirtis
> gale:jo bu:ti atitrauktas dar melodinio kirc^io egzistavimo metu, ir
> de:l to, pries^pskutiniam skiemeniui tampant kirc^iuotu, o
> paskutiniam netenkant kirc^io, pries^paskutiniame skiemenyje gale:jo
> susiformuoti auks^tas tonas, palyginti su buvusiu neauks^tu
> paskutinio skiemens tonu. ... Pats kirc^io atitraukimas is^
> paskutinio skiemens iN pries^paskutiniN, matyt, yra sukeltas
> tendencijos ture:ti koloniniN to paties paradigmos skiemens kirtiN,
> t. y. kirtiN to skiemens, kuriN kirc^iuoja daugumas paradigmos
> linksniuN". This is the kind of sophistry I can't understand
> completely, but since it (beg your pardon) somewhat resembles to me
> (mere impressionistically) your way of expressing yourself at least
> here on Cybalist, it would be very interesting to know your opinion
> on Kazlauskas' reasoning.
>
> Sergei
>
>
> Your use of Yahoo! Groups is subject to the Yahoo! Terms of Service.
>