[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: [jboske] Quantifiers and lo/loi

> la and cusku di'e
> >An XS solution not incompatible with SL is to say that
> >quantifiers are underlyingly of the form {PA fi'u roPA},
> >but that bits of the expression can be omitted and left
> >to glorking 
> I like that very much, but I'm not sure about the compatibility
> if fractionals in SL are supposed to be sizes of bits and not
> true quantification over bits 

There is not necessarily any "supposed to be", so long as we
can tell a story that is not downright inconsistent with CLL.
(I am of course discussing the BF solution here.)

As for fractionals and loi, there are various possible stories.
Regarding loi, the story could be that it is simply the form
that gadri take when quantified by a fractional. Or the story
could be that it means that it has cardinality &-ma'u (= the
number of jbomasses, not the number of constituents of the
jbomass). With regard to the fractional, either it quantifies
over bits, in which case {piPA loi} would be short for
{Q lo(i) piPA loi} (i.e. a jbomass containing piPA bits of), or 
else (which is incompatible with loi = thing of cardinality 
&-ma'u) it really is a cardinality and means "a piPA of a 
single", in which case it would be short for {vei pimu ve'o 
fi'u ro}.

I don't think we should rush into decisions among these
possibilities -- things have been too frenetic here & need
to settle down for a bit. But my general point is that, thanks
to Nick's work, it seems to me possible to construct a gadri
system consistent with some interpretation of CLL yet also
internally consistent and comparatively nonarbitrary.