Reading Comprehension

That every event has a cause is among the better-established working assumptions of natural inquiry. That human beings, themselves natural systems, exercise something properly called freedom of the will is among the more durable convictions of moral and legal life. The conjunction of these two positions presents a difficulty that is older than philosophy and has yet to be entirely resolved by it. If every action is the determined outcome of antecedent conditions over which the agent had no eventual control, in what sense can the action be ascribed to the agent rather than to the conditions? The question has the awkwardness of a problem that refuses to disappear under the kind of attention that usually disposes of awkward questions.

The compatibilist response, traceable through Hobbes, Hume, and a long subsequent line of philosophers, insists that the awkwardness rests on a misreading of what 'freedom of the will' was meant to denote. To act freely, on this view, is not to act in defiance of antecedent causation but to act in accordance with one's own desires and reasoning, without external constraint or coercion. A person locked in a room is unfree; a person choosing among courses of action on the basis of his own reflective preferences is free, regardless of what the deterministic story of those preferences would, in retrospect, look like. The freedom that matters for moral assessment is not the freedom to have done otherwise from a metaphysically identical starting state but the freedom to do as one's own deliberation directs.

Critics of compatibilism have insisted that the move is a redefinition rather than a solution. The original worry, on their reading, was that no one is the ultimate origin of his own actions, and rebranding the proximate causes of action as 'the agent's own' does not address that worry; it only relocates it. If the agent's deliberations are themselves the determined product of conditions he did not author, then calling those deliberations 'his own' is true in a thin sense and unhelpful in the sense that the original question demanded. The objection is sharp, and the reply that 'thin senses are senses too' is not, on its own, fully satisfying.

What survives, then, is a stand-off whose terms are clearer than its resolution. The compatibilist account captures something the ordinary moral practice presupposes; the incompatibilist objection captures something the metaphysical ambition presupposes; and the two presuppositions, on the most careful reading, do not coincide. The honest position is that the question of free will has at least two questions inside it, and that conflating them — as defenders of either side sometimes do — produces sharper rhetoric than light. The work of distinguishing them is, in the end, more interesting than any settled answer to the combined question would have been.