Re: [tied] Re: IE prefix "*s"

From: alex
Message: 28549
Date: 2003-12-16

Piotr Gasiorowski wrote:
> 16-12-03 18:53, alex wrote:
>
>> Since the discution took this path because of the supposed
>> simplification of "âi" to "i" , one has to underline that there is no
>> viable example which shows a such transformation, since - as I
>> mentioned in the begining - once being there, the group "âi" become
>> very strong and stable and does not simplify to anything more. Thus
>> the change assumed by Miguel *ex-cambio > *scâmba > *scâimba >
>> skimba appears to not be validated by known facts.
>
> Well, <inel> is a known fact. Romanian has phonological diphthongs
> only under stress or in final positions. Diphthongs that arose in
> _unstressed_ syllables were reduced to monophthongs a long time ago.

"inel" has the stress on "e": inél
The advantage here is that we do not work anymore with supposed data but
with recorded words.
In the XVI century the words are attested as "ânel" and "ânema"; from
the XVI century until nowadays there is no form recorded as "âinel" and
"âinema". That is the dissavantage of having recorded data; there is no
place anymore for suppositions.

Rosetti explains the change of "â" > "i" due influence of open timbre
from "e" respectively "i" in the next syllable.
For trying to make the knot with "schimba", one has to assume there has
been an word "*scâmbea" and under the influence of the "i" from the last
syllable, one could get the change "*scâmbea" > "*schimbea". That should
work I guess.
The reconstructed form "*scâmbea" has the advantage of explaining "â" >
"i" but the disavantage of giving head-aches for the last "ea" which has
no reason for being reduced to "a" in final position..

To be honest, my solution with *(s)kemb- doesn't work too very well.
Even if I assume the /e/ > /ã/ in a very early stage (for avoiding ske >
Ste), prior to any latin loans , the form which I get is the same:
*skâmb-.
Even assuming that the /e/ > /i/ before /mb/ is of no help since the
"ske" or "ski" became both palatalised and should have yelded "Ste" or
"Sti", thus "*Stimba" but not "skimba".

Hmmm... one has to see where the limits are and I guess regarding this
word I seen them by now.

Alex