LINGUIST List 5.1134

Mon 17 Oct 1994

Disc: Comparative method

Editor for this issue: <>


  1. , Re: 5.1128 Comparative method
  2. , Re: 5.1128 Comparative method

Message 1: Re: 5.1128 Comparative method

Date: Sun, 16 Oct 94 19:13:08 ESRe: 5.1128 Comparative method
Subject: Re: 5.1128 Comparative method

 Alexis Manaster Ramer is certainly right when he says that the limit
on the comparative method was used by certain scholars who refused even to
discuss the possibility of long-range comparisons. Quite often this limit
is cited by the same scholars who reject glottochronology as well. In this
latter case, I would like to know how they compute the particular dates of
8000 or 10000 years?
 Chance similarities are always a possibility, but to the best of my
knowledge I have never seen a proof to a widely circulating idea that one
can take any two languages at random and prove that they are genetically
related on the basis of chance similarities between them. It will be
absolutely impossible to find any REGULAR phonetic correspondences bet-
ween any look-alikes due to chance. Besides, how many chance look-alikes
it is possible to find among two unrelated languages? Very few indeed. Those
who disbelieve it are welcome to try, let's say, English and Mandarin. Even
if one succeeds in finding, for example, 10 English-Mandarin look-alikes
within for example 200 word list, I bet that no regular phonetic corresponden-
ces will be established on their basis. Therefore, as long as we can establish
these regular correspondences, I believe that comparative method is working,
and it does not matter on what level we apply it:on Germanic or Nostratic.
>From this point of view Nostratic has no less credibility than IE: it has
been established on a certain set of regular phonetic correspondences, each
of them supported by numerous examples. Amerind, on the other hand, is a
bunch of wordlists, not connected by any correspondences at all, and, there-
fore no more credible than another Greenberg's hypothesis, "Indo-Pacific".
 I think it might be useful to keep apart reconstruction of a proto-
language and proof of genetic relationship in a sense that there are cases
than we can do the latter and not the former. We actually do not need
hundreds or thousands of words from two or more languages to prove that they
are genetically related. Let me give the following example. In appro-
ximately 5th-8th centuries AD a language, which we nowadays call the Kara
language was spoken on the southern tip of Korean peninsula. Only 15 words
remain from this language, transcribed by Chinese characters, but I need
only five of them TO PROVE BEYOND ANY DOUBT that this language is related
to Old Japanese:
 Old Japanese Kara
"three" mi-tu m(r)it
"mountain" takey "peak" ta?
"high" taka- ta?
"sea" nami "wave" nami
"water" mi- m(r)i
One gets the following correspondences: m:m, t:t, k:?, a:a, i:i, n:n. All
correspondences except last one occur two or more times WITHIN FIVE WORDS.
As a matter of fact, we may need only FIVE words to prove the genetic re-
lationship. Now, if one can get demonstrate the same kind of thing for PIE
and PU, does not it mean that they are related? But I highly doubt that
it is possible to present the same case for Amerind, Sino-Caucasian, Dene-
Caucasian, etc. On the other hand, it would be impossible to reconstruct
Japanese-Kara proto-language, since Kara data are limited to 15 words. That
is a serious set-back, and in certain cases exactly this lack of materials and
data is likely to keep us from getting clearer picture of a certain proto-
language, which in its own turn may not allow us to find out its genetic
origin, or to delay our progress in that process. But it should be these natura
l causes and not scholastically set limits which make the usage of the CM
impossible. "General consensus" is not a serious argument, since today we
have one "general consensus", and tomorrow the other.
 Finally, I disagree with the statement that usage of reconstructed
protoforms obscures the CM and cannot be used in establishing long-range
genetic relationships. That may refer to an unaccurate reconstruction, but
I would like to insist that a careful one can only contribute to a successful
clarification and even dismissal of earlier hypotheses. Thus, e.g., one of the
best Benedict's etymologies in favour of his Austro_Tai "family", Thai ta "eye"
as compared with Malay mata "eye", has hardly any chances for survival in the l
ight of Proto-Tai *pra. This and similar examples make the whole hypothesis
rather dubious.


Alexander Vovin
Dept. of German, Russian, and East Asian Languages
172 Irvin Hall
Miami University
Oxford, OH 45056
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 2: Re: 5.1128 Comparative method

Date: Sun, 16 Oct 94 21:57:21 EDRe: 5.1128 Comparative method
From: <>
Subject: Re: 5.1128 Comparative method

In response to Teeter: I completely agree that there is no way of
prejudging what elements a language should have for comparative
work to be possible.

In response to Guy: 40,000 years is, I think, everyone will admit
very good. But we would need to be told what rate of loss Jacques
is assuming and what kind of branching (this is absolutely crucial).
Also, it is vital to know what vocabulary we are talking about. Does
anybody know of any counterexamples to the claim that the old
Swadesh list has a rate of loss of no more than 14% per millenium?
My colleagues and I have been doing calculations using that rate
as well as the 27% rate claimed by M. L. Bender for a different
100-word list. And the results are that for any reasonable-szie
family (i.e., not Basque or Sumerian or some other complete
isolate), we should expect to be able to recover enough for
comparative work for much longer than 10,000 years, but it all
depends on how many languages and, even more, how they branch.
I cannot see how Jacques arrived at 100 years or whatever it was
for certain examples.

Also, even if some languages evolve slower and other others faster,
that would mean that given a group of languages of unknown
history, we CANNOT a priori rule out the possibilitty of a successful
comparative efort, because we would not know whether they are the
fast or the slow kind of language. Hence, even what Jacques says
supports my position that we cannot impose some kind of methodological
chastity belt and insist that everyone say No to proposals that go
beyond 10000 years.

Alexis Manaster Ramer
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue