[hist-analytic] Filters, Chomsky, Composition

Baynesr at comcast.net Baynesr at comcast.net
Thu Dec 16 16:39:30 EST 2010

In "Filters and Control" (Linguistic Inquiry, 1977) Chomsky and Lasnik layed out a linguistic theory where filters and not rules of construction (transformation rules, etc.), controlled the possibilities of construction. In semantics, however, it is difficult to see, but worth considering, how filters might be used. What is particularly attractive in the filters approach (which in linguistic theory is not current) is that the idea of sentence "novelty" (essential to Chomsky's "Cartesian Linguistics") is to be accommodated in a generative approach. The two notions, at least, appear to work against one another. My rejection of compositionality may require a filters approach that is generative requiring ambiguity; but I don't accept the possibility of a generative semantics in the sense that Chomsky seeks to realize a generative syntax (or phonology: Chomsky/Halle). I think Montague came closest to the generative approach to syntax, but the problem of "novelty" was not an issue, and compositionality was accepted. 

Steve Bayne 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://rbjones.com/pipermail/hist-analytic_rbjones.com/attachments/20101216/b4da2e42/attachment-0002.html>

More information about the hist-analytic mailing list