Re: THEORY: Allophones

Steg Belsky wrote:
>
> On Mon, 5 Apr 1999 20:53:14 -0300 FFlores <fflores@...> writes:
> >Is there any natlang or conlang where an allophone
> >of a certain phoneme is also an allophone (maybe the
> >only one) of *another* phoneme? Do you think that's
> >likely or probable?
>
> >--Pablo Flores
>
> In natlangs, Hebrew has some.  In most dialects, _bet_ without a dagesh
> is pronounced [v], the same as the letter _vav_.  Ashkenazic
> pronounciation has the dageshless _tav_ ("saf"), and the letters _samekh_
> and _ssin_ all as [s].  All Hebrew dialects today have [s] for both
> _samekh_ and _ssin_....it's unclear what the _ssin_ originally was.
> Israeli and Ashkenazic dialects have [x] for _hhet_ and dageshless _kaf_,
> and [k] for _quf_ and _kaf_ with a dagesh.
>
In Modern Hebrew, I don't think that [b] and [v], [p] and [f], [k] and
[x] are only one phoneme (respectively) anymore, even when they're
spelled with the same letter, at least not anymore than [f] and [v] are
still one phoneme in Modern English. I'm sure that Israelis think of
them as different sounds, and with the loss of dagesh hhazak it's not
even very predictable which will occur without going into a lot of very
esoteric historical morphology that has little application to the modern
language. Not only all that, but also there are a lot of foreign words
coming into the language which don't have those as one phoneme. As for
minimal pairs, I can think of [Sabat] "Saturday" and [Savat] "struck"
(striked?, in the meaning of French "a fait gre`ve") where they used to
be the same phoneme. If they're not already separate phonemes, they're
well on their way to becoming them.

Actually, this makes me think. It seems to me that this change is
greatly influenced by the fact that in practically all (if not all) the
languages that were spoken by Jews before they revived Hebrew, these
phones are separate phonemes, or they don't exist at all. Since they
distinguished between the sounds in their native language, they thought
of these phones as separate phonemes and didn't really "internalize"
that these phones are the same phoneme, and so they inevitably became
separate phonemes. I checked out an old book, "Historical Linguistics"
by Winfred P. Lehmann, which calls this substratum theory, and says that
it's mostly unproven. Perhaps this is some evidence for it. Then again,
there were no native speakers of Hebrew to emulate, so it might not be a
better example than that of Indian English. The book was last
copyrighted in 1973 (The American Cultural Center here in Jerusalem is
rather limited in books on linguistics), and I wonder if any research
has been done since to show that language change can be influenced by a
substratum even when there is a majority of native speakers. Anyone
know?

I'm not a linguist and I'm new to this list; sorry if I made some
grievous error in linguistic theory.

--
Adam Raizen

araizen@softhome.net