[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
xorxes:
> la and cusku di'e
>
> > > For example, lu'i would be defined as something
> > > like:
> > >
> > > lu'i Qx = da poi ge Qx cmima ke'a
> > > gi no de poi na du x cu cmima ke'a
> >
> >The second x is unbound
>
> You're right, that's why I didn't want to use a lojban variable,
> I meant it to be just a word replacement. Let's see:
>
> lu'i Q de poi broda = da poi ge Q de poi broda cu cmima ke'a
> gi no de poi na du su'o di poi broda cu cmima ke'a
>
> Or more succintly:
>
>
> lu'i Q broda = da poi ge Q broda cu cmima ke'a
> gi no de poi na broda cu cmima ke'a
>
> > > and the same formula would apply without the quantifier
> >
> >In which case, we are also disagreeing about the meaning of Fx,
> >since F is not being treated as a function
>
> It reduces to a function when the quantifier is outside. In that
> case we have:
>
> lu'i ko'a = le pa da poi ge ko'a cu cmima ke'a
> gi no de poi na du ko'a cu cmima ke'a
>
> > > I'm not sure I see the point of having LAhE at all if they
> > > are transparent
> >
> >When you have true functions, treating them as predicates
> >involves redundant quantification (or some other redundant sort
> >of gadri use). Transparent LAhE avoid this redundancy
>
> Yes, but at too high a cost
On reflection, given that some LAhE patently aren't functions,
I will change my argument and say just that LAhE is the natural
home of functions and that LAhE should not be expected to all
work the same way.
> If I understand correctly, you would have {lu'i ko'a},
> {lu'o ko'a} and {lu'a ko'a} make sense only in the case
> where {ko'a} is a set. So for example:
>
> lu'i lo'i broda = lo'i broda
> lu'i le'i broda = le'i broda
> lu'i ro se cmima = ro se cmima
> lu'i loi se cmima =? loi se cmima
> lu'i ro mlatu = nonsense
> lu'i su'o mlatu = nonsense
>
> Is that correct?
I will sidestep the question, because it now transpires that
all existing gadri are supposed to be quantified, so LAhE +
gadri cannot do what I wanted it to.
No, hang on. Here's the idea:
Collective: LAhE-collective mo'ezi'olo'i broda
Kind: LAhE-kind mo'ezi'olo'i broda
where LAhE-collective/kind are transparent. IOW, where LAhE
is used instead of a new gadrow, the LAhE is transparent and
takes mo'ezi'olV'i as its argument.
Even this wouldn't work if lV'i is extensionally defined, as
John thinks it is.
All of which just makes me yet more fed up with SL and the
current debate. We have got the point where we have to start
making decisions, either for SL via BF, or for AL. Currently,
everything is sinking into a quicksand of indeterminacy, and
all we have achieved is showing that virtually every
conceivable issue is shrouded in uncertainty due to the
incompleteness of the current prescription.
--And.