Re: PLI: a triviality

[Jim Carter <jimc@MATH.UCLA.EDU>]

On Wed. 6 Dec 1995 Jorge Llambias <jorge@PHYAST.PITT.EDU> wrote:

> The difference between {le} and {lo}, as I understand it, is this:
>
> le broda = each of the broda that I'm talking about.
> lo broda = at least one of all the broda that there are.
> ....
> (Others view it somewhat differently, giving more importance to
> veridicality.)

As I understand it, your definitions of {le} and {lo} are orthodox.  But also,
you actually affirmed pc's position about veridicality!  "All the broda's that
there are" means that each of the candidate referents (from which certain ones
are selected) is truly a broda.  In contrast, with {le} the salient feature is
that you're talking about it.  It's customary for the referent to really be a
broda -- if you want your listeners not to get lost in a thicket of deep
metaphors -- but not essential.  For example in intemperate descriptions one
often says "the bastard", and it's obvious who the referent is, but without
any veridical involvement of the marital status of the referent's mother.

I believe Steve Hazel's original question was not trivial -- the exact
 definition
of the various <LE>'s is a foundation of the language and is the topic of
a tech harangue at least twice a year.

A confusing aspect of these tech harangues is the appearance of the word
 "claim",
as in "a veridical sumti makes a claim about its referents".  When I see "claim"
I think of an implicit subphrase stuck on the sumti: {noi ke'a broda} = "and
by the way, that referent really is a broda".  Other formulations use "the
 referent
exists in reality" for the implicit phrase.

I wish these implicit subphrases would go away for the following reasons:

1.  The point of the containing bridi is to express the relation of its selbri
among the given arguments.  Adding mandatory implicit baggage obscures the
speaker's intent.  Particularly if different people specify different baggages.

2.  With {lo}, referents being broda is implicit in the definition of {lo}, not
as a separate claim, and so no additional implicit phrase for that is needed.

3.  Peculiar interpretations can arise if a referent set turns out to be empty.
It's customary for speakers to specify nonempty referent sets, but Lojban is
supposed to be a "logical language", and I think its users should know enough
logic to deal with empty sets when they arise.  In other words, I don't like
a mandatory implicit subphrase saying "at least one referent exists in reality",
whatever can of worms "reality" is.

However on the last point, {lo} has an implicit outside quantifier of {su'o}
(at least [one]), giving rise to an implicit counting subphrase.  Thus were
the set of candidate broda's empty, any proposition including {lo broda} would
end up false, without the intervention of unicorn-containing realities.  The
implicit subphrase about existence is replaced by an implicit phrase about
how many candidates fit the s-bridi {broda}, which to my mind is logically and
practically far more defensible.

James F. Carter        Voice 310 825 2897       FAX 310 206 6673
UCLA-Mathnet;  6115 MSA; 405 Hilgard Ave.; Los Angeles, CA, USA  90095-1555
Internet: jimc@math.ucla.edu (finger for PGP key)
UUCP:...!{ucsd,ames,ncar,gatech,purdue,rutgers,decvax,uunet}!math.ucla.edu!jimc