WikiDiscuss

---

wrote:

> --- John E Clifford wrote:
> > --- Jorge Llambías
> > wrote:
> > > Mentioning something immediately introduces
> it
> > > into the discourse.
> >
> > But, in order to mention it and thus bring it
> > into the universe, it has to be somewhere
> already
> > (in some sense).
>
> Well, if it was somewhere already, we could
> define that
> somewhere as the one true universe of discourse
> fixed
> once and for all for every utterance in every
> context.
> I don't think there is such a somewhere.

Fair enough. This is mainly just terminological,
universes of discourse are normally given in
advance and then references are assigned within
them. If you want to build the uiniverse along
with the reference, I think that will work out
OK. But it does have one off-putting
consequence: The person who answers "What's in
the box?" with "Nothing" has given on xorxes'
view the right answer, there is nothing in the
universe of discourse in the box (I'm assuming
that air molecules have not been talked aobut
recently, since that would usually cue the
answerer to use them in the answer). Thus, the
response pointing to the air molecules in the box
(we do agree that they were there all the time
don't we?) is not an acceptable correction (and
by the way bringing something from the universe
into the light) but a piece of Gricean dirty
pool, changing the game in mid stream. Not a
move we ought to be recommending.
> > In particular, suppose I say
> > "some cows" or "there are cows that," what
> cows
> > have introduced? the ones that satisfy the
> rest
> > of the sentence?
>
> No, quantifiers don't refer. Saying "some cows"
> "all cows" or "no cows" just introduces cows.
> {ro lo bakni}, {su'o lo bakni}, {no lo bakni}
> and {lo bakni} will all require {lo bakni} to
> have referents in the universe of discourse.
> The quantifiers just quantify over those
> referents,
> they don't introduce them.

Were the cows there before or not? If not then
the use of the expression must introduce them, if
they were there already, where are the limits? I
pass over our differences about what "refers"
means.

> > But the rest of the sentence
> > comes after the introduction — the cows have
> to
> > be there in order to go on. And what if I am
> > wrong? I have introduced cows apparently but
> none
> > of them have the properties involved. So,
> maybe
> > by mentioning cows I introduce the lot of
> them.
> > But then, what is excluded — I mentioned
> > physical objects too, so are all of them in
> or is
> > it specific?
>
> {lo dacti} in principle could have a single
> referent, "Mr Object".

Not really. Or rather this won't work with Mr.
Object (in old Lojban {lo dacti} in a particular
context does have a single referent — the group
of dacti under consideration — but that doesn't
seem to be what "Mr. Object" means — it has been
rejected at least twice in the last few years).
So, if Mr Object is something that does all the
things it is supposed to do, it doesn't exist and
thus is no help at all.

> > I suspect this is all terminological, that
> > "universe of discourse" here means something
> more
> > restricted than usual (what has actually been
> > mentioned and perhaps what is going to be
> > mentioned in the foreseeable future,
> abstracted
> > from the broader notion of, say, the range of
> > quantifiers or the like). A smidge of
> > clarification might be handy here — and
> maybe
> > some new terminology as well.
>
> You're welcome to suggest other terminology,
> but I think "universe of discourse" is fairly
> standard.

It is standard, just not with this meaning, which
is why I had some trouble with it when you first
used it (and probably still do, since your
meaning is not very clear).

> > > >or ,
> > > > if so, what is excluded?)
> > >
> > > There is no general answer. When I say that
> > > there
> > > is nothing in the box, air molecules are
> > > normally
> > > excluded from the universe of discourse,
> but
> > > once
> > > I mention them, we have to admit that there
> is
> > > something in the box after all.
> >
> > I didn't ask for a general answer in the
> sense of
> > a list of allthe things that are always in
> the
> > universe of discourse, only criteria for
> deciding
> > whjat is in and what is not. Obviously
> > everything tha has been mentioned explicitly
> is
> > in. Almost as obviously, some other things,
> > implicit in what has been said or about to be
> > mentioned, are also in, but what are the
> limits
> > — by rule, not by list
>
> I don't know the rule. Relevance will probably
> be
> the most significant factor.

Not very useful, since "relevance" is surely as
obscure as this is. We only know what is
relevant when we see what comes up (possibly
including what comes up metaconversationally).
But by that time, you can look back at all the
things mentioned and say "That's the universe."
The task is to say something useful about it in
medias res.

> > > In {mi nitcu lo mikce}, the universe of
> > > discourse
> > > contains a single doctor, "Mr Doctor" for
> those
> > > who don't mind that picture. If you can't
> > > picture
> > > doctors as just doctors, and you
> necessarily
> > > must
> > > picture them as an aggregate of many
> individual
> > >
> > > doctors, each considered separately, then
> you
> > > must
> > > take another course, for example (mi nitcu
> lo
> > > nu
> > > da mikce mi}, "I need that someone treats
> me"
> > > or
> > > something like that. Here you are picturing
> all
> > >
> > > {lo nu da mikce mi} as one "Mr
> > > Someone-Treats-Me"
> > > that you need, but some people mind doing
> this
> > > abstraction less than doing the "Mr Doctor"
>
> > > abstraction. Events are somewhat easier to
> > > abstract
> > > than people.
> >
> > Ah, the history of Philosophy: those who
> ignore
> > it are doomed to repeat it. This is the kind
> of
> > metaphysical argle-bargle created by not
> paying
> > attenmtion to logic (or being to lazy to use
> it).
> > There is no Mr. Doctor (shouldn't that be
> "Dr.
> > Doctor"?) and if there were, he would be of
> no
> > help in satisfying a person's needs. Only a
> real
> > concrete doctor will do that, and further one
> who
> > is in the appropriate relation to the needer
> --
> > treating him, say.
>
> Mr Doctor (or Dr. Doctor if you prefer) is of
> course
> real and concrete. Just like an event of a
> doctor
> curing me has to be real and concrete in order
> to be
> any use to me. I have no need for an event of a
> doctor
> curing me if the event is not real and
> concrete.

This is the old cyclic dodge — not very
convincing. We have in Lojban that events exist
even if they do not occur (this is not my
favorite way of doing things, but it works and
solves the problems). So, what I need is for the
event to occur, meaning tht we should change the
definition of {nitcu} or — as we have always
done — sinmply take that as a give. The problem
is that if Dr. Doctor occurs, we still have the
problem To do what you want him to do, he has to
be abstract at least and so of no use at all. On
the other hand, if what you mean is that a
manifestation of Dr. Dr occurs, then we need to
change the definition of {nitcu} even more, since
that has not been what we have been doing
causually up til now. Further we have the same
problem: there is no manifestation that I need --
another one would always do, so on this reading
the claim is always false. And of course, the
doctor being real is not enough; we still need
the appropriate event, without which my need
cannot be met. So why not just say so straight
off — or at least recognize that this is
generally what is meant even when not said and
make the corresponding caveats in the text (just
as we do in English).

> > The fact that we cannot
> > identify beforehand who that doctor is,
> indeed
> > that the need is indifferent to that issue,
> does
> > not mean that there is an indifferent doctor
> who
> > is needed or an unidentified one. It only
> means
> > that the quantifier involved is within the
> scope
> > of the needing.
> > As for Mr. abstractions just being a variant
> of
> > event abstractions (and sometimes property or
> > truth function or... abstractions), the
> Lojban
> > answer is simply NO. The event, property,
> and so
> > on abstractions are inherent in Lojban; the
> Mr.
> > abstractions — assuming that it could be
> given
> > some meaningful interpretation (and all
> serious
> > attempts at this have failed so far) — is a
> new
> > thing, not already provided for. That it is
> > being used to hi-jack an existing
> construction,
> > which had a perfectly good but different
> meaning,
>

=