-------------------------------------------------------------
Comments:

 I changed intransitive, transitive and ditransitive verbs to take a [
 DEF-OPT + ] subject, since dropped subjects appear to have a definite
 interpretation. I left DEF-OPT underspecified on verb complements,
 since the interpretation as definite vs. indefinite depends on context
 (and perhaps on the specific verb).

The head-opt-subj rule is not sensitive to DEF-OPT and always
gives OPT+, because it is hypothesized that this is universally
the case.

 The "mod-order" type has two subtypes, which are used as values for
 the MOD-ORDER feature: (1) "dem-or-num-required" for when a classifier
 is used and needs to be preceded by a demonstrative or numeral, and
 (2) "dem-or-num-not-required" for all other situations.

I still think it's a better idea to have the classifier combine
with the demonstrative or numeral to its left and then have that
consituent act as a modifier.  (Among other things, this fits with
my intuition that the classifiers themselves are not contributing
any semantics.)

 But a couple of ungrammatical sentences, that have bad modifier order,
 still produce parses:

    *wo3 you3 liang3 zhe4 ben3 shu1
    *yi1 lao3shi1 kan4 wo3


 I have been looking at the parses, but haven't figured out what's
 wrong yet. Something about the constraints is working differently than
 I think it should (it's like some of the constraints seem to be
 ignored).

The MOD-ORDER value of the head daughter is not in fact getting
matched with the MOD..MOD-ORDER of the non-head daughter.  Although
the value of MOD is a list of synsems, the synsem on that list is
not in fact identified with the head-daughter's synsem as a whole.
Rather a bunch of smaller idenities are stipulated.  I'm not sure
exactly why this is, but you can see it in matrix.tdl.  They type is
basic-head-mod-phrase-simple.  


