The appeal of P&P

[Mike Maxwell]

Sergi Balari:
>This appears to me rather different from
>the idea of a universal body of knowledge
>(information) with a number of pre-defined
>options (parameters) that get fixed on the
>basis of external information. Anyway, this
>is not what I understand by a dialectical process.
>Dialectical, to me, is what Lewontin describes.
>Perhaps this what Chomsky has in mind, but, as
>far as I can tell he has never put it in such explicit
>terms as Lewontin.

I haven't read much Chomsky lately, but I wouldn't be surprised if he wasn't
very explicit about how this works.

I'm not sure exactly what you mean by "dialectical" (in particular, on how
the developing organism might change its environment), but there is an
interesting account of how P&P might actually work in acquisition in the
following paper:

    Dresher, B. Elan. 1994. "Acquiring stress systems." Pp. 71-92 in
    Eric Sven Ristad (ed.) Language Computations. DIMACS volume
    17.  American Mathematical Society.

The topic is phonology (stress systems), but one can imagine how similar
things might work in syntax.  Dresher looks at not only what the parameters
might be, but how individual parameters could get set in the absence of
knowledge about all the other parameters, how sequencing of parameter
setting might affect things, and what information ('cues') would be required
in the environment.

                                         Mike Maxwell
                                         Mike_Maxwell at sil.org