Opiniatrety

Who Denied It?

[UPDATE: I should make it clear that I have no evidence that Sarah Champion is the author of Belle de Jour, and in fact I doubt it. One of the CT commentators is now claiming to be the author of Belle de Jour and not Sarah Champion, if I read the post correctly. All this concerns a scenario, possibly counterfactual, in which Champion is the author of Belle de Jour.]

Chris Bertram reports on Belle de Jour, an anonymous blog purporting to be written by a British prostitute. (I think--I can't get the link to work, and if I could I'm sure it would be very naughty, so click at your own risk.) Literary detective Don Foster claims to have identified the author of BdJ as one Sarah Champion.

Why am I blogging this? Because Chris reports that Belle has denied being Champion. Commentator "jam" notes that "BdJ may be Sarah Champion. As the Times noted, Champion didn't deny it." Leading me to observe that the Times seems to be begging the question: If Belle denied it, and Belle is Champion, then Champion did deny it.

Well, what I said was a joke--we know perfectly well what is meant by "Belle denied it but Champion didn't," even if Belle is Champion. But the question is: Why do we accept it? It seems to me intuitive that we might say that each of the following three is true:

(1) Belle denied that Belle was Champion.

(2) Champion did not deny that Belle was Champion.

(3) Belle is Champion.

Yet they seem to form an inconsistent triad.

For me, the first natural response is to say that, in these circumstances, "X denied that p" implicates "X, speaking in the persona of X, denied that p." So (2) might be literally false but convey the truth of

(4) Champion did not, speaking in the persona of Champion, deny that Belle was Champion

while (1) is both true and conveys the literal truth

(5) Belle, speaking in the persona of Belle, denied that Belle was Champion.

(4) and (5) can both be true even if (3) is true, because "speaking in the persona of" is an opaque context.

Yet this analysis might have trouble with

(6) The author Foster named did not deny that Belle was Champion.

because it would require evaluating what the author Foster named under the persona of "the author Foster named." And it's not clear that she says anything under that persona.

A similar approach is suggested by Geoff Nunberg's paper "Indexical Descriptions and Descriptive Indexicals." Nunberg suggests that sometimes an indexical contributes one of the properties of its referent to the truth-conditions of a sentence. For instance, if I am a prisoner condemned to death, I may say:

(7) Traditionally, I get whatever I want for my last meal.

(7) doesn't mean that Matt Weiner traditionally gets what he wants... because there's no tradition about Matt Weiner's last meal; rather I contribute my property of being a condemned prisoner, and (7) is true iff traditionally a condemned prisoner gets whatever he wants for his last meal. I don't think Nunberg extends this to names (I can't open that link now either), but if you do extend it to name you could say that in (1) Belle contributes the property of being the person who posts on the Belle de Jour website, while Champion contributes the property of being the person who speaks publicly as Champion; and perhaps in (6) "The author Foster named" also contributes the property of being the person who speaks publicly as Champion.

Another possibility is to deny (3) in some form or other, as commentator Keith M. Ellis did in response to my claim that if Belle denied it, and Belle is Champion, then Champion did deny it:

It depends upon what your definition of is is (as used in this context). Although we might use the word identity in this context ("Belle's identity is Champion”) it would not have its mathematical meaning. Belle and Champion are not mathematically identical, if for no other reason than that Belle calls herself “Belle” and Champion calls herself “Champion”. The context makes clear that we must necessarily think of “Belle” as someone distinct from Champion even if they are in fact the same person. Thus, whether or not Belle is Champion, what Belle says Champion does not necessarily say.

I think what Keith is getting at here is something like Hector-Neri Castaneda's guise theory, though I don't know much about that. Of course it's in a way begging the question to say that Belle calls herself "Belle" and Champion calls herself "Champion"; if we accept that (3) licenses intersubstitutability in transparent contexts, then if (3) is true Belle calls herself "Champion" when speaking publicly and Champion calls herself "Belle" when blogging. But you could say that Belle and Champion are the same person but not the same voice, and that might license saying that what Belle says Champion does not say.

Perhaps the most economical way is to say that Belle is a fictional character. Then (1) is only true within the fiction, (3) is decidedly false within the fiction as well as in the real world, and (2) is true in the real world. Hence, no inconsistency. This is bolstered somewhat by the idea that Belle's exploits are indeed fictional--that is, untrue--but it seems as though Belle is not presented as a fictional character.

Anyway, it may be worth pondering why (1)-(3) seem acceptable.

Posted by Matt Weiner at March 18, 2004 07:32 PM