I just ran across the sentence “Unicorns don’t have wings.” It strikes me as a good example of the sticky problem of truth-conditions. We say that a statement is well-formed iff it describes a possible world. It is true iff it describes an actual situation in that possible world. You can say “Harry Potter has a scar” and this is true because in the possible world of Harry Potter (yes, we have a loose definition of “possible” here) he does have a scar. You can say “Harry Potter is married to Hermione,” and although well-formed, this statement is false. The statement to be true need not describe this world, merely one that is “possible.” In a possible world with unicorns, the statement “Unicorns don’t have wings” is true.
Similarly, we can say “There is no such things as unicorns” and this statement is true.
But what if you say *”Unicorns don’t have wings, and there is no such thing as unicorns.” That doesn’t even seem well-formed; it’s neither true nor false, but meaningless. And if you reverse it, it’s even more clearly ill-formed: *”There is no such thing as unicorns, and unicorns don’t have wings.”
I think the problem is that the two statements are not consistently describing the *same* possible world. To say “unicorns” is to presuppose the existence of unicorns, and to immediately shift to a possible world different from this one.
But wait. Let’s say “The King of France drives a sports car.” Fine, there might indeed be a possible world where the King of France drives a sports car, but in our world there is no King of France. My instinct is that this sentence doesn’t invoke a possible world the same way that the mention of the unicorn does. Is it because we could plausibly assume that someone in our world might not realize that France does not have a king? If I say this sentence someone would be justified saying “Wait, there is no such thing as a King of France.” If I say “Unicorns don’t have wings,” I’d find it at best oddly marked for someone to gravely inform me that there is no such thing as a unicorn.
What if I say “The king of France rides a unicorn.” Huh, my instinct, purely my instinct alone, is that this is not as ill-formed as “The king of France drives a sports car.”
Huh. Is there a semantic feature of certain words that invokes a possible world, thus asserting a different set of truth conditions? If so, what? +[mythological], maybe. That would mean that semantics interact with pragmatic schemata. But if that’s the case, it’d seem possible and maybe even reasonable to imagine that other features might trigger other schemata. In fact, it seems — just guessing here, I’d have to figure it all out formally — but maybe semantic features are just schematic triggers or something of that nature.
Normal people lie in bed thinking about their bills. Me, it’s semantics. I have a vague recollection of reading something along these lines in grad school. I should dig through my files and see if I can find it.