Re: lo

[jorge@PHYAST.PITT.EDU]

la pycyn cusku di'e

>         Somone suggested that I had special insight into dscriptions and
> quantifier.  I sure don't in English other than years of experience
> and even then I am often fooled, especially by isolated examples.  But in
> Lojban it is a snap: descriptions start with a member of selma'o LE and
> quantifiers start with members of RO or PA or with just a bound variable,
> DA. Quantifiers involve DA, explicitly or implicitly, cases like
> _su'o_broda_ (from _su'o_da_poi_broda_, not _lo_su'o_broda_, where _su'o_
> is a cardinal, not a quantifier)

I agree that {su'o broda} should be equivalent to {su'o da poi broda},
which in turn is the same as {da poi broda}, because {su'o} is the default
quantifier of {da}. It is also by definition equivalent to {lo broda} =
{su'o lo ro broda}. I believe you don't agree with this definition, but
that is how the grammar papers define it.

> pick up -- if need be -- the latest
> available bound variable (I forget the reset time on running through the
> unused ones -- probably "reasonable" is about right).

I think it was decided in the last Logfest that {da} only remains bound
within one single bridi, unless it is in a prenex outside of {tu'e}, but
John can probably give more details. I'll try to remember what was the
problem of just leaving it to context.

>         I take it we all agree that quantifiers are always -specific and
> hence automatically -definite as well (you can't know/depend on an
> identification that has not been made).

If you exclude {ro} from the quantifiers, certainly. Otherwise, no, because
since every sumti is explicitly or implicitly quantified, we would have
no possible +specific.

> They also are and have to be
> +veridical, since otherwise they could not hook up to a reference at
> all.

Yes, although I'm now a bit uncertain. For example, say I use {le ninmu}
to refer non-veridically to a number of men. Then I could say

        mi viska pa le ci ninmu
        I see one of the three "women".

This is -specific. Is it also -veridical? It looks like it is.

>         I take it also that we all agree that _le_  and its ilk in LE for
> sets and masses are +specific, +definite and -veridical, since reference
> is all there is here, the words used are just a guide to the referent,
> not what determines it, a fancy _ti_, if you will.  The only comment I
> would add is that +definite is less about whether we know what the
> referent is but whether that knowledge is essential to understanding what
> is being said.

Yes, that sounds right. I guess the definition is changing all the time,
hopefully we'll finally settle on one :)

>         We also agree that _lo_ and its ilk are +veridical and
> -definite.  I argue that, both because it is a description and to fill a
> gap in the pattern, _lo_ and its ilk are +specific.  The +veridical is
> then essential, for without a known referent (-definite), the referent
> cannot be determined except through its properties. I agree with Xorxes
> that this means that the default quantifiers are wrong and that set me
> wondering how those were set.  I have a memory of issuing a bunch of
> obiter dicta on questions like that on the basis of 30 second
> presentations of issues while I was in my Lojban oblivion phase.  If that
> is the history, I'd like to say I have more information now and would
> like to change my vote.

What would you change it to? Would {lo broda} be {ro lo su'o broda} =
"Each of the at least one real broda of this context"? I think I prefer
the status quo.

>         In passing, I note that when we do know the identity of the
> referent, it is often a good idea to use _le_, because the _lo_ reference
> may not get who we think it does.  If we think bunker White (sorry about
> being dated) is the richest man in the world, we may say that the richest
> man in the world takes his lunch to work in a brown paper bag.  But the
> Sultan of Brunei does not.  But it is amazing how much we can say about
> something just on the basis of its veridical description, not knowing who
> it refers to but that the referent is fixed.

All the examples have been of broda with a single element, like "the
assassin of Ferdinand", or "And's first wife", or "the richest man in the
world". It is easy to think of these as specific because it doesn't really
matter, they are singular and that is really enough for logical purposes,
you could use {lo} as it is now without need to change anything.

The problem is with descriptions that fit more than one referent, and yet
we want to use them with {lo} as +specific. If {lo} was changed to +specific,
I would have to do a lot of relearning, because I'm used to thinking of
it as -specific.


>         But all of this has absolutely NOTHING to do with opacity.

That's true :)

Jorge