(no subject)

[Pycyn@xxx.xxx]

this was bounced on 12/10 -- maybe it clears something up?

The "for x, if x went to the party then John knows that x went to the party" 
version (maybe with the added "and if x did not go to the party, John does 
not believe that x went to the party" or maybe with some other sort of 
relative completeness marker) must be used with care, because of the 
intensional nature of "know." We should not infer from the fact that Paul 
went to the party that John will say "Yes" to the question "Did Paul go to 
the party?" even if John is totally truthful and cooperative, for he may not 
know Paul under the name "Paul." So the real expansion seems to be 
something 
like "for every x there is a concept y which selects x such that John knows 
that (the referent of) y went to the party," etc. The external 
(extensional) 
quantifier on x should manage that, but once the substitution is made inside 
the intesional context, the external connection is lost or at least 
weakened. 
This tends to separate indirect questions slightly from direct ones, 
perhaps 
partitioning the answer set into equivalence classes with isosemic terms, at 
least one of which the knower knows. But it is the knowers isosemy, for he 
may think that Paul is Bill's father and thus believe that Bill's father was 
at the party and that neither Paul, who he has never heard of, nor Joan's 
father, who isd who Paul really is. 

A couple of pages on from all of this in the Handbook of Philosophical 
(i.e., 
flaky) Logic is a reminder that every natural language sentence is the 
dependent of a (usually unexpressed) performative and that some of these 
performatives may also be intensional, putting all of the terms in the 
surface sentence into that never-never land where Leibniz's law doesn't work 
and existential generalization and universal instatiation don't work. In 
particular, the suggestion is that all sentences that mention no existenct 
objects and that are held to be true are under such performative, say "I 
recite to you a myth that" or some such. This being so (and it sure solves 
a 
lot of problems), Lojban ought to dig into its small stock of unused cmavo 
to 
come up with a sentence length warning flag, when context is not sufficient. 
Remembering that the term length form is something like tu'a, I suggest the 
corresponding form with x, xu'a?
pc