Re: [tied] Re: septm [was: Caland [was -m (-n)?]]

On Wed, 26 Nov 2003 14:02:59 +0100 (MET), Jens Elmegaard Rasmussen
<jer@...> wrote:

>Hey, hey, Miguel and I are not all that identical: He is completely
>impossible if one wants to persuade him of a crazy idea, no matter how
>intensely one burns for it.

I'm not sure what you mean, Jens. I've been persuaded by a number of your
crazy ideas, even if I was unwilling to do so at first (nominative *-z, the
*-o- infix).


To return to the topic of Semitic *sab`atu(m) vs. PIE *septm.', I have done
some asking around elsewhere.

Yusuf B. Gursey on sci.lang informed me that in pre-Classical Arabic, the
nunated singular forms deleted the nunation in pausal position, so that
MuHammadun became MuHammadu:. The same for sab`atun -> sab`atu: (or
sab`ahu:, if the taa' marbuutah thing happened before the lengthening of
the final vowel) [The Classical and post-Classical pausal form is sab`ah,
with deletion of both the nunation and the vowel. In Modern
Spoken/Standard Arabic, sab`ah is the only form extant].

I think (but no-one has confirmed this so far) that it's conceivable that
the same thing was customary in Ethiopic. The difference between Ge'ez
forms with and without final -u (e.g. fem. sab`u and s&b`) could then also
be explained as pausal vs. non-pausal. That the lengthened final vowel
subsequently attracted the accent is, I would think, a secondary
development, perhaps developed only after Ge'ez had ceased to be a living
language. In any case, as in Arabic, the modern colloquials have deleted
the final vowel altogether (Amharic säb`att).

I also asked Bob Whiting (of the dept. of Assyriology at Helsinki
Univeristy). Here are excerpts from the correspondence (I forgot to ask if
it was alright to quote him on cybalist, but I don't suppose it's a
problem):

>[bw]:
>I'm afraid that I may not be much help. Although I take the
>relation between PIE *septm.' and the PS word for 'seven' for granted,
>I've never been able to reconcile to myself the difficulties of getting
>from one to the other. It seems fairly obvious that if the relationship
>is true then the PIE form must have come from PS because of the presence
>of /t/ and /m/ in the former. In Semitic (as you well know) these are
>not part of the root but are additional morphology. Much of the
>difficulty comes from the presence of the /t/ which is essentially a
>feminine marker, but because of the Semitic feature of polarity in
>numbers (from 3 to 10) is actually the masculine marker in this
>numeral. But in any case, it is a marked form rather that the unmarked
>form. Why the unmarked form *saba` or *sab`u(m) wasn't taken instead
>is just a mystery to me, but then again there aren't necessarily any
>rules for what gets taken as a loan form.
>
>[mcv]:
>>The PIE word is stressed on the final syllabic nasal, which is pretty
>>peculiar, and it was suggested that this may somehow reflect the Semitic
>>accentuation of the word.
>>
>[bw]:
>I would tend to think it more likely that it has to do with the /`/ and
>its loss in PIE pulling the accent to the end, but I have no
>justification for this. One might think of the basis of the loan being
>a form *sabi`tu(m) (noun with a real feminine marker which would have
>meant something like 'a group of seven', 'heptad'). Then one could
>posit something in PIE like loss of the /`/ with compensatory lengthening
>(> *sabi:tum) drawing the accent to the /i:/ then shortening and/or loss
>of the /i:/ (> sabtúm) with the accent remaining on the ultima. This
>would have had to have happened after the borrowing into PIE because
>it is hard to think of this happening in Semitic. It's pretty hard
>to think of the elision of a stressed long vowel in any case.
>
>[mcv]:
>>As far as I can see, the accentuation in PS would have been *sab`átu(m) or
>>*sáb`atu(m), and I don't think **sab`atú(m) is a possibility. Is that
>>correct?
>[bw]:
>I would find it very difficult to accept an accent falling on the case
>ending in PS.

If there's anything else, I'll keep you posted.

=======================
Miguel Carrasquer Vidal
mcv@...