Mark Volpoe: Numeral Quantifiers in the l-node hypothesis

[Martha McGinnis]

Dear Listers,
    Greenberg identifies the universal tendency for
obligatory Number morphology to be in complementary
distribution with Numeral Quantifier systems. English
and Japanese are examples good examples of the
contrasting systems. One therefore expects that if NQs
are analogous  to Number, as Greenberg's
generalization seems to suggest, NQs are phenomena
associated with functional heads.
    Number however, is able to be characterized in
binary features, but not NQs. Far from it, there seem
to be approximately 27 different classifiers (which
suffix to the numeral) in current usage in modern
Japanese. Additionally, these classifiers are closely
identified with semantic properties of the Nouns they
are used with, e.g., '-mai' is used for flat objects,
pieces of paper, dishes, towels, etc. '-hon' is for
thinnish, elongated objects such as pencils, trees,
cucumbers.
    In traditional approaches to morphology, one would
assign each Noun to a Noun Class, based on the NQ, and
enumerate the 27 or so classes. Has anyone been
working on NQs in Japanese or other languages within
the DM-framework? Anyone have an intuition or a
conviction about how to deal with the idea?
                      Thanks,---Mark

__________________________________________________
Do You Yahoo!?
Yahoo! Finance - Get real-time stock quotes
http://finance.yahoo.com