Dear Rob, Stefan and Mike,
>What is in "The language instinct debate"? Once again does it dispute the
>observation of a lack of generality in distributionally derived
>representations, or only the innatist conclusions Chomsky drew from it?
The way I read the book, I think it really disputes that observation: in fact Sampson's claim is that a Popperian model of try out new hypotheses on your language data is what a child does while learning his/her mother language, and that the Chomskyan claims of the lack (or shortage) of enough language data to learn from do not hold water.
In any case, I should point out that I consider the two books I suggested (directly to Stefan) essential reading to participate in this debate. "Empirical Linguistics" is one the best attempts to provide a foundational basis for corpus linguistics, while "Educating Eve"(now recast in a new edition as "the 'language instinct' debate", but I read the former one) is essential for understanding the Chomskyan and Pinker ideas and have them empirically falsified (and much more).
You may not agree with the ideas there, but I think anyone interested in linguistics, in language and/or in leaning should read them. Here are the correct references:
Empirical Linguistics: http://www.grsampson.net/BEmpLj.html
The 'Language Instinct' Debate: http://www.grsampson.net/BLID.html
Of course, this is just my opinion, but since I have been (mis)quoted I thought I would make the claim again to a wider audience.
Cheers,
Diana
============
Diana Santos
Linguateca node at SINTEF ICT
www.linguateca.pt
This archive was generated by hypermail 2b29 : Mon Oct 17 2005 - 19:04:03 MET DST