Newsgroups: sci.lang
Path: cantaloupe.srv.cs.cmu.edu!rochester!cornellcs!newsstand.cit.cornell.edu!portc01.blue.aol.com!news-peer.gsl.net!news.gsl.net!hunter.premier.net!news.mathworks.com!howland.erols.net!netcom.com!netcom16!alderson
From: alderson@netcom16.netcom.com (Richard M. Alderson III)
Subject: Re: Invention of Language
In-Reply-To: mcv@pi.net's message of Tue, 05 Nov 1996 21:48:33 GMT
Message-ID: <ALDERSON.96Nov6150920@netcom16.netcom.com>
Sender: alderson@netcom16.netcom.com
Reply-To: alderson@netcom.com
Organization: NETCOM On-line services
References: <55lmbo$g2m@scream.auckland.ac.nz> <seagoat.677.01B9F229@primenet.com>
	<55octl$3nh@halley.pi.net>
Date: Wed, 6 Nov 1996 23:09:19 GMT
Lines: 44

In article <55octl$3nh@halley.pi.net> mcv@pi.net (Miguel Carrasquer Vidal)
writes:

>In other words: proto-languages are descended from proto-proto-languages.

No, that's not how it works.

You are making the common mistake of considering "proto-language" to be synony-
mous with "unattested ancestor language."  A proto-language is a *model* of a
language built by means of comparative and internal reconstruction; as a model,
it stands in for the unattested language which is the *real* object of study.
It is the unattested language which is in turn descended from an even earlier
language (not a proto-language).

The methodological mistake, and I do think it a mistake, that most long-range
comparativists make is to try to use individual-language data from too many
sources, rather than comparing proto-languages, to set up their first sets of
correspondences.  (If, indeed, they follow any set methodology at all; I have
no truck with the "mass comparison" folks.)  Because the proto-languages are
already shorthand for large sets of correspondences, detecting patterns shared
by proto-languages is a stand-in for handling massive amounts of data.

(NB: There are those within historical linguistics who believe that comparative
reconstruction is a "lossy" operation, that proto-languages are not proper
comparanda.  Obviously, I disagree.  A number of good studies within Native
American language families had already been done by 1970; cf. Mary R. Haas,
_The Prehistory of Languages_, for discussion and bibliography.)

The result of such a set of correspondences can *then* be applied to individual
languages within the respective daughter families, to see whether further
generalizations, new etymologies, and the like, can be found.  But the result
will still be a proto-language, modelling an unattested ancestral language just
as its original inputs did.

One other source of confusion is the use of the term "proto-language" by Derek
Bickerton to refer to the unrelated concept of the predecessor behaviour(s) to
language in one or more early hominids; it is unclear to me whether he believes
that this stage can be recaptured by any study, whether of creole formation or
of language history.
-- 
Rich Alderson   You know the sort of thing that you can find in any dictionary
                of a strange language, and which so excites the amateur philo-
                logists, itching to derive one tongue from another that they
                know better: a word that is nearly the same in form and meaning
                as the corresponding word in English, or Latin, or Hebrew, or
                what not.
                                                --J. R. R. Tolkien,
alderson@netcom.com                               _The Notion Club Papers_
