Re: Odp: Proto-World, Nostratic, etc.

----- Original Message -----

From: Mark Odegard

To: cybalist@egroups.com

Sent: Thursday, November 11, 1999 3:13 AM

Subject: [cybalist] Re: Proto-World, Nostratic, etc.

Piotr and I are not in really in disagreement [...]

Well, we ARE partly in disagreement, and there's nothing wrong with it. In science, one grows accustomed to being in friendly disagreement with one's colleagues practically all the time; wise people do not expect everybody to see eye to eye with them. If they tried to avoid public disputes and clashes of opinion, science would degenerate into private speculation.

With 'proto-World', I think language was invented just once. You have to be immersed in language from birth (even before birth, it's been suggested). Whichever way it developed, once a group of our ancestors had real language, the adaptive value of language, the things it did for them made them infinitely more successful than any of there pre-linguistic relatives. Without real language, you cannot exchange information except by physically showing someone else what it is you want to tell them. With language, you can tell your grandchildren what your grandmother told you her grandmother said about food and water sources at some distance. With language, you can discuss hunting and gathering stragegies, not just the tried and true ones, but some innovative ones, ideas that require a high level of co-operative pre-planning. With language, you can fed you and yours much better than those without language. More food, healthier food, means more babies, healthier babies. You've got a population explosion. Those who were not co-opted by the speaking humans were marginalized into extinction. I'm saying it's likely some pre-linguistic humans were incorporated into the speaking group, but find it difficult to believe full-fledged language developed more than once. It is conceivable that pre-linguistic humans hung on until rather later, perhaps even into historic times in isolated corners of the world.

It is obvious that the development of language offered its users a selective edge in the further evolution of the genus Homo. H erectus and H. neanderthalensis lingered on until quite recently (ca. 30,000 yrs before present, i.e. after some 70,000 yrs of coexistence with H. sapiens in Eurasia), perhaps occasionally interbreeding with H. sapiens; but they eventually died out. Other less known and less securely defined species (H. heidelbergensis, H. antecessor) never achieved much evolutionary success. Fully-fledged linguistic skills are often supposed to have been the difference that mattered, because in other respects there is no obvious advantage that H. sapiens could have over its cousins. But when you speak of H. sapiens as "infinitely more successful", and of a "population explosion", one would like to see the palaeontological evidence for that. Let me reiterate that the biological monogenesis of language is logically independent of, and thus compatible with, its cultural polygenesis (with different, though typologically constrained, forms of communication developing parallelly in different communities making up the primitive population of H. sapiens). I find it very difficult to reconcile primordial monolingualism with what we know about the speciation of H. sapiens and the social life of early humans. Even at present, the more primitive conditions you have, the more diversified the local languages are -- think of New Guinea or South America, where isolates and very small families are the norm. Sub-Saharan Africa is a vast region, with enough Lebensraum to support hundreds of languages well before the first H. sapiens stepped out of Africa.

[...] The most graceful way I've heard the issue described is that, for all the problems, it's now unthinkable to consider Afro-Asiatic in any other way.

"Afroasiatic" is no doubt a convenient label (by the way, the non-hyphenated spelling is preferred by most specialists, while some still speak of "Hamito-Semitic" and in the Russian tradition the term Afrasian is used). But convenience is not the same thing as validity. In biology, it's convenient to talk of "fish" or "reptiles" even if biologists know that these "classes" are not valid taxa from the point of view of a strictly phylogenetic classification. The general public, of course, is not even aware of such problems. Whether Afroasiatic is a single family or not depends on how successfully all the Afroasiatic data can be correlated. I can assure you that many eminent etymologists working in the field of Afroasiatic have questioned its status as a family. Some prefer to call it a "phylum", just to express their reserve. Some contest even the validity of smaller-scale groupings, claiming that e.g. Cushitic is an areal league of a few genetically unrelated families.

One thing is sure: African and Afroasiatic reconstruction has not reached the degree of certainty and acceptability even remotely like that enjoyed by Indo-European (or Uralic, for that matter). For many reasons it is an exteremely difficult domain, and progress in it requires patience; there are no short cuts in such cases. Attempts to solve all its problems at one fell swoop will always fail.

I simply do not understand how people can speak confidently of reconstructing 2000 Nostratic etyma if the number of reliable Uralic reconstruction does not exceed two hundred and there are problems even with Semitic comparative dictionaries.

The Eastern Bloc scholars are less well known in the West. Part of this was do to the previous political regime. Another part of it is the lamentable fact that if it's not published in English, in an English-language journal, it essentially doesn't exist. [...] I've heard of Dolgopolksy, but not of Shevorshkin. I confess to being a self-taught amateur who has come to IE studies in only the last couple of years.

A lot of it HAS been published in English, though Illich-Svitych's works haven't. But the McDonald Institute for Archaeological Research (Cambridge, GB) has published Dolgopolsky's book The Nostratic Macrofamily and Linguistic Palaeontology (1998), with a long introduction by Colin Renfrew. It's nice to know that somebody hasn't heard of Vitaly Shevoroshkin. He's a kind of press officer for long-range comparison, and he's based at the University of Michigan, Ann Arbor, MI, USA. (Aharon Dolgopolsky, incidentally, works at Haifa, and of the great names of the Moscow Nostratic school only S. A. Starostin actually lives in Moscow, I believe). It was Shevoroshkin's conference shownmanship and hyperbolic statements in the popular press a few years ago that provoked an American colleague (Eric Hamp, I think) to remark: "He's already found out what American salesmanship is all about".

This is the the criticism consistently leveled against all such studies. It's not that there is anything wrong with the idea of Nostratic, etc; in fact just about everyone admits that something like this is indeed the case.

Well, I'm one of those who do deny there is much of a case there, and I'm in a good company. I've seen Dolgopolsky's most recent etymologies, as well as those published by Bomhard, and I remain unconvinced -- not because I'm stubborn and don't wish to see the light, but because the putative links are really very weak, probably indistinguishable from chance similarity, given the vast number of languages being compared. The belief that some version of the Nostratic hypothesis must ultimately be true whatever the weaknesses of its demonstration results from tacit assumptions which I find questionable -- e.g. the extension of the IE model of language spread and diversification to pre-neolithic times. I'd say that the Indo-Uralic case is marginally better, or at least the comparison is more rigorously constrained, being done pairwise between reconstructible protolanguages. Still, even the hypothesis of Indo-Uralic unity is based on may-have-beens and "intuitive judgements" which are not equally acceptable to everyone.

[...] At best, all that historical linguistics can do today is reach back perhaps 7500 or so years, and then only when there is a lot of evidence to go on (as is the case with Indo-European).

No disagreement here. I wouldn't even insist on imposing a time-depth limit: it's the quality of the evidence that counts, not how far back the recontstruction can reach. What I'm against is not "deep reconstruction" as such (I'd be happy it it were possible) but the exaggerated claims of its proponents. Interested colleagues from other disciplines (archaeologists or geneticians) tend to take those widely circulated claims at face value and try to correlate their own findings with them, only to feel duped when they find that what they bought as gospel wasn't worth the paper it was printed on. They may refuse to take linguistics seriously afterwards.

Piotr