[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
On Sat, 14 Dec 2002, Jordan DeLong wrote:
> On Sat, Dec 14, 2002 at 04:08:10PM -0500, Invent Yourself wrote:
> > On Sat, 14 Dec 2002, Jordan DeLong wrote:
> >
> > > On Sat, Dec 14, 2002 at 03:46:54PM -0500, Invent Yourself wrote:
> > > [...]
> > > > A few hours ago, Jorge wrote
> > > >
> > > > i na vajni fa le du'u makau catra la lauras
> > > > It doesn't matter who killed Laura.
> > > >
> > > >
> > > > Had we an identity abstractor Q, with lambda ce'u, then we could have said
> > > > {na vajni fa le Q ce'u catra ly.} We weren't given one, which is why I use
> > > > {su'u ce'u broda keibe lo kamse'i}.
> > >
> > > That would mean "The person who killed laura is not important". It
> > > would not mean "It is not important who killed laura".
> >
> > su'u...kamse'i is not supposed to return the person, but the identity of
> > the person. If it returns the person, it's useless! My self-ness is my
> > uniqueness, but it's not me. I have hair, my self-ness doesn't.
>
> I see; So I was confused over what you mean by identity (I was
> thinking in the math sense where the identity just returns itself
> (1 * 4 = 4, etc)).
>
> So; what *is* the identity of something, if not the thing itself?
> In some logics it is viewed as the class containing only that thing,
> but I don't think that works here...
Maybe it's a quality that only they have. Bill's identity is ka ce'u me by.
> > > Identity abstract is nice (and we apparently have 3 proposed ways
> > > to do it: seka, poi'i, jaika), but it's not what you mean.
> > >
> > > [...]
> > > > So jaika is like the seka, which is actually pretty useless, since it does
> > > > not provide identity (at best it, like every selbri, allows us to
> > > > narrow-down the range of possibly sumti) and usually is used by people
> > > > desiring jei. (I can expound on that last point if needed.)
> > >
> > > I don't understand that at all. What does se/te/ve ka have to do
> > > with jei?
> >
> >
> > My redness gets rendered as ka mi xunre (lacking a ce'u: no good) or ka
> > ce'u xunre kei be mi (redness, and stick mi in the ce'u). It should be
> > rendered either as jei mi xunre (the amount of my redness) or li'i mi
> > xunre (experience of my redness)
>
> No.
>
> "ka mi xunre" is precisely the same as "du'u mi xunre", because
> there's no places for ce'u to go (it would normally be "ka mi xunre
> ce'u" but there is no x2 for xunre).
You can't trick me into defending ce'u-less ka! But this is not how it was
interpreted the CLL, or the older generations who liked it.
> "ka ce'u xunre kei be mi" is precisely the same as "du'u mi xunre" because
> you reduced the lambda variable. (it is (\x: xunre(x))[mi] == xunre(mi)).
This isn't how ka2 was intended by Nick or And or whoever invented it.
> "jei mi xunre" is the truth value of "I am red". In a lot of
> contexts this will simply be a T or an F. This is the same as "ni
> mi xunre kei be lesi'o jetnu". This is *not* affected by any
> concerns as to how red you are.
>
> What you *actually* want is "ni mi xunre". "ni" is a more general
> version of "jei" which allows you to use whatever scale makes sense
> instead of just the scale of truth of the proposition (in this case
> it's probably a scale of luminousity or whatever). To be frank 90%
> of the usage of "jei" which I see from you should be "ni", because
> you're almost never actually talking about truth values, but rather
> the extent or degree of whatever it is (which has nothing to do
> with the truth of a proposition[1]).
I agree that my use of jei "should" be ni. But I've already explained why
I prefer jei, why I think jei gives physically meaningful values instead
of useless numbers no one needs, and that I believe that "he is tall" is
truer statement for a 6-footer than a 5-footer.
> [1] even in fuzzy logics this has nothing to do with anything. My
> understanding is that a logic with infinite truth values ranging
> 0-1 considers the value of the expression to be a measure of our
> certainty of its truth (or whatever). It has nothing to do with
> whether the thing is 10lumens brighter or whatever. (I have no
> idea how much a lumen is, btw).
An alluring argument, but compare it to the single example of non-boolean
jei in the book, in 11;6, and try to make it fit. It doesn't make sense to
decide that George is 58% likely to be a criminal, but rather, on the
scale of criminals, with 0 being the innocent and 1 being pedophile
cannibal holocausters, George rates a 0.24.
Also, the CLL offers the caveat that the details of jei have yet to be
worked out. I am working them out.
It would be nice to have an abstractor that returned the statistical
likelihood of the bridi.
--
jipno se kerlo
re mei re mei degji kakne