[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Presuppositions



[I'm still massively behind with mail. Apologies if I've said stuff
I've not yet come across in unread mail.]

Lojbab to Chris:
> These presuppositions sound like what we have been meaning by the word
> "veridical" when defining "lo".  The use of lo does make explicit a
> presupposition that the referent ahs the indicated property.  But "le"
> descriptions do not, and djuno should work with non-verdicials as well.

I agree with Chris about presuppositions. Logically, the way I think
they should be handled is as predications OUTSIDE the scope of the
illocutionary force of the sentence.
For example, "She wonders why he left her" would be:

   He left her, and I ASSERT THAT she wonders what the reason for
       that is.

This is precisely what veridical poi and lo do NOT involve, but it
*is* what nonveridical voi and le involve. Stuff in a voi clause or
in a le sumti is not *asserted* by the speaker.

However, while it is generally though not uncontroversially held that
the complement of _know_ is presupposed to be true, I am actually
opposed to this for Lojban. I think "ko`a djuno ko`e" should mean:

A    I ASSERT THAT ko`a justifiably believes ko`e (about x3, with
      justification x4) and ko`e is true

--which does not *presuppose* the truth of x2-- rather than

B   ko`e is true and I ASSERT THAT ko`a justifiably believes ko`e (about x3,
 with
      justification x4)

The difference would show up in negation. I think "ko`a na djuno
ko`e" should mean

A'   I ASSERT THAT it is not the case that ko`a justifiably believes ko`e
         (about x3, with justification x4) and ko`e is true

rather than

B'   ko`e is true and I ASSERT THAT it is not the case that ko`a justifiably
          believes ko`e (about x3, with justification x4)

(B' is the meaning of English "X does not know that Y" - unless the
"not" is metalinguistic negation.)

The reason I favour the non-presuppositional version is that I think
the presuppositional one but be unnecessarily (and therefore mabla)
glico, and more generally any presuppositions, since they behave in a
special way logically, should be *explicitly* flagged (as they are by
voi & le).

--And