LINGUIST List 5.929

Mon 29 Aug 1994

Disc: Altaic

Editor for this issue: <>


  1. , Re: 5.926 Altaic
  2. "Reinhard, 5.926 Altaic

Message 1: Re: 5.926 Altaic

Date: Sun, 28 Aug 94 13:14:42 ESRe: 5.926 Altaic
Subject: Re: 5.926 Altaic

I am very glad that Marshall Unger testified himself that contents of his
1990 article were twisted. I believe that Unger's approach is typical for
the absolute majority of historical linguists on Japanese/Korean side of
Altaic in North America: nobody doubts Japanese-Korean genetic relationship,
and everybody looks with a great deal of sympathy on the the further connec-
tion with Manchu-Tungus. This is, for example, as far as I know, Samuel Martin'
s position, too. As far as Mongolian and Turkic are concerned, opinions vary,
but even here to the best of my knowledge we have uncertainty or various
shades of doubt, but never direct denial of a possible genetic link. However,
this uncertainty is used by anti-Altaicists here and there as a support of
their own position!
 I can't help but to draw a political analogy between anti-Altaistics and
Marxism-Leninism. Marxism-Leninism is "omnipotent because it is true" or vice
versa is"true because it is omnipotent". In other words, neither bothers
with providing a proof, while the opponents are classified as "backwards",
"sectarian", "enemies of the people (or scholarship)". Both love to use
slogans (see Doerfer's works). Lenin devised a theory that a proletarian
revolution can breach a chain of capitalism in a single country: so did the
theory of LOANWORDS breach the comparative method in Altaistic. The results
were disastrous in both cases: a number of peoples of Europe and Asia strayed
away from the main highway of development, and found themselves after 70
years exactly where they embarked on the marxist ship. Altaistic, too, was
almost strangled for thirty years, but the tide has turned now. Marxism-Le-
ninism has brought into being and cherished for many years Nikolai Marr's
"New Language Teaching", which turned comparative linguitics upside down,
claiming that any modern language originated from hundreds of thousands
genetically unrelated languages. It seems that theory of omnipotent loanwords
so popular in Altaistics, leads us exactly in the same direction. Twisting
other peoples works, and citing certain passages from them which distort the
general picture and do not reflect author's opinion, but may be used in one's
own interests is very typical for marxism-leninism. The case with Unger's
article is far from being the exception. In 1986 I personally heard Shcherbak
claiming that G. Ramstedt's "Einf"uhrung" is a student's work, and that later
in his life(?!) the great Finnish scholar turned to anti-Altaistic as he
realised that this is the only true approach to the data: Anti-Altaistic is
omnipotent, because it is true; it is true, because it is omnipotent. The reade
rs of Altaic discussion in the forthcoming JSFOu will definitely enjoy
seeing Samuel Martin's name cited as an opponent of not only Altaic, but Ja-
panese-Korean relationship as well. REcently at the "international Circle of
Korean Linguistics" in London I was asked by one of colleagues: "I have heard
that you are an opponent of Altaic, since you disproved Ainu-Altaic relation-
ship, is it true?" The chain of these anecdotes may be easily continued. The
problem is, however, that there are too many of them, and the more anti-Alta-
istic loses its ground recently, the more we hear them. And, therefore,
I am far from being sure that Unger is right saying that "these things happen":
it more looks like a deliberate desinformation -- another trick frequently
used by marxists-leninists. Finally, the nature of scholarly discourse among
anti-Altaicists also reminds me closely a Marxist-Leninist one: any kind of
Altaistic research is branded "fallacious", "futile", "mad", "crazy", "Omni-
comparativismus", "incredible" etc. without providing any basis for such
claim. On the other hand, any attempt to argue or criticize anti-Altaistic
approach usually is met as "attacks ad hominem" -- very convenient way to
avoid any real scholarly polemics.
 The only constructive criticism I have seen recently was B. Comrie's
review of Starostin's book. Comrie seems to be alone among anti-Altaicists
who tries to criticize the system of correspondences proposed. Thus, he
demonstrates that Tungusic dilacaa "sun" cannot be related to Japanese tosi
"year" and Korean twols "anniversary", because it violates the system of
correspondences proposed by Starostin himself. I cannot but agree with Comrie
on that issue. That, however, does not prove that Japanese and Korean words
are not related to each other and to Turkic *yal2 "year of age" and Mongolian
*dzil 'year'. Starostin is not a God, and he cam make mistakes, and this is
not the only one in his book, but this, however, does not diminish the overall
quality and importance of his book. And, certainly, disproving one or even
ten etymologies (really disproving, like Comrie does on the basis of violation
of proposed correspondences, and not "disproving" by application of the epithet
s cited above) does not yet demonstrate that the whole theory is wrong. But I
will certainly like to see constructive criticism of Altaic -- it is stimula-
ting and helps to improve the reconstruction.
 Both Comrie and Janhunen accuse Starostin of easily dismissing "competing"
etymologies, such as Austronesian for Japanese. Obviously, it was not in Staros
tin's mind to provide detailed criticism of Benedict's or Kawamoto's opuses --
his book has quite different goals. Anyone reading Benedict or Kawamoto can
easily detect that there is no REGULARITY in the correspondences they propose
between Japanese and Austronesian (for detailed analysis please see my two
review articles in Diachronica XI-1, 1994 and forthcoming in December issue of
"Oceanic linguistics"), and, therefore, Austronesian hypothesis cannot even re-
motely stand a chance in competing with Altaic theory, where all correspondence
s are regular. Moreover, different Altaicists may disagree one with another
on certain number of etymologies, but agree on the main bulk which constitutes
at least 80%. The case of Benedict and Kawamoto is quite different: they more
often propose completely different Austronesian etymologies for one and the
same Japanese word than they don't.

Alexander Vovin
Miami University
Dept. of German, Russian, and East Asian Languages
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 2: 5.926 Altaic

Date: Sun, 28 Aug 1994 21:44:27 5.926 Altaic
From: "Reinhard <>
Subject: 5.926 Altaic

Thanks to Alexander Vovin for his contribution to the current discussion
about the Altaic Hypothesis (AH). Unfortunately, it seems to me that he,
and hopefully only he, has been under the mistaken impression that I
disagree with his fundamental premises, ideas and arguments as well as
with those of Alexis Manaster Ramer. I do not. I apologize in case it
was I who unwittingly created this impression.

One of the basic purposes for my initial posting was to support Alexis
Manaster Rames' call for discussion of the AH at a time when certain
non-Altaicist linguists unquestioningly choose to follow the
anti-Altaicist route. Another purpose was to provide a brief overview of
what I believe to be the basic premises of the AH and the major trends and
dynamics in its history. I deemed this necessary as a courtesy toward
those in this extensive forum who have less or no familiarity with the
subject matter. Although I made no attempt to hide my pro-Altaicist
position, I tried to refrain from making my overview too personal, save
for a few terminological and strategical suggestions. It is a mistake to
assume that my outline reflects my personal picture of the Altaic family
of languages.

I had been under the impression that the anti-Altaicist camp has a
tendency toward wanting it both ways: on the one hand they make much of
the alleged lack of correspondences among numerals and body part terms
(which many of us understand to have arisen from Indo-European-based
expectations), and on the other hand they dismiss Turcic-Mongolic-
Tungusic cognates as "loanwords" and the pronominal correspondences as
basically irrelevant. Certainly, Alexander Vovin's point regarding the
"bizarre" methodology of the anti-Altaicists is well taken, though it
seems to be more obvious to him than to me that it was actually *in
opposition* to comparative methodology as such. This is an interesting
point I am sure many of us would like to learn more about.

I did not intend to impose my own opinion with regard to extent and depth
of "Altaic." In fact, I did not express my opinion, nor did I as much as
say that I *have* an opinion about this. In my simplified overview I
merely intended to show that *everyone* who does not reject the AH
definitely considers Turkic, Mongolic and Tungusic to be Altaic. While it
may be regrettable to Alexander Vovin and others, it is still a fact that
so far not *everyone* is totally convinced that Korean and Japanese are
related to them or are closely enough related to them to be called

As for terminology, I suggested the non-compound forms for the sake of
simplicity. Certainly, there is good reason to use the compound
"Manchu-Tungusic" in place of simply "Tungusic." However, there is also
good reason to use "Chuvash-Turkic" or "Bolghar-Turkic" in place of simply
"Turkic," but few people do so. As for my suggestion to use generic
"Mongolic", it is after all consistent with "Tungusic" and "Turkic" ("-ic"
being the typical generic marker), and it is partly for this reason that
my suggestion has been welcomed by many who kindly sent me private
responses. Furthermore, "Mongolic" can be easily distinguished from the
specific language commonly known as "Mongolian." As for Alexander Vovin's
suggestion to keep "Mongolian" as a generic term and to refer to the
specific language as "Khalkha", it would not only be inconsistent and
create confusion, at least among non-linguists, but it is also incorrect,
since Khalkha is only one among several dialects of what is generally
known as the Mongolian language, other dialects being Chakhar, Urdus,
Urat, Tumut, Kharchin, etc. This suggestion could be compared to
referring to the Germanic languages as "German" and to start referring to
German by a dialect name: e.g. "Franconian."
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue