Re: mo'e

[jorge@PHYAST.PITT.EDU]

la lojbab cusku di'e

> >{la'e mo'e da} is not grammatical.
> It should be grammatical!.  Look at the last rule in operand_385.  haven't
> checked the parser, though.

You are right, it is grammatical. I felt that it had to be ungrammatical
because it is not a sumti. I have no idea what kind of operand is
{la'e <operand>}. It still feels ungrammatical to me, even if it is
accepted by the parser.

> 1) li ci is not a "description" - no selbri involved.  So you HAVE to do
> something else.

I didn't mean to use "description" in any technical sense. {li ci} is
a sumti, and like all sumti it has a referent. Its referent is the number 3.
For apples, {mo'e <sumti>} takes the number of referents as the number
and the type of referents as the dimension. {li ci} is just like {lo namcu},
but more specific. If {mo'e lo namcu} is "one number" then {mo'e li ci} has
to be "one 3". Otherwise, {mo'e} has different definitions for different
sumti.

> 2) In the case of the description, the number of referents PLUS the dimension
> is "the number".  It happens that in ordinary arithmetic, the result of
> operating on two dimensioned numbers happens to be the same as if the numbers
> were separate from the dimension.  This iss not necessarily the case for all
> mathematical operations.

When is {mo'e <sumti>} NOT a dimensioned number? Only for {li} type sumti?
Is {mo'e da} a dimensioned number?

> So I sstill say that in both cassess, you are "converting the sumti into a
> quantifier" - a ssingle quantifier, which may or may not have multiple
> components, multiple types, etc.

I'm not sure what you mean. Is it true that the number of referents of the
sumti gives the resulting quantifier number? If yes for apples, but no for
{li ci}, then we are using different definitions in each case, and I don't
know which one to choose for {mo'e da}, because {da} could be standing there
for {li ci} as well as for {lo plise}.

> Obviously, what I was trying for was to convert set membership into a
> "quantifier" (i.e. something that can be operated on by a mex operator).
> I am looking for a selection from a set of 7 numbers, andthe quantifier that
> results will not actually be used in a mex expression, but will be used to
> create an ordinal expresssion.  Exactly what conventionss need to be used to
> make this clear are not esstablisshed, but I still say that mo'e is doing the
> same thing in all of these cases.

Could you explain what {mo'e da} means then? {mo'e} is not doing the same
thing when the sumti is of {li ci} form as when it is of {lo plise} form.
In one case it takes the referent of the sumti as the quantifier. In the
other it takes the number of referents of the sumti as the quantifier.
They are different things.

> I guess we have some kind of different underssstanding of "ssame thing", when
> it applies to objects of different types.

I guess we do. I don't think they are objects of different types, they are
all sumti.

> The conversion of a description
> into a quantifier may use somewhat different rules than the conversion of
> a quantifier ssumti into a quantifier, which may be different form the
> conversion of a free or bound variable into a quantifier.

Why? Why shouldn't there be one rule for the conversion of sumti into a
quantifier?

> These are all sumti,
> but they are sumti of different types, and they are seldom in the language
> considered to be semantically similar for purposess of interpretation.

Seldom? They should always be treated equally. {ko'a}, {ri}, etc can stand
for any of them, why shouldn't they follow the same rules?

> They
> are similar only in grammar, unlesss we find it conveniemt to make them
> ssimilar in interpreattion.

Then what is the meaning of {mo'e da}? I still don't know the answer.

Anyway, here I have another reason to completely ignore MEX. It doesn't
seem to fit well with the rest of the language if {la'e} can flag something
that is not a sumti and there are special rules of interpretation for
each different sumti.

Jorge