But CVC is by no means the only (or even the clearly predominant)
phonotactic pattern in I-S/Dolgopolsky's Nostratic. Many of their
roots contain three or even four consonants. Since Glen has
invoked Tagalog, one could remark that the phonotactic structure of
Tagalog stems is very similar to what the "Moscow School"
Nostraticists propose for Proto-Nostratic. Apparently even 16
consonants and 5 vowels are not too few.
I-S may be dead, but his soul goes marching on. With so many
staunch disciples and the personality cult that surrounds him, I
wouldn't be afraid for Slava. As for I-S's notation, I have this vicious
impression that it may be too complicated even for insiders.
Dolgopolsky, in particular, seems to be confused by his own
diacritics and special symbols. In the Nostratic dictionary samples
published so far there are numerous notational inconsistencies.
As for the "split or merger" problem, there is also the lurking
possibility that the correspondences are spurious. This is
especially likely if both A : B and A : C are supported by just two
or three instantiations (in such cases it's rather straining the truth
to talk about regular correspondences). Correspondences cannot
be atomised with impunity, since very small correspondence sets
won't convince any critics. If we seem to require really massive
mergers (or systemic meltdowns, as R.L. Trask has called them) --
for example, when a dozen or so Nostratic stridents, most of them
tenuously instantiated, merge as IE *s -- I smell a rat.
Piotr
To:
cybalist@yahoogroups.com
From: Miguel Carrasquer Vidal <
mcv@...>
Date sent: Sat, 27 Jan 2001 06:39:33 +0100
Send reply to:
cybalist@yahoogroups.com
Subject: Re: [tied] IS's "regular roots"
... My point was simply that one cannot have a simple
phoneme inventory (say,
5 vowels and 20 consonants) and at the same time CVC
roots (20*5*20
makes 2,000 roots or less, which is clearly
insufficient). I also
objected to the dismissal of Illich-Svitych's work
simply because (a)
he's dead and (b) one doesn't understand his notation.
A problem which may be partially responsible for the
proliferation of
proto-phonemes in Nostratic (and other large time-depth
proposals) is
that when confronted with a set of correspondences like:
Lang 1 Lang 2
A B
A C,
the easiest reconstruction to make is *B, *C, with
merger to A in
Language 1. To reconstruct *A, with split into B and C
in Language 2,
one needs to know what the conditions of the split
were, and, when
dealing with large time-depths, such information is
more likely to
have been irretrievably lost. This makes *B *C not
only the easiest,
but in fact the only possible reconstruction, given the
information
that we have.
=======================
Miguel Carrasquer Vidal
mcv@...
><><><PIOTR><><><
[pyotr gonshorofski]
School of English
Adam Mickiewicz University
Poznan, Poland
><><><>BYE<><><><