Re: Cognitive Linguistics, "The Language Instinct", and High-Functioning Autistics

I agree with David Peterson I think. One thing that really bothers me
about the Chomskian theory of built in language is that:

(i) The poverty of stimulus argument has never been properly supported
by evidence, despite claims to the contrary

and

(ii) Chomsky insists on taking an all or nothing view. Even if the
poverty of stimulus argument is correct, that is not proof that a
complete grammar is built in: what is wrong with the view that children
may be biased towards language, have some basic cognitive enhancements
(like an area of the brain good at tree processing to guide them in the
right direction), but don't have the whole damn thing built in? There
are infinitely many shades of built-in-ness that are obscured by this
naive assumption that if human beings are biased towards language
learning from birth then it must all be built in.

I myself take the minimalist view of built-in-ness. Looking at nature,
you rarely find examples of animals having everything programmed in..
nature tends to take the elegant route, and that means doing the least
work possible. I think human beings may have extra general cognitive
structures of some sort that bias them towards language learning, but I
don't accept the bizarre "everything is built in and babies just figure
out what the right settings for their grammar are" approach, which not
only jars with my intuition but does severe damage to the continuous
variation of language by trying to present things as the result of
discrete settings.
 So I think babies have a bias towards learning to communicate (not
necessarily by speaking though.. I suspect some children grow up with
sign language), but the central part of that learning process is social
interaction in the form of talking to reinforce that bias and provide
material to learn from.

Reply