Philosophy 730                                                                                                                                                                                 W. Lycan
Fall, 2007

Objection to Ludlow's Position

         In class I maintained that Ludlow's example-based argument is unconvincing (because each of the examples is most naturally understood in a way that upholds the absolute distinction between the actual and the merely fictional).   Here I shall offer a brief objection to his view itself.

        First, two questions not answered in the text:  (1) In just what way is the view modeled on or inspired by pretense theories?   (2) In just what way(s) does context make true either a purely fictional sentence such as "Buffy slew a vampire" or a predication of vampire-slaying to Sarah Michelle Geller?

       Question (1) arises because, as Ludlow says, his view departs from pretense theories precisely by dropping the whole notion of a pretense.  So what's left?

       Question (2) arises because until we are told something about what contextual factors control the truth-values of fictional utterances and how they do so, Ludlow's idea amounts only to saying that a fictional sentence can be just plain literally true when "context makes" it so.  (Oh, I see.)

        (Remember, we all agree that a fictional sentence can be true or treated as true in an appropriate context, by way of a tacit story operator or an illocutionary context of pretense.  Ludlow's radical claim is that a  fictional sentence can be true without benefit of any story operator or pretense.)

       Ludlow assimilates fictional predications to "socially dynamic" predication.  What's characteristic of the latter is that the context-dependence is "extremely subtle to the social environment, often with surprising results."   So in response to question (2), he might at first say, "Oh, yes, well, very complicated, very subtle, no simple answers, mumph."  Fine, but he owes us at least the beginnings of a description of the relevant contextual features and how they push judgments of truth-value.  In the case of knowledge predications, (I agree) the context-dependence is more subtle and complex than at first believed, but we can cite some factors: what sources of evidence are available to the subject, gravity of the consequences of being wrong, background of other knowledge claims accepted in the context, subject's own doxastic history, and more; and we can give examples to show how each of these affects our willingness to ascribe knowing. 

        So what contextual features would (rightly) make us accept "Buffy slew a vampire," as true, without qualification by a story operator and without pretense?  It's hard to think of any that don't amount to: that we're entertaining a fiction and we're aware that it's a fiction.  (I add the awareness clause because if we were entertaining a fiction but were unaware that it was fiction, we would accept its sentences as true because we falsely believed them to be real-world true; no contextual explanation would be needed.)  As Ludlow says, he's committed to identifying the relevant contexts without reference to "fiction" or pretense, but--autobiographically--I don't see how anyone might do that, and he gives no hint.

        He says, "This move might have some merit if there was some single property or identifiable class of properties which we could identify as the property of being fictional" (et seq.).  But my point is that that cuts both ways:  The burden is now on him to limn an identifiable class of contextual features that would rightly make us judge a fictional sentence to be plainly and literally true.       

       A natural suspicion is that Ludlow's view really is a pretense theory.  Not a semantic theory, of course, but a standard illocutionary or speech-act pretense theory.  What's obviously special about his special contexts is that a kind of pretense is going on.  Which brings us back to question (1) above:  What "certain important similarities" does his view bear to pretense theories, if it is not itself one?  He specifies none.  Interestingly, one of his expository sentences is (also interestingly) ambiguous: "The core notion of pretending is dropped altogether."  That could mean either, "The notion of pretending, which is the core notion, is dropped," or "The core notion is dropped while leaving a cored or somewhat deflated notion of pretending in place."  The latter reading would be less mysterious, but I doubt it's what Ludlow meant, because there is nothing deflated about standard illocutionary pretense theories.  (Remember that they're already contrasted with fictionalist or other semantic theories.)  It's conceivable that he's really an illocutionary pretense theorist, but his rhetoric suggests otherwise.  Also, it looks to me as though, so far as his arguments work against the PRETEND operator theory, they would work as well against a standard pretense theory. 

       Concluding metaphysical note:  I said Ludlow agrees with me that his examples involve no metaphysical weirdness.  But his view itself seems to have a shocking metaphysical implication.  Suppose that fictional predications are subtly context-relative in much the way that knowledge ascriptions are.  Then, just as the distinction in nature between knowing and not knowing is not absolute, the distinction in reality batween being actual and being merely fictional is not absolute either.  Is Buffy a merely fictional person?  (Of course she is the topic of what we usually call a "fiction.")  In one of Ludlow's special contexts, it would be plainly and literally  false to say that Buffy doesn't really exist.  So whether she exists or not depends on context--period.   Startling enough.

            If whether Buffy exists depends on context, does the same apply to you, and me, and Descartes?  Why, on Ludlow's view, would it not?  And if it doesn't, that would be an absolute ontological difference between us and fictional characters.