From: Miguel Carrasquer
Message: 27285
Date: 2003-11-16
On Sun, 16 Nov 2003 17:20:39 +0000, elmeras2000 <jer@...> wrote:
Some more thoughts...
>This makes the Arabic accent as useful for comparative purposes as
>that of Latin.
It's not to be excluded that the Arabic rules reflect the Proto-Semitic
ones without much change. That makes the Arabic accent still as useful for
comparative purposes as that of Latin... but as it is in the context of
Romanistics.
>For now, I have not yet seen anything to make me discard the thought
>that a special kind of Semitic accent in a form with mimation could
>be reflected in *septm.ยด . Perhaps it should be remembered that it
>does not have to be pan-Semitic, it just has to be some existing
>language old enough to serve as the donor of a loanword entering
>some prestage of Proto-Indo-European.
The only attested Semitic language with pervasive mimation (in all but the
status constructus) in the singular (and therefore, in the word sab`at-) is
Akkadian.
North West Semitic (already Ugaritic) has no mimation/nunation at all in
the singular. Arabic only has nunation in the indeterminate state of the
singular (-u in the construct and determinate). Some varieties of South
Arabic apparently generalized a pattern -m (indeterminate) -n
(determinate), also in the singular, although there is no trace of that in
Ge'ez, as far as I know.
So the most likely source (also for geographical and historical reasons)
would be some East Semitic (Akkadian/Assyro-Babylonian-like) language.
North-West Semitic, at least from 1500 BC onwards, and Arabic, are
excluded. South Semitic might perhaps just be possible as far as mimation
goes, but it's of course highly unlikely that the PIE'ans got their word
for "7" from Yemen.
=======================
Miguel Carrasquer Vidal
mcv@...