pc answes

["John E. Clifford" <pcliffje@crl.com>]

        OK, so read _ro_ for _su'o_ throughout the examples (_lo_ has
changed so often
that I can't tell if I am a tick behind or a tick ahead of the curve).
The absurdity of the
standard rules remain.
        Back to the main point.   Assuming that _ci nanmu cu pencu ci
gerku_ is the nine-
dog sentence, we need independent _ci_s for the three-dog one.  But a
look at the logical
form shows that the prenex forms of "three" are independent and capable
of going in any
order relative to other numbers and to the particular  The universal
creates a problem
here, but we can deal with that by simply using the right order in the
prenex.  Thus, a
fairly light cost, we can use _ci da poi nanmu ci de poi gerku [whatever
the hell the
prenex comma is] da pencu de_ or maybe even _ci nanmu ci gerku [comma] ny
pencu
gy_ .  Or maybe even _[leaper] ci nanmu cu pencu [leaper] ci gerku_ (I
can't even
remember what the x-perimental form of leaper was, let alone what it
might have been
finally lexed as).  We might even get by without the first [leaper], on
the pragmatic
ground that it is already set at the head of the sentence in terms of
processing -- but that
would need some experimenting (in particular to assure that it does not
get the nine-man
form).  None of these will work if one of the objects involved is defined
in terms of the
other, but then the possibilities of their being independent is cut off
anyhow.
        If we take the bare form to be the three-dog one, I do not see an
equally easy way
to get to the nine-dog form.
pc>|83