Saul Smilansky gave one of the best talks of the conference. Smilansky made a clever observation about the literature: even those who hold the most pessimistic views about free will emphasize the positive. This probably makes their views more attractive. But the dark side has its own attractions and I’m glad that Smilanksy called attention to them. One lingering issue, which Thomas Nadelhoffer and Tamler Sommers discussed, was the extent to which the logically impossible can be valuable. I think there is a strong prima facie case that the logically impossible is not valuable. But consider this counter-example: a man demands that you become both A, and not-A, at the same time or else he will shoot you. Here the logically valuable becomes valuable because another believes, mistakenly, that it already is valuable. This mistaken belief transforms the contradiction into something valuable. Perhaps Smilansky would analogize the free will problem to this case: ordinary responsibility practices require the logically impossible and so the logically impossible is valuable to the extent it preserves these practices. Of course, it would be best to convince Smilansky, like the man who threatens to shoot one, that the logically impossible is not valuable. Indeed, if the disease theory of crime is accurate, then abandoning this quest for the logically impossible will bring its own treasures.
Finally, Paul Russell’s discussion of hard compatibilism and the design argument impressed me. Russell and I had corresponded over email, before the conference, exploring Hume’s inconsistent writings on this subject. During the session, he mentioned me by name, an article I had recommended to him (Gary Watson’s “Soft Libertarianism and Hard Compatibilism”), and used me, together with three others, in an example to illustrate his point. Russell wants to find a compromise between the ad hoc nature of soft compatibilism (according to which one’s freedom might hinge upon whether God designed your life) and biting the full bullet of hard compatibilism (which denies this). Russell suggests his view is “nibbling on the bullet.” To reach this compromise, Russell claims that all designed agents are morally responsible but that the designer, alone, is forbidden from holding such puppets responsible. Holding responsible, but not responsibility itself, is agent relative. Later, over drinks, Manuel Vargas claimed that his own revisionist view (to be detailed in “Building a Better Beast” and his forthcoming book with Pereboom, Kane, and Fischer) is similar to Russell’s. I think this approach has a superficial appeal but, upon rigorous inspection, is untenable. This is so because, as Russell notes, the internal experiences of both a designed and non-designed agent are identical. But it seems unfair to hold one agent responsible for doing the exact same things, given the exact same situation, that another agent, not held responsible, did. Vargas argues that not all external properties are irrelevant. For example, whether a fish is in water is relevant to evaluating the fish’s life processes. But even if that external property is relevant there, it does not follow that the external property “somebody else designed my life” is relevant to moral responsibility. Consider the fact that the external property “somebody else designed my life”, unlike “currently submerged in water”, is purely historical. Perhaps the folk will settle the dispute. I suggested to John Fischer that we should ask the folk about the design problem and he noted that the folk may use a different concept of “free will” than philosophers do. But the folk conception is, if anything, less demanding than the philosophical one. So semi-compatibilists like Fischer might welcome the folk response to the design argument. Fischer favors an externalist concept of blameworthiness and distinguishes it from moral responsibility. On the internal conception of blameworthiness, moral responsibility (and holding responsible), however, which I would defend, the compatibilist cannot soften the sting of the design argument.