[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
Nick:
> I spent three months advocating tu'o for Kind.
That's a good thing then, isn't it, because it should make the agreement
we have been approaching more robust.
> The reason I gave up on it was, a, it was also coopted for Substance,
> and b, generalising it from Mr Individual to Mr Anything Else was
> acutely painful. Remember Mr xodium?
>
> Misterhood needs to be orthogonal to ontological type. While glorking
> tu'o will do for individuals, it will not do in the general case. How
> do the XSs come up with Mr xodium? (I do not have fond memories of
> Jorge's attempts to kludge around the subject.)
>
> Is Mr xodium {tu'o loi marjrxodiumu}?
tu'o lo tu'o marjrxodiumu
> Is Mr Couple {tu'o loi re lo prenu} -- or in your XS, {tu'o lo re prenu}?
Yes.
> And is this really the way you want to go, rather than a general LAhE?
Yes. XS is simpler, more elegant and more consistent, and the only change
to CLL it requires is to the interpretation of inner PA from something
unuseful and unused.
Plus, there are no alternatives currently on the table.
> The bpfk is supposed to decrease the kludges not increase them? For the
> thousandth time, no, that is not the bpfk's mission. The bpfk is to get
> things clarified and/or sayable that haven't been. You want elegance,
> take it to LoCCan. I should not have to keep saying this.
XS is not kldugey, requires only a single change to CLL, is perspicuous.
makes everything sayable, and drastically eases the task of sorting out
the rest of the gadri.
> I popped in to the wiki, saw the equation lo = Kind, and nearly walked
> out again. No, lo != Kind. I will not accept that. I will only accept
> lo = Individual or Kind of Individual. gadri usage may be confused, but
> some parts of the gadri system are well defined, core, and central to
> the language.
Since {lo} has no meaning of its own in CLL, lo = Kind is not a change
to CLL. But if you accept lo = individual constant, that is good enough,
since there are no rivals to the equation Kind = individual constant.
> LAhE mess with quantification already,
That hasn't been agreed. Current LAhE work by means of handwaving magic.
I.e. they don't work.
> and the sense of re mikce is
> already pretty damn close to that of LAhE3 re mikce: the possible
> denotation remains "all doctor pairs". LAhE3 merely suspends the
> exporting of the prenex. If you insist on doing tuples differently,
> fine, I don't really care (though I do care about Mr Substance); but I
> don't buy your counterargument.
You could kill all the unacceptable inconsistency in this by moving LAhE3
to LE. Then it would become comparable to XS, and you wouldn't have to
have a rule saying that quantification is not quantification when marked
on a sumti following LAhE3.
Alternatively, the abbreviation rule whereby {re lo broda} can be shortened
to {re broda} could be made context sensitive so that it doesn't apply
after LAhE3 and {LAhE3 re broda} is instead an abbreviation of {tu'o LAhE3
tu'o LE-kind re broda}, where LAhE3 merely copies the meaning of its
complement, and LE-kind is a 'virtual' gadri that exists in logical form
but is not itself lexicalized.
So, I don't reject LAhE3 tout court. There are ways of getting it to work.
But are they better than XS?
--And.