Re: arguments

Classical Nahuatl does something unusual along these lines, iirc.  The
verb always agrees with the subject, but it agrees with the object
*furthest down* the relationial hierarchy!  So if there's a
benificiary, the object prefix agrees with the benificiary; if there's
none, then it agrees with the indirect object; if there's no indirect
object it agrees with the direct object.  This is typologically very
rare; I know of no unrelated languages that do anything like this.

As for your proposed agreement rules, they look basically okay --
unusual, but okay.  But if you're going for naturalism, consider
something like Nahuatl instead -- agreement with the furthest down in
the hierarchy, which is sorta what your rules do.  That's still not
*usual* -- it's downright bizarre -- but there's precedent, at least!
;)

The last two rules are rather strange.  The second-to-last one because
it agrees with two objects instead of one, as the rest do, and the
last one because it gives up on agreement altogether.  It'd be
slightly more realistic to have either (1) every verb agrees with one
argument, if available, or (2) every verb agrees with two, if
available, or (3) every verb must agree with one, but may agree with
two.

(But this just reflects my personal bias towards naturalism.)

Anyway, it sounds pretty cool.  Good luck!

--
Patrick Littell
PHIL205: MWF 2:00-3:00, M 6:00-9:00
Voice Mail: ext 744
Spring 05 Office Hours: M 3:00-6:00