Consequentialism and Deontology: Two Theories of Right Action

A runaway trolley is hurtling toward five workers on the track. You stand beside a lever that can divert it onto a side track, where it will kill one. Should you pull the lever? Most people say yes. Now imagine instead that you stand on a footbridge above the track, beside a large stranger whose body would stop the trolley if you pushed him over. Same arithmetic — one death to save five — and yet most people recoil. The two great traditions in modern moral philosophy try to make sense of why.

Consequentialism holds that the rightness of an action depends entirely on the goodness of its outcomes. The classical version, utilitarianism, was sharpened by Jeremy Bentham and John Stuart Mill in the nineteenth century: act so as to produce the greatest sum of well-being across all those affected. On this view, the lever and the push are morally equivalent. Both trade one life for five; both are obligatory. Anything else is sentimentality dressed up as principle.

Deontology, most famously associated with Immanuel Kant, denies that outcomes are the whole story. Some acts are wrong in themselves, regardless of what good they might bring about. Kant's categorical imperative asks whether you could will the principle behind your action to be a universal law, and whether your action treats other people as ends in themselves rather than as mere means. Pushing the stranger off the bridge uses his body as a tool — a trolley-stopper — without his consent. That, for the Kantian, is the morally decisive feature, and no quantity of saved lives can wash it out.

The split runs deeper than the trolley case. Consequentialism is forward-looking: what matters is the world your action helps bring about. Deontology is constraint-based: certain things you may not do, even to bring about a better world. A consequentialist can in principle endorse framing an innocent person if doing so prevents a riot that would kill many. A deontologist holds that the prohibition on framing the innocent is not the kind of thing that can be outweighed.

Each view buys its clarity at a cost. Consequentialism is haunted by the worry that it is too demanding and too permissive at once: too demanding because it seems to require you to maximize good at every moment, leaving no room for ordinary projects and attachments; too permissive because, in the right circumstances, it appears to license betrayals, lies, and even killings that strike most people as monstrous. Defenders respond by distinguishing act-consequentialism from rule-consequentialism, or by arguing that, properly calculated, the long-run consequences of monstrous acts are themselves monstrous.

Deontology faces the mirror complaint. If certain acts are forbidden whatever the stakes, the theory can seem to prize the agent's moral cleanliness over the lives of strangers. Bernard Williams's worry runs the other way: a morality that demands you ignore your own deep commitments — that treats you as a neutral calculator of universal duties — may misdescribe what it is to be a person at all. Deontologists reply that absolute constraints are precisely what protect persons from being absorbed into someone else's calculation, and that a moral life without such limits would have no spine.

It is tempting to declare one theory the winner, or to split the difference with a tidy compromise. The harder and more honest move is to notice that the two theories are tracking different features of moral life that really are in tension. We do care about outcomes; a moral view indifferent to suffering is not serious. We also care about how those outcomes come about, and about what we are willing to do to other people in pursuit of them. The trolley problem is not a puzzle with a hidden solution. It is a diagnostic, and what it diagnoses is that our moral concepts were not built to give a single answer when these two cares pull apart.