Sino-Tibetan (was: Re: Arabic and IE)

[Alexis Manaster-Ramer]

----------------------------Original message----------------------------


On Thu, 4 Feb 1999, Scott DeLancey wrote:

> ----------------------------Original message----------------------------
> On Mon, 1 Feb 1999, Alexis Manaster-Ramer wrote:
>
> > > Is Sino-Tibetan controversial  (Roy A. Miller fights it ruthlessly, but he
> > > seems to be the only one) ?
> >
> > I seem to recall Sagart attacking ST as well. Certainly the
> > state of ST lx is not satisfactory.
>
> Hmmm ... "not satisfactory"?  You could say that about any language
> family, I suppose.  I won't be satisfied with Indo-European until
> there's some consensus on the subgrouping of the major branches,
> for example.  And the state of Altaic linguistics has been discussed
> often enough here, and elsewhere ...

I was afraid of that. Just as Hockett got yelled at twenty
years ago or so for pointing out scandalous state of Athapaskan
or Nadene linguistics (I think I have this right).

As I point out in my Part 2 (to appear here soon), all
too briefly, I agree re IE (but not because of the branching
so much as because of the all-too-great disregard for
the regularity of sound laws, morphological analysis,
and semantic responsibility even there).  Altaic the
same story but more so, except that I do not see how
anyone can deny that the Altaic languages are related,
which is distinct from questioning the validity of the
current reconstructions (which leave a lot to be desired
obviously than Altaic does).

But what I meant was precisely that just as people who
doubt Altaic altogther seem to be misled by this
very elementary cnfusion and say in effect that
if the current reconstruction is so bad (this of
course was true 20 or 30 years ago much more than now)
than the whole family is in doubt, so too I think
that the few critics of ST are really reaction to the
unsatisfactory state of the actual work on ST (and
again just as in Altaic to some extent I suspect that
Miller is reacting to ST as it was some decades ago
esp. to Benedict's work much as the anti-Altaicists
are still really responding to the errors of Ramstedt
in the 50s and Poppe in the 60s).

I myself to the extent that I have a right to an
opinion (which being a mere mortal who can barely
read a little Chinese and has no Tibetan or any
other ST language at all, though I did once know
a bit of whatever the language of Mizoram is called
(Mizo is it) but have now forgotten it down to the
last morpheme is a very small extent) do not qustion
the validity of ST, and have repeatedly urged Sagart
(and just the other day Miller) to reconsider.  But
does anybody listen?

I certainly support 100% Baxter's recent elegant
demonstration of (a) how obviously ST lgs must
be related and (b) how to do probabilistic testing
of language relatedness if you are gonna do it at all
--though I do not in general
believe the latter is (so far anyway) at all necessary
or (more to the point) useful.

But Benedict's methods and those of other ST scholars
have not been free of problems.
>
> But if you mean to imply that there is serious room for doubt about the
> genetic unity of Sino-Tibetan, I don't think that's the case.
I dont mean to imply that. I agree with you, as far as I am
entitled to say anything about ST.

>  The
> older opposition to the idea stems primarily from a gross misunderstanding
> of the relevance of typology.  I can't imagine that on this list we
> have to go very far into the argument that Chinese and Tibeto-Burman
> can't be related because they are so typologically dissimilar, or
> the converse (often held by exactly the same people) that Chinese
> must be related to Tai because they are so typologically congruent.

I agree about the early history of the subject (I have
written something on this general problem in East/SE
Asian languages though I talked by the Maspero brothers'
celebrated obsessions with separating "tonal" from "atonal"
languages (as in Austroasiatic) rather than with ST),
but I do not agree about this list.  Sure, in theory,
everybody knows that typology has no place in lg classification.
But I dont think this is so in practice necessarily. I would
think that some of the opposition to Nostratic also has
in part typological roots.

>
> More legitimately, I've heard the argument made (by Chris Beckwith,
> I think, among others, though he shouldn't be held hostage to my
> imperfect memory) that ST can't be considered *proven* because the
> lack of morphology in Chinese makes it impossible to find the kind
> of nice syntagmatic and paradigmatic morphological correspondences
> that make us so confident of Indo-European or Semitic or Algic.

Thank you very much for bringing this up. I have both in
print and here excoriated Godard and others for this false
methodological "principle" which seems to go to a paper in
which Meillet got carried away in his fight with Kroeber
and asserted it but later even Meillet saw he was wrong
and specifically said that it is the languages of E Asia
that show that the principle is a false one--although he
underestimated the role of morphology in E Asia.

> This of course isn't really an argument *against* ST unity, only
> a healthy cautionary note.

Not even that. It is simply a false principle.  Of course,
more evidence is always better than less and morphology
usually makes for better evidence but that is a matter of
detail not of principle.  We have know this ever since
Rask and it is one of the things we really do know
and which will not change.

> But let us remember that at least
> certain morphological *processes* can be reconstructed for pre-Old
> Chinese which are strikingly parallel to attested TB morphology,
> most strikingly an *-s suffix with a range of functions, especially
> derivation of nouns.

Totally.  That's just what I was alluding to.  But it would
be nice if thre really WERE no morphology in Pre-Old chinese
to compare to TB languages, for the sake of the theoretical
points, But Comecrudan will make the same point.

[snip]

>
> As for Sagart, he is indeed convinced that the Chinese-TB link is
> a chimaera, but as far as I know he is the only working Sino-Tibetan
> linguist who takes that view, and I cannot for the life of me see
> what his argument is.  (And I'm far from alone in that).

I have argued with Sagart too, many times. But I would like
to say something sociological here. I have attended ST meetings
but I know Sagart as well as some leading ST people (I knew
Benedict too) and I have seen Sagart argue with e.g. Baxter
at other meetings and I have seen the literature where the
debate about Chinese-TB vs. Chinese-Austronesian goes on
and I am impressed by the professional and scholarly tone
of teh debate, no matter how ANNOYED most of you must
be with Sagart and how FRUSTRATED he must be so alone.

This is such a sharp and refreshing contrast to the tone
of similar debates elsewhere in this field that I think
it important to call attention to.  No one as far as I
can see calls for "shouting down" Sagart, as campbell
famously did Greenberg, no one villifies and libels
the other side, above all no one tries to suppress
and censor the discussion of opposing viewpoints
as has so long been the case with Nostratic where as
I said even to this day most journals will not admit
that Illich-Svitych ever existed.

> On the
> one hand, he has identified some significant Austronesian elements
> in the Chinese vocabulary, but this is hardly surprising (or new;
> Tsu-lin Mei and Jerry Norman pointed some of that out 30 years ago).
> And he would like to build a case for a special genetic relationship
> between Chinese and Austronesian, which would indeed require splitting
> Chinese off from TB.  But I have to say that the few arguments I've
> seen of his *against* ST are not impressive, to put it diplomatically.
>
I think they are incorrect too but he has in my opinion
done an impressive job with so little and with a theory
seems to me hopeless.  He is a certainly an impressive
scholar all around, although I think that his most important
contribution by far, and which will one day revolutionize
the study of ethnology and anthropology and indeed socail
sciences generally (and bring them to the level of linguistics)
is his study with Immanuel Todd on the prehistory of certain
types of family structure in Eurasia.  It is not linguistics
but it uses methods first developed by linguists and opens
up the possibility of an anthropological science no less
successful and important than comparative linguistics.

Alexis MR