DM-LIST Query: Gender in DM

[Martha McGinnis]

Fellow DM-interested people:

I have a question for which I seek references and/or comments. Many
thanks in advance:

DM distinguishes between syntactically relevant features and
diacritics. The former are inserted early and feed the syntactic
computation. The latter are associated arbitrarily with individual
roots, and thus come in with vocabulary insertion, and thus cannot
affect syntax.

Gender seems to be a diacritic introduced by vocabulary insertion, in
that it is (at least in some languages) an unpredictable feature of
individual (noun) stems. In Italian, libr- 'book' is masculine (il
libro) while scatol- 'box' is feminine (la scatola). In Russian it's
the reverse, knig-a 'book' (fem) jashchik- 'box' (masc).

However, there is gender agreement outside the Noun-word, e.g., with
the determiner, adjectives, and (in some languages) even with past
participles and verbs. This makes gender look syntactically active,
hence pre-vocabulary insertion, hence NOT a diacritic feature
arbitrarily associated with vocabulary insertion. Across words looks
like it should be outside the domain for contextual allomorphy (or
does it?)

I don't think making the distinction between gender and declension
class helps here (though I believe it is independently necessary).

Thoughts, pointers, etc... ?

Thanks,

-Jonathan

_______________________
Jonathan David Bobaljik
Department of Linguistics
McGill University
1085 Dr. Penfield
Montréal PQ H3A 1A7
CANADA

tel: (514) 398-4224
fax: (514) 398-7088

http://www.mcgill.ca/linguistics/faculty/bobaljik