[YG Conlang Archives] > [jboske group] > messages [Date Index] [Thread Index] >
Nick:
> > Date: Wed, 18 Dec 2002 12:52:56 -0000
> > From: "And Rosta" <a.rosta@hidden.email>
> > Subject: nitcu x2 (was: RE: RE: lo'ie != lo'ei
> >
> > John:
> >> the x2 of nitcu is an intentional context, just like the x2 of sisku,
> >> and the same fix should be applied: switch to "x1 needs something that
> >> has the property x2". After all, you can need a cure for Ebola even
> >> though
> >> there isn't one, which is exactly parallel with "Galadriel seeks a
> >> unicorn."
> >
> > I say:
> >
> > "the x2 of nitcu is an intentional context, just like the x2 of sisku,
> > and the x2 of djica, the fix used for djica should be applied to
> > nitcu and sisku: switch to "x1 needs the realization of abstraction
> > x2" (x2 to be nu xor du'u)
> >
> > You can need/want/seek an individual, such as Nick or your
> > computer, and rather than representing this as "ka ce'u du/me
> > Nick/your computer", it is better (in the sense of being closer
> > to the underlying meaning) to represent this as "tu'a Nick/your
> > computer", "nu/du'u co'e Nick/your computer"
>
> We have two alternatives, both of which throw Lojban into upheaval:
> allow gadri-like intensionals into siksu, or allow clausal intensionals
> into gismu like djica
In my gimste, djica x2 already is a nu. I would like the predicates
for seeking and needing to work like djica already does. That doesn't
necessarily mean sisku and nitcu have to change, but it does mean
that if they don't change then they aren't proper predicates for
seeking and needing.
> What we cannot allow is the continuation of the current status, which
> allows naive statements like {mi djica lo mikce} for "I need any
> doctor". We are damnably lucky noone's realised until now in the
> mundane world we've been doing this
That's just a mistake. Lojban for Intermediates would point it out
as an easy pitfall -- a lexical one in the case of djica, and
a logical one in the case of nitcu.
> So, to see how feasible things are:
>
> (1) Does anyone know how many intensional preds we have?
Do you mean *gismu*?
> (2) Is anyone's linguistic intuitions going to be terribly hurt if we
> do start saying {mi djica le ka ce'u mikce} and {mi nitcu le ka ce'u
> mikce}?
Yes -- mine.
> (3) How big on the richter scale of gismu adjustment would such a move
> be?
A well-motivated change to place structures that is a big improvement
but fixes something that is not totally broken. So for me, a minor
change, but for Conservatives, a major one.
> (4) What is the compelling current rendering of the gadri-like
> alternative?
There is no real gadri alternative that can make the distinction
between the two nonspecific readings of "I need a doctor".
> (5) Do we want intensional arguments like this used freely anywhere in
> a place structure, or strictly susbcategorised and appearing only with
> a handful of gismu?
Intensional arguments should be expressible only via an abstraction sumti. Most
predicates don't create intensional contexts, but an infinite number
of predicates do. We only need to fix nitcu if we think it should be
possible to use it to say "I need a doctor, any doctor", just as
we already can say "I want a doctor, any doctor".
> Btw, And? Underlying meaning my eye. English does intensions by
> nominal, Lojban by clause. They're different. Underlying meaning? We're
> making up the underlying meaning; it's all construct. That's not
> underlying meaning, that how English happens to do it. And even if all
> languages in the world say "I need a doctor" like that, rather than "I
> need that there be doctoring" (and I doubt it), why should that mean
> Lojban should?
This is silly. English syntax *doesn't* reflect the underlying meaning,
and it is precisely where Lojban is like English that it too fails to
reflect the underlying meaning. Yes, we define the underlying meaning,
but if the underlying meaning is defined in terms of the *concepts*
'seeking/needing/wanting' (NOT the English *words*), then we can
translate them into logical form and thereby see how Lojban should
express them.
--And.