The Confluence of Character and Contingency: Re-evaluating Responsibility in the Face of Moral Luck — Epoche C1
A prevalent intuition in moral philosophy holds that individuals are morally responsible for actions and their consequences only insofar as these are under their control. Call this the control principle. It underwrites what philosophers term an internalist view of moral responsibility — "internalist" here meaning that moral assessment attaches only to what lies inside the agent: the will, the intention, the settled traits of character — and not to the external upshots that the world happens to deliver. On such a view, an agent's moral worth is determined by internal states and choices rather than by circumstances beyond their volitional grasp. The principle sounds like common sense; it is, after all, the reason we do not blame people for the weather, and the reason duress, accident and insanity function as excuses. Yet it encounters considerable friction when confronted with the phenomenon of moral luck . The term was introduced by Bernard Williams in a paper delivered to the Aristotelian Society in 1976, and the problem was given its now-standard formulation in Thomas Nagel's reply at the same symposium: moral luck obtains wherever a significant aspect of what someone does depends on factors beyond their control, and yet we continue — stubbornly, and on reflection not unreasonably — to treat them as an object of moral judgment for that very aspect. Where the control principle comes from The purest statement of the principle is not Aristotelian but Kantian, and it is worth seeing why, because the distinction matters later. Kant opens the Groundwork of the Metaphysics of Morals (1785) with the claim that nothing in the world is good without qualification except a good will — the settled resolve to act from duty. Crucially, he adds that if some special disfavour of fortune were to deprive that will of any power to accomplish its ends, it would still shine "like a jewel" in its own right. The image encodes a complete doctrine: success adds nothing to moral worth and failure subtracts nothing, because worth resides wholly in the willing, which is the one thing fortune cannot touch. Luck, for Kant, is morally inert. Whatever one thinks of the doctrine, its appeal is plain: it makes moral standing something no accident can confiscate, and it makes moral assessment fair in the way an examination is fair when every candidate sits the same paper. The difficulty is that our actual practices of praise and blame refuse to honour it — and that the principle, applied with full consistency, consumes far more than its defenders intend. Two drivers, one lapse: resultant luck Consider the stark contrast between two drivers, both negligent in the same way. Driver A, distracted by a momentary lapse of attention, swerves but narrowly misses a pedestrian, and drives home shaken. Driver B, distracted in precisely the same manner, swerves and strikes and kills a pedestrian. By construction the two lapses are identical: same inattention, same speed, same road. Everything within the drivers' control is the same; the only difference is whether a pedestrian happened to be on that stretch of pavement — a fact neither driver controlled. This is what Nagel calls resultant luck: luck in how one's action turns out. If moral responsibility were solely a function of controllable factors, the two drivers would be exactly as culpable as one another, and any difference in our response would be irrational. Yet the difference in response is profound, and it is not confined to unreflective sentiment — it is built into law. English law, to take one instrument, grades the two cases as different offences: dangerous driving is one crime under the Road Traffic Act 1988, while causing death by dangerous driving is another, and the second — the same driving, plus a death the driver did not choose — carries a maximum sentence many times heavier than the two years available for the first. Most jurisdictions draw the same kind of distinction, most visibly in punishing completed crimes more severely than attempts that failed for reasons external to the agent. The killer-driver also condemns himself more severely: he will feel something the lucky driver is spared, and few of us regard that extra anguish as a confusion on his part. Nor is the observation new. Adam Smith, in The Theory of Moral Sentiments (1759), devoted a section to what he called the irregularity of our moral sentiments: we officially profess that only the intention matters, yet in practice our gratitude and resentment swell or shrink with the actual consequences. Smith did not think the irregularity could be reasoned away; he even argued that it is useful, since it directs human justice at deeds and their effects rather than licensing an inquisition into hearts. The problem of moral luck, then, is not a modern paradox conjured from thin air; it is a standing feature of moral life that the control principle has never managed to legislate out of existence. The problem stated exactly It helps to see that the difficulty has the form of a genuine trilemma — three claims, each individually compelling, that cannot all be true together: The control principle: agents are morally assessable only for what depends on factors under their control. The stipulation: the two drivers are identical in every respect that was under their control. The practice: Driver B is more blameworthy than Driver A. Any two of these entail the falsity of the third. One might therefore hope to escape by simply dropping (1). But Nagel's deeper point is that the control principle is not an optional philosopher's axiom; it is itself an entrenched intuition, the very one that grounds our system of excuses. We withdraw blame when we learn the harm was caused by a seizure, or that the agent was coerced, precisely because those factors were outside the agent's control. Worse, the principle is expansionist. Once we begin subtracting from the moral ledger everything the agent did not control — the outcome, then the