Recently I’ve been noticing a spate of proposals designed to base philosophy on a more “rigorous” understanding of human behavior and beliefs (known as “experimental philosophy” or “X-phi”). Here’s a report on an example of this trend:
Traditional philosophy relies on certain intuitions, presented as “common sense,” that are presumed to be shared by everyone. But are they? For example, can people be morally responsible for their actions if they don’t have free will? Many philosophers have assumed that all sane people would of course say no. Experimentalists don’t assume. They ask. Recently, they presented the following scenario to two groups:
Bill and his wife were flying home from vacation with their friend Frank, who was having an affair with Bill’s wife, as Bill knew. Kidnappers injected Bill with a drug that forced him to obey orders, then told him to shoot Frank in the head, which he did.
They told the first group that Bill wanted Frank dead and so grieved little for him. To the second, they said that Bill hated what he’d done. Both groups were then asked if Bill deserved blame for Frank’s death.
Traditional philosophers have argued that Bill shouldn’t be blamed in both cases because it’s common sense that moral responsibility requires free will. But, in fact, the first x-phi group did blame Bill in the scenario in which he welcomed Frank’s death.
So what exactly does this prove? Perhaps that individuals transfer “blame” from their concern about a bad mental state to their evaluation of an act (a sort of reverse transfer of merit.) Does the mere prevalence of such reactions increase the validity? I think not….and in fact, I would despair of attributing any objective meaning to such reactions. “Social practices cannot be brute data, for they are ‘partially constituted by certain ways of talking about them.'” (Berger, Self-Interpretation, attention, and language, in Lavoie, ed., Economics and Hermeneutics 263 (1990)).
I admit that X-Phi has been well-explored in a post on the Leiter blog here, and I’m afraid that I can’t properly respond to all the points made there. Nevertheless, I want to register some concerns about the degree to which we take as given existing assumptions about morality and justice. Are these really the basic foundational building blocks of philosophy, as Rawls’s “method of reflective equilibrium” seems to assume? Or merely the targets for a rhetoric designed to improve society on more objective grounds?
David Papineau suggests the latter in an essay called “The Tyranny of Common Sense.” Here’s his comment on the methodological conservatism at the heart of projects built on “everyday intuitions:”
Sometimes I despair of my philosophical colleagues. . . . When it comes to philosophical ideas, they are inveterately suspicious of intellectual innovation.. . . [For example,] on the political level, Rawls’s moral system endorsed the principles of liberal social democracy. But his philosophical methodology was far more conservative. Rawls urged the method of “reflective equilibriumâ€, which measured moral systems against pre-theoretical moral intuitions about particular cases. Such intuitions could on occasion be jettisoned, if they stood in the way of some overarching moral principle. But, by the nature of the case, any system that satisfies the requirements of reflective equilibrium must respect the great preponderance of everyday moral reactions.
As Papineau counters, “It can’t possibly be a good idea to assess philosophical theories by the extent to which they preserve everyday intuitions. The trouble is that everyday intuitions are often nothing more than bad old theories in disguise.”
So in the case of “Bill” above, perhaps the respondents’ negative reaction to Bill matches some expansive (ab)uses of the federal mail fraud and RICO statutes–i.e., to make someone guilty of the “crime of being a criminal,” or convictable on the basis of a long but inchoate pattern of bad behavior. But doesn’t such free-floating culpability represent a “bad old theory” of the criminal justice system that civilized peoples reject?
In conclusion: such “data” from experimental philosophy can never be brute. They only make sense in the context of a larger narrative. The narrative does the work, not the “data.”