LINGUIST List 3.100

Mon 03 Feb 1992

Disc: Proto-World (Part 1)

Editor for this issue: <>

Directory

1. Swann Philip, 3.87 Proto-World
2. Eric Schiller, Re: 3.87 Proto-World
3. , 3.87 Proto-World
4. Dan Everett, Re: 3.87 Proto-World
5. Richard Sproat, 3.87 Proto-World

I'm not sure it's really fair to describe Scientific American as part
of the 'popular science press' - the articles are nearly always written
by the researchers who did the original work, not by journalists.

Interestingly, the only reference to Language in the 1991 index of 'Nature'
is to a discussion of Language Origins by Robert Foley (vol 353, 114-15).
He refers to Cavalli-Sforza and reprints a figure from the 1988 paper.
His main topic, 'though, is a fascinating paper by Nobel and Davidson
(Man, 26, 223-54) on the probable date for the emergence of language in
hominids (circa 40,000 years ago). If 'Nature' thinks that the discussion
is worth recording, then linguists should perhaps hesitate before dismissing
it as pseudo-science (people in glass houses ...).

Philip Swann
FPSE-TECFA
University of Geneva

Message 2: Re: 3.87 Proto-World

For the record, my remarks were directed at the July 1991 Article more
than at the Proto-World one, and I do believe that outrage is appropriate
when one point of view, not currently accepted by a plurality of experts
in the field, is presented as gospel, without any mention of the tentative
status of the proposal. Austro-Tai is one such myth. I have no objection
to scholars examining the hypothesis, testing it against accepted
standards of reconstruction, and comparing it to established alternatives
(Austric predates Austro-Tai by almost 40 years).

Summing up the panel of experts at the 6th International Conference
on Austronesian Linguistics would take more space and time than is
available, but among the hypotheses:

Austro-Tai (Benedict)
Austric (Reid, Diffloth, Schiller and of course Pater Schmidt!)
Sino-Austronesian (!?) (Sargent)
Nostratic link (Shevoroshkin)
healthy skepticism toward any link (most of those assembled)
Macroasutric (suggested as a possibility - Schiller 1987, based on
 hints and discussions derived from Diffloth, Haudricourt, not
 that either should be held accountable!)

My complaint is not that Austro-Tai is wrong (though I believe it is).
It is simply that there is no reason why it should be assumed in so
much of the literature. From linguistics, the assumption now makes
its way into the Scientific American literature, and those looking
into the questions of deep genetic affiliation are concentrating
on a single (probably incorrect) path, rather than using their methods
to see if the evidence supports one of many theories. Bad science
generated by ignorance - should we not react with some degree of
emotion?

It is if one were to treat Government and Binding (pick your vintage)
as THE theory of grammar! (<grin>)

Eric Schiller
University of Chicago

Just wanted to respond to Grover Hudson's remarks. I really do think
it is OFTEN the case that choking off a wrong-headed line of research
is a proper thing to do, and that reprimanding researchers for shoddy
thinking is often proper.

[That said, I decline to take a position on this particular issue.]

As professionals we have an OBLIGATION to police our disciplines and
weed out snake-oil salesmen.

I agree, we should refrain from tromping too hard on attempts to
explain phenomena which we see as unorthodox or unlikely to be
correct. We want to `let some large n of flowers bloom' to rephrase
Chairman Mao. However, there is no reason for us to encourage
attempts to explain phenomena which do not exist! That's not
scientific freedom --- that's absence of standards.

Message 4: Re: 3.87 Proto-World

Andrew Carnie's objections to the `play' that Historical Linguistics
is receiving in the press puzzle me. I am not a Historical Linguist
and I find descriptive linguistics, formal syntax, phonology, and
issues of learnability to be the most interesting areas of the
discipline. But, I recognize that these interests are simply my
personal baggage. Just because I do not do something or because it is done by
fewer people in the field does not mean that it is any less important
to the field or that it should not receive priority by the press - it
is good for us to know that others find this interesting. And since
when does the raw numbers of people working in an area determine that
it is central or peripheral to the discipline? This sounds rather
ignorant. If this were true, then maybe the central issue of the field
right now is Chinese linguistics - who has counted the number of
people studying Chinese around the world? This reasoning might lead
us to conclude that, within the U.S. at least, computational
linguistics is becoming the central area (more money, fastest-growing
number of researchers (?), more interdisciplinary interest, etc.).
(Perish the thought!) Maybe we should take a lesson from this - maybe
society is right about what the most interesting areas of the field
are and the specialists are wrong. I doubt it, but it's worth
considering. I am certainly impressed at the number of very
intelligent people that I meet from other disciplines and even within
linguistics that do not think that `tree-oriented syntax' is central
to the field.

That said, I agree completely with the suggestion that
psycholinguistics, syntax, phonology, etc. should be reported on more.
It isn't entirely ignored though - US News & World Report has reported
recently on Steve Pinker's work. If people outside the discipline do
not share our views of what is significant, we should take this as a
challenge. We cannot blame them if we are too arcane. Moreover, my
appreciation of other areas of the discipline and their `centrality'
usually diminishes in direct proportion to my ignorance of them.
So, Andrew, don't worry about it. Just do it if you like it. I used
to get mad because there were no Nobel prizes in linguistics, which is
to me the most interesting and significant discipline in existence. But,
my father still cannot believe that I get paid to do this stuff at all!

Dan Everett

In response to Andrew Carnie's point about what Scientific American
should be publishing that relates to modern linguistics, it should be
noted that back in 1983 they did publish a paper by Bickerton on
Creole languages and their relevance for the issue of innateness.
Without going into the issue of whether Bickerton is right about such
points, which I know to be a bone of much contention, the article did
at least serve the function of bringing up the issue of innateness of
linguistic knowledge. Also, in previous issues they had published
papers on speech errors, on computational linguistics and speech
processing, and a few other areas of current interest. The coverage
has been spotty, it is true, but it has not been nonexistent. Besides,
it is not clear that Scientific American has been any worse at giving
a representative sample of work on linguistics than it has been for
other disciplines. For example, by looking at what get's published on
physics, one might think that current interest is centered solely on
cosmology, string theory and a few other topics of cosmological
importance. True, these are hot topics, but they are not the only
topics. The point is that one has to remember that Scientific American
is first and foremost a magazine for popular consumption, and they are
therefore naturally bound to produce things that stir the popular
imagination. Linguistic origins is certainly at the top of any such
list of soul-stirring themes, at least as far as language is
concerned.

Also, I can't help feeling that there is a large measure of Political
Correctness involved here: if one can claim or at least imply that all
languages originally came from one common parent, that seems somehow
more palatable than the implication that language, for all we know,
may have evolved separately in different places by different (and
presumably ethnically distinct) groups of people. So the Proto-World
theme has a particular appeal. What professional linguists need to do
(though I am by no means sure how to achieve this) is to popularize
the message that not all appealing stories are necessarily true.

Richard Sproat
Linguistics Research Department
AT&T Bell Laboratories
600 Mountain Avenue, Room 2d-451
Murray Hill, NJ 07974
tel (908) 582-5296
fax (908) 582-7308
rwsresearch.att.com