Turning now to the ancient mystery of moral responsibility and free will, I can imagine no better place to begin than Herman Melville’s Moby Dick or The Whale:

Now that I recall all the circumstances, I think I can see a little into the springs and motives which being cunningly presented to me under various disguises, induced me to set about performing the part I did, besides cajoling me into the delusion that it was a choice resulting from my own unbiased freewill and discriminating judgment (7).

 This much was plain to Melville over a century ago.  And today, we have good reason to suspect that conscious control, the ideal agent, the dispositional self, and timeless propositional content are to some extent illusions. It is therefore not clear to me why we insist on framing ‘will’ and ‘judgment’ as empirical targets of exploration.  Now that the veil has been rent, why continue in the law of the philosophers?  The situation, not dependent on man’s will or effort, hardens whom it will harden and shows mercy to whom it will show mercy.

I do not know if Al Mele has studied Moby Dick at length over the last few years or dedicated himself to the proper Calvinistic interpretation of Pauline literature. What I do know is that he has just received a $4.5 million grant to study “Free Will: Human and Divine”.  It is not yet clear to me that there is something we call “will” to study, specifically, and so I am mildly skeptical about the project of investigating “free will”.  To empirically study ‘white goo’, for instance, as apposed to black goo, there must be something certifiably tangible in the world that goes by the label ‘goo’. Yet, for 2010, $132,000 is allotted to the theology of free will, $165,000 is allotted to the conceptual underpinnings of free will, and $2.8 million is allotted to the science of free will. I am not skeptical about the conceptual underpinnings of free will, as this just might be all there is in the world to study as relevant to our encounters with the linguistic item ‘free will’.  As Dennett once put it: “Philosophers have a choice: they can play games with folk concepts (ordinary language  philosophy lives on, as a kind of aprioristic social anthropology) or they can take seriously the claim that some of their folk concepts  are illusion generators” (2007).

Mele was kind enough to attend my paper presentation on propositions and consciousness a few years ago for a graduate conference at FSU, but from the discussion that followed it was clear that we shared somewhat different intuitions about the existence of certain abstract entities that flutter within the imagination of philosophers. I have not studied Mele’s past work – and I should – yet I will for the purpose of this entry assume that a similar clash of intuitions explains my mild skepticism over this new proposal for a science of free will.   Just as there is no fixed propositional content in my possession when I “know that”, so there is no performance of a discrete action by the faculty of will arising from free, conscious deliberation of fixed, determinant reasons.

Dennett (2003) has offered some critical commentary on Mele (1995), but merely pointed the way to John Fischer for further research on the Harry Frankfurt tradition.  As it turns out, just a couple weeks ago I was fortunate enough to attend John Fischer’s UCSD presentation, in which he presented one of his forthcoming papers. Fischer convinced me that he had arrived at a fairly conclusive argument against moral responsibility’s requirement of alternative possibilities or indeterminism or any other such thing. The argument goes something like this (as far as I recall anyway): 

World 1 (W1) is reality. World 2 (W2) is an invention for the sake of the thought experiment. In W1, a person S considers Action V at Time 1 (T1), decides to perform V at T2, and performs V at T3. Further, S is morally responsible for V.  Consider, however, an alternative scenario: S decides not to perform V at T2, even when S’s conscious deliberations at T1 remain unchanged. Must this alternative scenario be genuinely possible for S to be sufficiently free and therefore morally responsible for V? 

Now consider W2, in which there is a random generator that produces genuinely random outputs of either M1 and M2.  If the output is M2 at T2, nothing happens that does not happen in W1.  The choice at T2 leads to V in the same way as it does in W1.  However, if the output of the random generator is M1 instead of M2, the normal processes from T1 to T3 are bypassed, externally forcing an alternative decision at T2 – perhaps through remote control signals from the generator to a probe in S’s head.  Because of the presence of this random generator, W2 is characterized by the relevant form of indeterminism.  

In terms of moral responsibility, is there any important difference between W1 and W2 when the random generator’s output is M2 in W2?  The intuition is that there is no important difference and what accounts for this intuition is that what is required for moral responsibility is not genuine alternative possibilities or indeterminism or any other such thing, but rather the right kind of “glue” holding together the deliberations at T1, the decision at T2, and the action at T3.  Moral responsibility depends on the right kind of “dispositional features of the sequence” leading from deliberation to action.  The action must be “suitably” related to prior processes. What determines whether S is morally responsible for V or not depends on the “intrinsic” way prior processes lead to V. 

 Fischer’s paper will be published at some point and so my somewhat thin recollection of Fischer’s presentation might need revision.  But minor revisions should not mitigate the modest purpose I have in summarizing this new thought experiment. This is because the important question I have in mind pertains to just what, precisely, these “dispositional features” or “intrinsic sequences” are (I jotted these phrases down during the presentation).

Fischer need not answer this sort of question.  His work is done. I am largely convinced. However, these rough, concluding intuitions must now point us to empirical research. But what if such research reveals aspects to this sort of intrinsic, sequential “glue” that significantly undo the very intuitions that gave rise to Fischer’s thought experiment?  History suggests that this is precisely what we will encounter.

According to Harvard psychology professor Daniel Wegner  (2002), for example, the very idea of the conscious will is an illusion. Dennett (2003) sees no need to fuss about the word ‘illusion’ here, but regardless of which semantic route ends up most empirically healthy, we might end up as concerned about the semi-folk concept of ‘free will’ as we now are about the causal relation between the brain and the immaterial mind – at least once we begin coming to grips with the sophisticated unconscious and the interpenetrating – and in this case competing – parallel cognitive processes that cajole us “into the delusion that it was a choice resulting” from our “own unbiased freewill and discriminating judgment”.

I agree with Fischer. It is all in the glue.  But classical Frankfurt thought experiments were not necessary for the contemporary mind sciences to assume just this – at least so long as we keep Moby Dick nearby.  And it will not be to such thought experiments that the mind sciences must now turn.  Or perhaps I missed something. . .