LINGUIST List 3.110

Tue 04 Feb 1992

Disc: Proto-World and Popular Linguistics

Editor for this issue: <>

Directory

Joe Stemberger, Re: 3.101 Proto-World (Part 2)

Message 1: Re: popular linguistics

Date: Mon, 3 Feb 92 11:54:58 PSTRe: popular linguistics
From: Jim Scobbie <scobbieCsli.Stanford.EDU>
Subject: Re: popular linguistics

On Fri, 31 Jan 92 rwsmbeya.research.att.com (Richard Sproat) noted:

>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.

This weekend I checked the new books shelf in a local shop and found two science-like books. One was a primer, a 'all you need to know about science but were afraid to ask' volume. It had two entries on language, there was a one page discussion on language and thought and a longer piece on the languages of the bible, the Aramaic dictionary project and such like things. I thought this was rather sad: surely there is more about the science of language which our well-rounded book-reader ought to know?

The other book was called 'The third chimpanzee' (I think) and had a section about the origin of language. It asked howcome humans and the animal kingdom are so far apart in linguistic ability. After reviewing vervet vocabularly and holding out hope that natural animal languages (though simpler than ours) may perhaps exist, the author moved on to Creoles, and what that tells us about our genetic inheritance. Perhaps the author had read the Scientific American, or is the same.

If the authors are different then we can learn something at once -- that popular science recycles itself, and if we can present our findings in such a way that they appeal to the public and are accurate, then linguistics would quickly bootstrap itself into popular culture. Such a culture would then be able to handle the more esoteric matters with which we occupy our time :-).

Every day on the radio I hear 90 seconds of astronomical info put out by some university. Everytime I hear it I wonder if people would not be more interested in hearing a bit about language. Certainly *usage* columns and vocabulary spotting are perennial recreations. Why not something a little more abstract, something based on professionals' work? I'm sure the appetite is there, at all levels of interest. All we need are some appealing problems about language that we already know the answers to. -- James M. Scobbie: Dept of Linguistics, Stanford University, CA 94305-2150 scobbiecsli.stanford.edu

Message 2: language origins

Date: Tue, 4 Feb 1992 13:42 EST language origins
From: Mark H Aronoff <MARONOFFccmail.sunysb.edu>
Subject: language origins

State University of New York at Stony Brook Stony Brook, NY 11794-4376

Mark H Aronoff War and Peace Linguistics 632-7775 04-Feb-1992 01:15pm EST TO: Remote Addressee ( _linguisttamvm1.tamu.edu)

The lead article in the science section of today's (2/4/92) New York Times concerned the relation between Neanderthals and modern humans. I would guess that articles on human origins appear in the Times science section about four times a year, certainly more often than on just about any topic. If we compare this to the fraction of all scientists that comprises those working on human origins, we find that far more articles on this subject appear than we would expect if the two were proportional. What lesson do we learn from this very elementary exercise? It is that human origins is a fascinating topic to most people, certainly more fascinating than most scientific topics, which makes sense, since we are all, like the Jew of Venice, human. Sooooo, when someone starts to wax scientific about the origins of human language (which is, after all, the most salient of human activities/attributes), this is bound to attract attention, especially when the waxer is a respected scientist, as Cavalli-Sforza is (he also has a wonderful and impressive sounding name). The same goes for all the proto-world stuff. Note that I am not commenting on the scientific merits of these views, merely on the fact that the public is so interested in the matter. On the other hand, if some of these physical anthropologists are right, then modern humans have only been around for 40,000 years, which is bringing us into range of at least some respectable historical linguistic methods. Now I'll go back to my morphology, happy in the thought that even most linguists find that topic totally uninteresting (so that I get more paradigms to myself).

Message 3: Re: 3.101 Proto-World (Part 2)

Date: Mon, 3 Feb 92 13:48 CST
From: Joe Stemberger <STEMBERGER%ELLVAXvx.acs.umn.edu>
Subject: Re: 3.101 Proto-World (Part 2)

It's interesting that so many linguists don't like the work that attempts to get at Proto-World, and that many of us actually get angry about it. As Grover Hudson points out, such reactions may suggest to outside parties a dogmatic attempt to suppress radical opinions that disagrees with standard, dominant ideas.

I think we need to point out clearly to others WHY we don't like it.

It isn't that we necessarily think that the answers the Proto-World people have come up with are wrong, that they disagree with particulars that we all believe. OK, maybe some historical linguists have different ideas about language families, and there's some of that. But most of us don't work in historical, and we have no real axes to grind about what answers are right and what answers are wrong.

The Proto-World IDEA is exciting and interesting. If it could be demonstrated, most of us would be happy with it.

So why don't we like the Proto-World research?

It's the methodology. Basically, we have no faith that the particular methods being used are reliable, that the answers that the researchers are getting are much more reliable that grouping language families at random. Maybe the Proto-World people are right in almost all details --- but we have no reason to think that that is likely.

Grouping languages is trickier than grouping animals or plants. Chance similarities are far more of a problem when comparing languages than when comparing species. So much so that syntactic characteristics are viewed as unreliable guides to grouping languages even over comparatively short time periods. And you have to worry about borrowing between distantly related languages (but not between distantly related species); if you find words in common between two reconstructed proto-languages, it may point to a common ancestral language OR to a period of contact between two distinct ancestral languages, with a lot of lexical borrowing. Some "shared" words may be onomatopoeic. Some "shared" words are random similarities. People in other fields don't realize all these difficulties, or that the problem is so much harder than in biology.

So, we get upset about Proto-World because of the methodology.

Should we express outrage and displeasure about this to the rest of the world? Well, as linguists, we are in general subject to attack from other quarters claiming that our methodology is horrid. The bulk of researchers in Psychology and Communication Disorders are skeptical of results coming from theoretical linguistics, BECAUSE THEY DON'T TRUST OUR METHODOLOGY. They read any randomly chosen article on English syntax and find that it relies on the grammaticality or ungrammaticality of sentences of a sort that they don't even recognize as English. They don't agree with our notions of simplicity in phonology and hence are skeptical of the results (and are reinforced when they look at the analyses of English given by Chomsky & Halle, or by Halle & Mohanan, etc., which they view as crazy). Oh sure, linguistics may have somehow stumbled on the right answer, but they don't think that there's any reason to think that's true, given their complete lack of faith in our methodology. And the rapid rate of change in our theories suggests to many that we are wildly thrashing around in the dark, with no real idea about what is really going on in language. To be sure, some people in those fields don't share that skepticism. And they react well to research inspired by linguistics but using THEIR methodologies.

Some of the ideas that get the LEAST enthusiastic response from these other disciplines are things like innateness and modularity, as linguists use them. (Which, of course, are the things which have been MOST OFTEN written up in the popular press.)

I think that we need to make clear to the world at large that the methodology used for Proto-World studies in weak. But I think that we have just ignored such charges against linguistic methodology in general, and I don't know that we can speak from a solid position ourselves.

---joe stemberger