Nostratic, Afro-Asiatic, and so on

[Alexis Manaster-Ramer]

----------------------------Original message----------------------------
The more I reread what has been said on this list
by various people, the more I feel I was right at
one point to suggest that perhaps some people do
not know the relevant literature at all, and if so,
I was castigating them for the wrong reason (and
apologize if that is the case).

It may therefore be useful for me not merely to
complain that a number of leading scholars and
so less leading ones like me are being ignored
and our work passed over in silence that to summarize
some of the things which are apparently not known
at all to a lot of people and which I was taking
for granted.

1. The idea that languages can be classified
into families on the basis of facts which we
take to indicate that the members of each family
come from a single ancestral language which is NO
LONGER spoken (a protolanguage) is what many
scholars assume we say that there is a family
we call Indo-European, another one we call
Semitic, another one (subsuming Semitic) which
used to be called Hamito-Semitic and still is so
by some but usually is called Afro-Asiatic, etc.

Hence, when we talk about families that are more
controversial than these three, e.g., Nostratic,
Altaic, Nadene, Pakawan, etc. (more of these later),
we mean that some people say that there once was
a Proto-Nostratic, a Proto-Altaic, etc., while others
say either that there was not, or that we do not
have good enough evidence that was, or that there
cannot in principle be such evidence (for reasons
I will go into later).

However, and this is important, there have at all
times been (and there are now) scholars who deal
with languages (including some who one clearly
has to count as linguists and many who one does not,
for not all scholars of language are linguists)
and who are/were very good at what they do/did who
either deny the reality of ANY and ALL protolanguages
and hence have an entirely different notion of
'language family' and/or who deny that some particular
ones among the widely-accepted protolanguages ever
existed.  Some of the names here include(d) Prince
N. N. Trubetzkoy (better known for his work in
phonological theory than in historical linguistics,
but still clearly knowledgeable in the latter),
Kotwicz (whose first name escapes me at the moment,
one of the leading specialists in Altaic languages
in the thirties and fifties), Denis Sinor (a leading
Altaic scholar of the present day), Gerhard Doerfer
(one of the great, and deservedly great, names of
Altaic studies, but especially Turkic and Mongolic),
and from what I can gather many if indeed not most
people who deal with Semitic languages except those
who are linguists rather than Semiticists and perhaps
with some other exceptions.

It is often hard to know if anyone denies ALL
protolanguage as a matter of principle or only some
as a matter of fact.  Trubetzkoy denied Proto-IE
and I think wanted to deny them all but am not sure.
Kotwicz denied Proto-Altaic but I am pretty sure never
spoke to the general issue of principle.  Sinor, who
is more a historian than a linguist but knows more
about more languages than many linguists, seems to
deny (or be skeptical about) the whole idea of
proto-lgs and has been a vocal critic of Proto-Altaic.
Doerfer seems pretty clearly to accept some proto-
languages but not others.  At one point he published
the idea that the Anatolian languages (like Hittite)
are not descended from a common protolanguage with
the rest of IE, for years (as I have mentioned) rejected
and still seems to reject the idea that all the Afro-
Asiatic languages have a common origin (and likewise
re Uralic) and is of course widely celebrated for
his tireless campaign against Proto-Altaic, but
I think he clearly accepts a Proto-Turkic, for
example, and a Proto-Semitic. It is another matter
that as far as I know, and I think I would know,
he has no credentials whatever which would entitle
him to an opinion re IE-Hittite, Uralic, or Afro-
Asiatic.  Many Semiticists, whom again I would not
generally (but there may be exceptions) describe
as linguists but who again know more than many
linguist about all kinds of languages, seem to
deny the reality of a Proto-Semitic.  As for Afro-Asiatic,
many Semiticists in my experience do not reject it
opnely, but simply pretend as though it had no
relevance in practice to their work and say little
about it, but there have been some strident voices
recently (and not so recently) hotly denying that
Semitic is related to the other Afro-Asiatic languages.

(2) Given (1), it is hardly meaningful to speak
of "established" or "proven" language families
(and protolgs) like IE or AA or Semitic as opposed
to ones which are not "established" or "proven".
One must of course distinguish degrees of controversiality.
IE and Semitic are less controversial than AA, and
AA is vastly less controversial than Nostratic or
Altaic.

(3) The Nostratic theory, which seeks to relate
IE to some or all of the following (depending
on the particular version of Nostratic, since
there are many, as we will see) has, like most
theories, complex roots, but the term goes back to
Holger Pedersen, a Danish Indo-Europeanist of the
late 19th and the first half or so of the 20th
century, who was (this is one thing that UNcontroversial)
one of the very best Indo-Europeanists of his or any
age, known in particular for massive contributions
to the Celtic, Armenian, Anatolian, Tocharian,
and several other branches
of IE (like Meillet, his great French rival and
rough contemporary, he seems to have focused more
than some others on the branches of IE which were
not widely or deeply known to many IEnists then and
indeed today).  He had some experience with
the questio of language classification inasmuch
as he was involved with the battles within IE
about the IE affiliation of the Anatolian languages
(many people do not know this, but this was once
almost universally denied by Indo-Europeanists
and took a very long time to become universally
accepted).  However, I do not know that he had
any great knowledge of the other Nostratic languages,
though he clearly knew some Turkic and some Semitic
at least, and I do not know that he actively worked
in any Nostratic family other than IE or even on
the problem of establishing the validity of Nostratic.


(4) For the next sixty or so years (roughly 1900+
through 1960+), as now, most scholars ignored
the Nostratic issue and such work as was being done
on it.  And the work that was being done (one
particularly important name here is that of Bjorn
Collinder, one of the leading specialists in
Uralic linguistics) tended to involve pairs
of Nostratic subfamilies (most often Indo-European
with Uralic) and tended to have trouble finding
precise sound correspondences even there.  Nor
was there much work on reconstructing a Proto-0
Nostratic.  In the 1960's Vladislav ("Slava") M.
Illich-Svitych and Aron (Aharon) B. Dolgopol'skij
(Dolgopolsky) apparently independently convinced
themselves of the reality of Nostratic (each had
a slightly different list of language families
in mind and I believe Dolgopolsky long resisted
the term 'Nostratic'), but each told of this
to one and the same person who finally brought
them together. I-S (and to a much lesser extent)
Dolgopolsky then proceeded to (a) compare several
language families all at once, and (b) try to
reconstruct a common Nostratic proto-language.
Furthermore, (c) realizing that the existing
body of knowledge about the proposed components
of Nostratic was very inadequate, they proceeded
to make changes (ranging from minor, as in IE,
to major, as in Uralic and Dravidian, to really dramatic,
as in Altaic) to the existing reconstructions
and in the case of Afro-Asiatic, got busy with
the basics such as trying to reconstruct some kind
of picture of the major branches of AA (I-S worked
on Chadic, Dolgopolsky on Cushitic) and laying
the groundwork for a reconstruction of Proto-AA
(which neither really accomplished to a very
great degree).

I-S died before most of his work was either done
or even published, and Dolgopolsky after doing some
important work emigrated from the USSR to Israel
and for a long time published almost nothing of
relevance as he was adjusting.  I-S's unpublished
work was published by a changing team led by
the man who had first introduced these two, Dybo,
and while it was a heroic and maybe even superhuman
effort, I have elsewhere noted that it was in many
ways a far-from-successful one.  In any case it
took a very long time, during which for all practical
purposes very little substantive work (for many
years none at all) on Nostratic was published
or even I suspect done by anyone of this school.

An American scholar, Alan Bomhard, quite independently
came up with a whole series of arguments for a
somewhat (actually importantly) different view
of Nostratic, but after learning of the I-S and
Dolgopolsky work, some of the gap narrowed.  His
work, while not widely and certainly not deeply known,
did receive a certain amount of exposure, incl.
a number of book reviews.  I-S's and Dolgopolsky's
work on the other hand was widely ignored even in
Russia, where to this day as far as I know the only
published review of I-S's posthumous Nostratic
dictionary is a recent translation of a review
I myself published earlier in English in Bernard
Comrie's Studies in Language (in 1993 I think).
In some European journals, especially in one
pubslished in Czech, notice was taken of the work,
but in the English-speaking world and especially
in the US, that was not the case.  Comrie's
invitation to me to write to the review
was a  brave and radical departure from tradition, some
twenty years after the first volume of the
book being reviewed was published.

To this day, despite private and public appeals
from some leading American linguists, some
historical but some not (incl. Hamp, Fromkin, Wasow,
and others), the journal Language (edited by
Sally Thomason and then by Mark Aronoff) has
refused to do a review of this enormously
important book, and as far as I know, the first
(and maybe only?) discussion of the school
of Nostratic founded by I-S and Dolgopolsky
in that journal was in a brief review I have there
of a Russian-lg encyclopedia of linguistics,
whose treatment of Nostratic I got a chance to
pan.  Other journals of course have done little
better.

Speaking of encyclopedias of linguistics, some
have chosen to ignore or dismiss Nostratic
(e.g., Bill Bright's, which is probably the
best known one in the US), some have begun
mentioning it, as does (or will) the new
edition of the Encyclopedia Britannica.
Most textbooks of historical linguistics have
only recently treated Nostratic or still do not
and if they do, they generally purvey disinformation
(there may be exceptions).

While essentially no substantive work on Nostratic has
been getting done by the self-styled school of Nostratic
in Moscow (which is why I have often said that there
are no Nostraticists at the present time, although
this is slightly less true today than when I said it
a few years back) or very little by Dologopolsky isn
Israel (although I keep hearing that he has been
DOING a lot, but only publishing very little), Bomhard
in this country has done an impressive amount of
research, most of which I find not compelling (but
some improtant scholars have praised it highly) and
little bits and pieces have been done by various
scattered people in Europe (of very uneven quality).
But back in the US, Vitaly Shevoroshkin, a personal
friend of I-S and Dolgopolsky's but himself an Indo-
Europeanist (mostly an Anatolianist in fact) found
himself through the late 1970's and the 1980's
the possessor of one of the best-kept secrets in ling
and tried in various ways to tell American (and other
Western) linguists about, with mixed success. He also
at length did a small amount of work on Nostratic
himself and trained one student, Mark Kaiser, who
did some work as well but has now left the field.

Shevorshkin also managed to hold the first conference
outside of the USSR that dealt to a significant
extent with Nostratic and related issues, to get
some coverage of the Russian Nostratic work in
the popular press, and the like, and he inspired
a few people (incl. me) to continue making noise of this kind
after he himself largely gave up in disgust at the
modest gains.  As a result, there was some more
press coverage, more conferences (one at Ann
Arbor, the next at Ypsilanti), the failed effort
to get Language and other journals to review I-S's
work, etc.  I believe that it was this noise that
inspired Comrie to make his move to get I-S's
work reviewed in Studies in Lg, although I do not
now recall the events clearly.

A small amount of (to my mind entirely misguided)
criticism of Nostratic was also provoked, esp. on the part
of Donald Ringe, which is better than the silent
treatment, of course, but only marginally so.

On the other hand, I do not know what got
Brent Vine (now at UCLA) to write the only critique
of Nostratic to date which neither the author who
wrote it nor the editor who accepted it should
feel deeply and mortally ashamed of.  This is not
to say that I agree with almost any of it, but
it is a piece of honest scholarship by an honest
(and excellent) scholar (some rebuttal of it
appears in the paper which I coauthored with
Michalove, Adams, and Baertsch appearing in
the just-published Joseph-Salmons anthology).

This brings us into the early or mid-1990's, by
which time I had become somewhat tired of merely
fighting for the right of Nostratic to be heard,
and decided to see if I could actually believe
any of it.  As I recall, my first project involved
looking at a proposal of I-S's involving a proposed
Nostratic etymology for the Armenian plural ending -k`.
To do this, I had to learn IE, which took a bit longer
than it should have since I was teaching computer
science full time and had to learn THAT first,
and ended up writing an unbelievably long paper
on the problem, which turned out to be a neat
problem for IE (with Pedersen and Hamp being my heroes
on this occasion, as on many later ones) but without
finding out anything useful about Nostratic.
I then did some work which started out trying to
evaluate (and as it turned out usually rebut) all
published critiques of Nostratic, many quite
obscure, that I was able to find (except Vine's,
which I intended to deal with separately).  I
think I did manage to deal with all the arguments
involving typological issues in a paper in JIES
not too long ago, and many otheres elsewhere, but
a few may remain.  (A series of replies to Ringe's
series of attacks has started, too.)

And in the process I found myself doing two
other things I had not envisioned: (a) criticizing
more and more of the I-S and Dolgopolsky work
and (b) making some proposals of my own for
improving on it or adding to it.  A couple of years
ago I had all but convinced myself that Nostratic,
or at least the cluster IE-Uralic-Altaic, was
really related, but was too ill to publish
the arguments (some of them were publicized by
George Johnson in the New York Times, and some
are discussed in the paper in the Joseph-Salmons
anthology).  Some other bits appear in various
papers, and some remain unpublished.

Most recently, Dolgopolsky has published a little
book which makes me cringe and which almost convinced
me that Nostratic was all a matter of a set of
coincidences and a set of very old borrowings.
On the heels of the book, Lord Renfrew and others
organized a conference on Nostratic whose proceedings
will be out soon.  Also, Brian Joseph and Joe
Salmons, two honorable and excellent Indo-Europeanists,
put out an anthology of papers pro, con, and on
Nostratic based mostly on the Ypsilanti conference
of a number of years ago. And I cannot see how
the main journals will be able to avoid reviewing it,
so some word will surely go out now.

Also, even as some of the so few people working on
Nostratic have been dropping out (not usually of their
own choice), a small number of excellent minds from
different backgrounds have in various ways come to
deal with Nostratic.  If I have made any real
contribution, in fact, it is having gotten Peter
Michalove to work on Nostratic, and having gotten
Bill Baxter (in my view one of the best linguists,
esp. historical linguists in the US) and Christopher
Hitchcock (a philosopher of science at Cal Tech)
to grapple (with some small help from me) with the
mathematical (or as I hold voodoo mathematical)
attacks that have been on Nostratic, esp. by Ringe.

Of course, there are other names too that should be
mentioned, esp. perhaps those of the people (usually
quite misguided in my view) whom Joseph and Salmons
got to write anti-Nostratic papers for the anthology,
e.g., Lyle Campbell, Brent Vine again, etc., and the
various people (whom I have not had the pleasure
to meet, due to illness) whom Renfrew et al. assembled
at their recent conference.

(5) So what families belong to Nostratic?  In
I-S's formulation, IE, Afro-Asiatic, Uralic,
Altaic, Kartvelian, and Dravidian.  But there
are many variations on this, though I think
IE, Altaic, and Uralic are part of all schemes.

END OF PART 1.  If there is interest and I can
find eht energy, there will be a part 2 later.

Alexis MR