LINGUIST List 9.453

Tue Mar 24 1998

Disc: State of Comparative Linguistics

Editor for this issue: Martin Jacobsen <martylinguistlist.org>

Directory

1. manaster, Re: 9.431, Disc: State of Comparative Linguistics
2. Kawagashira Nobuyuki, Disc: State of Comparative Linguistics
3. Karl V.(van Duyn) Teeter, Re: 9.431, Disc: State of Comparative Linguistics

Message 1: Re: 9.431, Disc: State of Comparative Linguistics

I must say that the response to my posting on this subject and the
ensuing discussion have exceeded my wildest hopes. I think all
linguists know the frustration that comes from our inability to combat
effectively teh misconpceitons about language that the general public
has and which the popular press and the popular books about lg pander
to. I have long felt the same frustration with regard to the
misinformation about comparative linguistics (esp. but not only
linguistic classification) that goes uncombatted INSIDE the field of
linguistics.

John Koontz is of course right with his suggestions, which I welcome
and which I hope will one day be realized, about how accurate
information about comparative linguistics might be collected in a
convenient and accessible way. But to do that, we need a certain
critical mass both of possible contributors AND possible readers. I
hope that the current discussion can help on that score.

In the meantime, of course it is wonderful that LINGUIST itself can
serve as a fast and effective way to combat misinformation, such as
the monstrous libel of Niger-Kordofanian by those who have suggested
that there is no basis for this family. I daresay that those of us
who have addressed this point here have now reached at least as wide
an audience as the original libel had.

What is more important is that there a number of issues on which there
has emerged in effect a consensus to either conceal the true state of
affairs or to spread misinformation, so much so that nearly everybody
in linguistics just repeats the stuff as though it were divine
revelation or a priori logiv when in fact we are just dealing with
particularly entrenched urban myths.

An example: who would question the "well-known fact" that after a few
thousand years (given variously by various texts as somewhere between
6 and 12) languages change so much that any evidence of their original
genealogical connections will have been lost? But does anybody know
where this bit of urban mythology come from? Or does it not strike
anyone as odd that different authors give different numbers without
ever discussing the discrepancy? Or that almost no author ever gives
the actual computation required to get the number? Or cites a source
where sucha computation mightbe found?

It was by the chanciest of acceidents that I found what I believe to
be the source of this urban myth, namely, a rrelatively obscure
passage in a relatively obscure work of Swadesh (a name unknown to
many younger linguists or only known as that of someone who invented
that disreputable bit of mathematical ling known as glottochronology)
in which he indeed argues for such a cut-off point on how far back
into the past GLOTTOCHRONOLOGY can reach. Somehow, in the rush to
prove that the end of comparative linguistics is at hand, however, all
kinds of people who really should know better decided, however, that
they could apply this result (although without citation and w/o the
decency to get the exact number right) to COMPARATIVE LINGUISTICS
generally.

This of course is analogous to say thing that anything invisible to
the naked eye is inaccessible to observation, forgetting that there
are for example such things as microscopes, for the relation of
glottochronology to the other methods of comparative linguistics is
like that of eye to microscope.

Alexis MR

P.S. A small point raised by Mohamed Diriye Abdullahi has to be
addressed. He questions my point that the Afro-Asiatic language family
is a counterexample to Doerfer's "universal" which holds that
genuinely related languages always have cognate numerals 2-4 and
certain (I did not list them) body part terms. It is of course true
as MDA says that some some numerals and some body part terms can be
reconstructed for Proto-AA, but it is also true (and that is what is
relevant here) that there are pairs of AA languages which share no
cognate numnerals except maybe '5' (e.g., Hebrew and Burji) and
similarly (although the details differ) for body part terms. The
numerals story is discussed in a recent paper by Sidwell and myself in
Wiener Zeitschrift fur die Kunde des Morgenlandes, and the body part
story in one by Sidwell, Vovin, and me coming out sometime in
Ural-altaische Jahrbucher.

	Dear LINGUIST Subscribers,

Manaster wrote:
 
 > A third concern I have is that recent textbooks
	> of historical linguistics, written typically by
	> people who (while excellent at other things) have
	> done little or no work on language classification
	> have been spreading all kinds of misinformation
	> about this, the most difficult in some ways,
	> area of comparative linguistics. Misinformation
	> in textbooks of course is the hardest kind to
	> combat, because introductory level students
	> will not read technical articles where the real
	> information is to be found.
	
I agree with his idea. I have had the same anxiety as he has.
Misinformation and questionable theories are written even in many
introductory books. This situation confusing students and scholars.
Maybe many (I hope, some) writers, including linguistic specialists,
does not notice it.
	
A mongolic specialist says that Altaic theory is not proved. Such a
situation is the same in Chino-Tibetan. As far as I remember, A famous
Chinese linguist Wang Li insisted that Chinese and Tibetan has no same
linguistic root. But I felt that this information is not known to all
linguists, causing no arguments.

 Nobuyuki Kawagashira
 General Linguistics
 University of Tsukuba
	

Message 3: Re: 9.431, Disc: State of Comparative Linguistics

	Concerning the state of comparative linguistics, a rather
diffuse debate I have been following, it is not true that there are no
criteria for genetic relationship, for goodness sake, as suggested by
Benji when he writes, "Over time any language can change and/or borrow
anything (or everything?) from other languages under conditions of
contact." In part it seems to me that this notion comes from the
mistaken assumption that genetic relationship is a matter of
statistics with words, an American fallacy I have written about
previously.

	In fact, it seems to me that the linguist seeking to study
genetic relationship does very much what the linguist approaching a
new language does. One gathers data and writes a grammar to account
for it. In descriptive studies this results in a grammar of the
language, in comparative studies a grammar of a
protolanguage. Historical/comparative linguistics is the construciton
of grammars for protolanguages. As for borrowing, there is indeed no
limit, since in fact one can learn a foreign language, which involves
a hundred per cent borrowing. But, as Meillet knew, there is a
difference between borrowing lexical items and borrowing grammatical
structure. You can hear words, but nobody has ever heard a grammar,
which is a system constructed anew by every speaker, and this is the
crucial difference which allows us to do the history of a language and
distinguish borwoings from retentions. So in principle, while one can
learn a new language and thus incorporate a grammar which makes it
look as if everything has been borrowed, in practice there are limits,
which is precisely why we can do the history. Words can be borrowed
but not structure -- where it looks as if there has been structural
borrowing, it must be seen as a case of partial language learning.

	May I throw in a casual remark relevant to the debate? On the
matter of how insulting it may be to call a professor "Mr.", my
teacher Murray Emeneau, the foremost authority on Dravidian languages
in the West and thriving in his nineties, is known by everybody as
Mr. Emeneau, and I know none who do not consider it a term of honor.
Yours, kvt