[tied] Re: Slavic palatalistions: why /c^/, /c/?

From: tgpedersen
Message: 41908
Date: 2005-11-08

--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...> wrote:
>
> On Sat, 05 Nov 2005 09:57:08 +0000, tgpedersen
> <tgpedersen@...> wrote:
>
> >--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...>
wrote:
> >>
> >> On Wed, 02 Nov 2005 09:38:41 +0000, tgpedersen
> >> <tgpedersen@...> wrote:
> >>
> >> >--- In cybalist@yahoogroups.com, Miguel Carrasquer <mcv@...>
> >wrote:
> >> >>
> >> >> On Tue, 01 Nov 2005 13:55:13 +0000, tgpedersen
> >> >> <tgpedersen@...> wrote:
> >> >>
> >> >> >Here's a version with even fewer rules:
> >> >>
> >> >> But they're all wrong.
> >> >>
> >> >
> >> >Because...?
> >>
> >> Because the rules you gave were:
> >>
> >> >PIE > pp-PS
> >> >*ai, *oi > *e
> >>
> >> Obviously wrong. *ai gives /ê/, not /e/, and only after the
> >> first palatalzation.
> >
> >*i, *e, *j palatalise.
>
> What's that to do with the fact that *ai does *not* give
> /e/?

I'll have to ask with a question: is /ê/ documented or
reconstructed? If it's there only to be differentiated from original
PIE *e, it wouldn't be necessary if PIE *e > *je first.


> >The whole point of the exercise is to "tag"
> >the vowels *i and *e with a *j which to simplify the palalisation
> >condition (as ' / *j'),
>
> Except that this is demonstrably wrong.

Demonstrate it then.


> >which has the added affect of distinguishing
> >original *e (> *je) from later by-products (here *ê) such that
the
> >two Slavic palatalisations don't have to be done in any
particular
> >sequence. You would have noted that if you had taken time to
think
> >over the proposal, instead of rejecting it for not conforming to
the
> >traditional scheme.
>
> You mean not conforming to the facts.

If that's your perception of what you're doing, why don't you cite
the relevant facts then?


> >> >pp-PS > p-PS
> >> >*V(:) > *jV(:), /V front
> >>
> >> Wrong as I explained (*ji would have given /i/, not /I/, in
> >> OCS).
> >
> >No you didn't. You explained that *ji would have given /ji/. I
> >wonder what you will claim to have explained the next time? I
assume
> >it occurred to you in the meantime that the /j/ was consumed in
the
> >process of palatalising the preceding consonant?
>
> No it wasn't. Cf. ORuss. spellings like c^judo, etc.

What phonetic sense is there in that? How would you distinguish
between /c^/ and /c^j/? in short: how much phonetic reality is there
in this extra /j/?


> An easier way

Than which?

>to demonstrate that *e and *i never acquired a
> j-glide (except in the Anlaut) is the fact that for instance
> *te and *ti develop differently from *tj(V). Cf. the verbs
> metoN, metes^I (Pol. mioteN, mieciesz) vs. xUtjoN, xUtjes^I
> (Pol. chceN, chcesz).
>

The former might be regularization (-tje- > -te-). Isn't the latter
an Iranian loan?


Torsten