WikiDiscuss

pc:
> The last remark is an interesting point, but I
> am not sure it is right. I did not *say* that I
> gave the box to John, I only said "I gave to
> John."

We do understand {zo'e} differently, then.
In English, you can't drop the "it" to say
"I gave it to John". In Lojban, you can
say {mi pu dunda zo'e la djan} or
{mi pu dunda fi la djan} to mean the same
thing, i.e. that I gave *it* (i.e. the obvious
thing) to John.

> Based on a variety of pragmatic factors,
> you *understood* that it the box I gave. If it
> was something else (not otherwise dealt with in
> the context), you can accuse me of a number of
> pragmatic failures, of being misleading in
> general, but not of saying something false.

I'm not talking about accusations of saying something
false. You could also use {le tanxe} to refer to
something other than the box and then claim you were
not saying something false. All I'm saying is that
{mi dunda fi la djan}, in that context, and when there
are no misunderstandings, is true iff I gave the box
to John, not if I gave something else to John. Just
as {mi dunda le tanxe la djan}.

If there are misunderstandings, such as the participants
disagreeing about what the referents of {zo'e}, {le tanxe},
{la djan}, (even {mi} maybe) are, then that's a different
matter.

Pragmatic failures aside, I find that {zo'e} has a referent
just like {le tanxe} does.

> {zo'e} belong to that strange set of place
> pluggers (like {zi'o} "this place does not exist
> now") that fill places explcitly for a variety of
> purposes — getting the place count to work out,
> guaranteeing there is no {ce'u} in that place,
> scansion, and so on — but none of them is to
> refer to someone. {zo'e} says "there is no need
> to mention someone here" and so is semantically
> equivalent to a blank. And blanks don't refer.
> Neither does "it doesn't matter who" nor "you
> know who" (sentential, not nominal), the two
> suggested expansions of {zo'e}.

Well, if zo'e, like zi'o, reduces the number
of places of {dunda} by one, then that's another
valid way of seeing it. But in that case, what
the reduced relationship is will be context dependent.
{dunda be zo'e} in the given context means
"x1 gives-the-box-we're-talking-about to x2". Then the claim
is that that two-place relationship holds of (mi, la djan).
I don't have a problem with that view. The claim will
not be satisfied by my giving something else to John,
that would require a different relationship.

> The pragmatic understanding has here to be
> separated from the semantic claim (cf. the fight
> about whether {zi'o} refers to nothing --
> ignoring the error avbout "nothin" of course.)

I can see it either way, {zo'e} has a context dependent
referent, like {ta}, or {zo'e} changes the relationship
to a different one, like {zi'o}. The effect is the same
as far as I can tell.

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail is new and improved - Check it out!
http://promotions.yahoo.com/new_mail