Reclaiming Autonomy in Moral Responsibility — Epoche C2
The argument this essay is answering Galen Strawson's 1994 paper "The Impossibility of Moral Responsibility" contains a short argument, which he himself calls the Basic Argument, to the conclusion that no one has ever been or could ever be truly morally responsible for anything they do. A correction is owed at the outset: an earlier version of this essay called it the "true self" argument. That is not Strawson's label, it does not appear in the paper, and it misdescribes a structure that turns on a regress of self-shaping rather than on any account of an authentic self. In its most compressed form the argument runs: Nothing can be causa sui — nothing can be the cause of itself. To be truly morally responsible for one's actions, one would have to be causa sui , at least in certain crucial mental respects. Therefore no one can be truly morally responsible for anything. The phrase causa sui Strawson takes from Nietzsche, who in Beyond Good and Evil §21 calls the notion the best self-contradiction yet conceived and likens the craving for it to hauling oneself into existence by the hair. Put that way the argument looks like word-play; its force is visible only in the expanded version, which is a regress: What you do in any given situation is a function of the way you then are — your character, values, motives and momentary state, taken together as your mental nature at that time; call it $N_0$. So to be truly responsible for what you do, you must be truly responsible for $N_0$, at least in the respects that bear on the action. To be truly responsible for $N_0$, you must have brought it about that you were that way, and brought it about deliberately rather than by accident. But deliberate self-shaping is itself an action, chosen for reasons, in the light of some earlier nature $N_{-1}$ — the principles of choice you already had when you set about remaking yourself. So you must be truly responsible for $N_{-1}$; and step 3 applies to it, requiring $N_{-2}$, and so on without end. No finite agent completes an infinite series of self-shapings, and no member of the series is one the agent is responsible for independently of an earlier one. That is the regress, and it is valid. To resist the conclusion one must either find a stopping point in the series or reject the premise that responsibility for action requires responsibility for the nature the action springs from. Two features of Strawson's framing decide in advance what a reply has to do. First, he does not deny that there is a workable everyday notion of responsibility; he grants perfectly good compatibilist senses in which people are responsible. What he denies is true or ultimate moral responsibility, glossed with a deliberately theological image: the kind that could make it just for an agent to be consigned to eternal reward or punishment on the strength of what he has done. Anyone offering a lesser notion is, on his view, changing the subject. Second, the argument is built to be indifferent to determinism: if an action flows from your prior nature you are not responsible for it, because you are not responsible for that nature; and if part of it is undetermined, that part is not something you did, so indeterminism subtracts from responsibility rather than adding to it. This is why the classical compatibilist replies do not touch it. Two attempts to stop the regress, and why the essay takes neither The natural response is to find a point in the series that is genuinely the agent's own without being inherited. Two well-developed attempts exist, and saying why neither is the route taken here is the quickest way to make the present position precise. Harry Frankfurt's "Freedom of the Will and the Concept of a Person" (1971) locates freedom in the structure of the will rather than in its history. A creature has first-order desires; a person also has second-order volitions, desires that a particular first-order desire be the one that actually moves her — that it be her will. Freedom of the will is then having the will one wants to have, and the unwilling addict is unfree because the desire that moves him is not the one he wants to be moved by. Strawson's objection is short and, on its own terms, decisive: the question moves up a level. If the agent is not responsible for her second-order volition, identification with a first-order desire confers no responsibility on the act; and to be responsible for that volition she must have formed it in the light of a third-order commitment, which restarts the regress. Frankfurt's remedy, sketched in 1971 and developed afterwards, is that identification can be decisive — that wholeheartedness terminates the hierarchy without arbitrariness. That stops the series but does not deliver what the series was needed for. It tells us where the agent stands; it does not make her the author of her standing there. Robert Kane's The Significance of Free Will (1996) takes the opposite tack: it accepts the regress premise and hunts for a terminus with no prior sufficient reason. Kane's condition of ultimate responsibility states that an agent is ultimately responsible for an action only if she is also responsible for anything that is a sufficient reason, cause or motive for that action's occurring. His stopping points are self-forming actions: occasions of torn decision in which an agent is trying at the same time to decide in two incompatible ways — the business traveller who hears the assault in the alley and is pulled between conscience and her meeting. The indeterminism sits inside the effort of will, and because she is genuinely trying to do each thing, whichever way it comes out is something she wanted, endorsed and brought about, so the outcome is not mere chance. Kane locates that indeterminism in thermal noise amplified by chaotic neural dynamics, which makes the account naturalistic in the sense of requiring no soul, but does not make it compatible with determinism. This is where the essay in its earlier form need