[FOM] Possible Worlds

Max Weiss mmweiss at sfu.ca
Sat Dec 23 21:41:38 EST 2006

>>     (3) David Lewis, in conversation, in probably the late 1970s or
>> 1980s,
>> said he was attracted by an idea he attributed to Robert  
>> Stalnaker: that
>> what is really being claimed as possible when someone says "P  
>> might be
>> false" (P a mathematical statement) is the falsity of the  
>> SENTENCE: that
>> there is a possible world in which we so use language that the words
>> which
>> actually express the (for sake of example, assume true) conjecture
>> express
>> some falsehood.
> My problem with suggestions like this is that with such a notion of
> modality, any sentence might be false, and it is unclear to me just  
> how
> useful such a notion of possibility (and necessity) is.
> (Or perhaps I misunderstood the issue.)

The challenge Tim Chow raised is to account for truths of the form  
``it might not be the case that ..." where the ellipsis is filled in  
by the expression of a logical truth or logical consequence of  
necessary truths already known.  Or---a bit more generally---to  
explain epistemic possibility.

Allan Hazen describes a proposal to treat epistemic possibilities as  
alethic possibilities that are metalinguistic.  Such a proposal would  
involve analyses like:

(1) ``It might be the case that ..." is true iff it's possible that  
``..." expresses a truth.

Panu Raatikainen objects that any sentence might have meant something  
other than it does, and thereby might have had another truth value  
than it has.  Others will have different intuitions: for example,  
David Kaplan (``Words") and other philosophers (e.g. Peter Geach)  
have suggested that words have their meaning essentially.  The  
controversy indicates that our grasp on the application of alethic  
modality to semantical statements is worse than our grasp on the  
epistemic modals it is supposed to explain.

Of course, maybe some people can improve our understanding of alethic  
modality in relevant ways.

But still the explanation suggested seems wrong.  For suppose A is a  
formula of the predicate calculus that is complex enough that we  
cannot determine conveniently whether it is valid.  Then, A might be  
valid, and A might be invalid.  What makes this so?  The explanation  
suggested by (1) is, perhaps, that the connectives might have had  
other meanings than they have.  But doesn't the question of the  
validity of A presuppose that the connectives of A mean what they do?

Max Weiss

More information about the FOM mailing list