A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
Moral responsibility requires appropriate control over one's actions—typically understood as the ability to understand what one is doing and to govern one's behavior by moral and practical reasons. The control condition distinguishes responsible action from accident, coercion, or involuntary conduct. Debates arise over how much control suffices (compatibilist versus libertarian views) and whether responsibility requires the absence of prior causal factors beyond the agent's control.
Consider someone forced at gunpoint to commit a crime: are they morally responsible? Why or why not? What conditions would restore responsibility?
The control condition doesn't require absolute freedom from prior causes; compatibilists hold that responsibility and determinism can coexist as long as the agent's own deliberation (rather than external coercion) guides action.
Moral responsibility is not all-or-nothing — it is grounded in the kind of control an agent exercises over their actions. You already understand from your study of moral agency that being a moral agent means being the sort of creature who can understand reasons and respond to them. The control condition sharpens this: responsibility tracks whether that capacity was actually operative when you acted. When you trip on a loose stair, your body does something — but no one blames you, because your deliberative agency wasn't in the driver's seat. When a surgeon makes a careful incision, their agency is fully operative, and if the incision is negligent, they bear responsibility. The contrast reveals what control is doing: it marks the boundary between what we do as agents and what merely happens to us or through us.
Two forms of control appear in the literature. Guidance control asks whether the mechanism that produced your action was your own, reasons-responsive deliberation — not external manipulation, compulsion, or accident. Regulative control asks whether you could have done otherwise. Compatibilists tend to emphasize guidance control, arguing that what matters is whether you were acting through your own reasons-responsive mechanism, not whether some prior causal chain was involved. On this view, a heroin addict acting under compulsion lacks guidance control — their mechanism is hijacked by addiction — while someone who deliberates freely and acts on their reasons has it, even if determinism is true and their deliberation was itself the result of prior causes. The gunpoint case from the learning prompt is illuminating: the coerced agent's deliberation is operative (they understand the situation, weigh the threat, choose), but external force structures the choice so severely that responsibility is mitigated or eliminated. Notice that this is a matter of degree, not binary.
The connection to free will and determinism becomes vivid here. Libertarians about free will argue that genuine control requires the ability to have done otherwise in exactly the same circumstances — a kind of causal openness that determinism would rule out. If your action was the inevitable product of prior causes, you never really had a choice, and control is illusory. Compatibilists respond that this sets the bar for control too high. What matters for moral responsibility is the kind of control that distinguishes agents from billiard balls — the capacity to adjust behavior in response to reasons, evidence, and social norms. A person who would have acted differently if offered different reasons has the relevant control, even if in the actual circumstances they could not have done otherwise.
Moral luck exposes a persistent tension. Consider two drivers who both behave recklessly, but only one encounters a pedestrian and causes an accident. We tend to hold the unlucky driver far more responsible than the lucky one, even though their actions and intentions were identical up to the moment of the accident. Bernard Williams and Thomas Nagel argued this reveals that our responsibility practices track outcomes that were partly beyond the agent's control — a structural feature that sits awkwardly beside the control condition. If responsibility requires control, why does luck about outcomes matter at all? Compatibilists and control theorists respond differently, but the puzzle remains one of the most productive in ethics: it forces us to say precisely what the control condition is really protecting against, and how much moral weight we should give to agent-external circumstances when assessing blame and praise.
Topics in reflective domains aren't scored by quiz answers. Read, reflect, and mark when you've thought it through.