Newsgroups: sci.lang
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!newshost.marcam.com!usc!howland.reston.ans.net!Germany.EU.net!EU.net!sun4nl!mcv
From: mcv@inter.NL.net (Miguel Carrasquer)
Subject: Re: Language and genes
Message-ID: <D02y75.K2n@inter.NL.net>
Organization: NLnet
References: <634@percep.demon.co.uk>
Date: Wed, 30 Nov 1994 12:18:40 GMT
Lines: 71

In article <634@percep.demon.co.uk>,
rmallott <rmallott@percep.demon.co.uk> wrote:
>Miguel Carrasquer <mcv@inter.NL.net> writes: [abbreviated]
>
>>Language change is random, just like biological change. 
>
>Distinguish in both cases between the sources of the variation
>and the genetic or linguistic change which actually takes effect.
>The genetic variations which survive and produce a change in
>the gene pool are by no means random; their survival is tested by
>the genomic, physiological and ecological environments in which
>they occur.

The sources of biological change are random: some changes have
a positive effect on the phenotype and proliferate, some
changes are detrimental and disappear.  This is traditionally
called "survival of the fittest".  The great majority of changes
are neither detrimental nor advantageous: they just happen.
(However, when external circumstances change they can prove to be
detrimental or advantageous after all).

In the case of language, no biological advantage/disadvantage
can be derived from language change.  The environment in which
language is "tested" is socio-political.  It is socially advantageous
to speak Latin if you're a citizen of the Roman Empire, for instance,
and quit speaking Celtic.  But the changes that transformed Latin
into French, or Spanish, are still random.  The social environment
didn't care one way or another: anything might have happened
(and it did).

>To say that linguistic changes carried through into 
>the linguistic enviroment in which they occur are random is no more 
>than to deny almost as an axiom the possibility of ever understanding 
>the nature or processes of language change. Randomness cannot be
>proved.

What is there to understand?  You can list (in principle) the changes 
that took place in the transformation Latin => Spanish.  But there are 
no "meta-rules" (no interesting ones at least); everything is in the 
details.

>>Even more so, since linguistic 
>>change is _faster_ than genetic change. ...  
>
>But not necessarily than change in gene _frequencies_ within a population.
>The structures and functioning of language are modelled in the human
>brain, its realisation is achieved through the physiological
>structures of articulation and audition. Genetic evolution operates
>on human brains, on the articulatory organs, on the organs of hearing.
>For language changes e.g.in phonology to come about, they must 
>be accommodated to the neural and physiological structures that
>genetic evolution produces and changes.

???? I'm afraid I don't understand this at all.  There's nothing
in my neural and physiological structures that prevents me from
learning any language spoken in the world today.  Precisely
because of the slow nature of genetic evolution, our brains,
articulatory and auditory apparatus have been essentially the same
for the last 100,000 years or more.  From the level at which
linguistic change operates, these things don't matter at all,
and can be regarded as constants.  Language is what Richard Dawkins 
would call a "meme" (in fact it's the most powerful and basic "meme" 
of them all).  Our genes, by giving us brains and mouth, have
been instrumental in creating the language "meme", but they
(the genes) don't play any role at all in the "memetic" evolution 
affecting language.

-- 
Miguel Carrasquer         ____________________  ~~~
Amsterdam                [                  ||]~  
mcv@inter.NL.net         ce .sig n'est pas une .cig 
