Alec Marantz: phonological arguments for late insertion? (reply to Martha McGinnis)

[Martha McGinnis]

Along the lines that Piotr is suggesting, you might look at Bruce
Hayes's article, "Precompiled Phrasal Phonology," in _The
Phonology-Syntax Connection_ Inkelas and Zec, eds., U of Chicago
Press, 1990.  His solution to preserve lexical phonology is to
"precompile" the results of a "lexical" phonological rule in the
lexicon even though the context of the rule won't be met until the
syntax.

>On Fri, 16 Nov 2001, Martha McGinnis wrote:
>
>>I'm looking for references to anything that gives phonological
>>arguments <italic>for</italic> late insertion or, specifically,
>><italic>against</italic> Lexical Phonology.
>
>How about looking at clitic constructions which undergo 'lexical'
>processes, although they are created *after* syntax?
>
>Obviously, you can postulate that some lexical rules reapply *in the very
>same fashion* in *restricted contexts* in the postsyntactic component, or
>postulate some kind of equivalent of the late Clitic Group in order to
>make such contexts appear less restricted (and hence more natural for
>postsyntactic application), but this looks exactly like patching the
>theory in order to accomodate such paradoxical constructions. So if you
>don't decide to patch Lexical Phonology, then Late Insertion and its
>consequence: 'lexical' phonological applying exclusively postsyntactically
>seem to be a nice solution to the above-mentioned paradox.
>
>(I'm unable to attribute this idea to any particular work; I used it in my
>diss with a feeling that it 'has been around' for a while.)
>
>Wondering if this can be of use to you,
>
>Piotr
>
>--
>Piotr Banski
>University of Warsaw
>bansp at bigfoot.com

--
marantz at mit.edu