Questions What makes something right?Turning the trolley: one death rather than five

What makes something right?Consequences

Turning the trolley: one death rather than five

The trolley problem

Philippa Foot 1920 · 2010Invented the trolley problem and defended a virtue ethics grounded in human nature.

See profile

What it claims

The driver may, and even must, turn the tram onto the track where it will kill only one person.

Why

When every option causes harm, choose the one that causes least: five deaths are a greater harm than one.

The same idea through history

Philippa Foot1967In an article on abortion, Foot imagines the driver of a runaway tram who can only switch tracks: five workmen on one side, a single one on the other. She compares him with a judge who would have an innocent executed to calm a riot.

Judith Jarvis Thomson1976"Killing, Letting Die, and the Trolley Problem" gives the problem its name. In 1985 Thomson herself would say she was "blundering around in the dark" in it.

Judith Jarvis Thomson1985"The Trolley Problem": a mere bystander, not the driver, now holds the switch, and the "fat man" who could be pushed off a footbridge appears. The version discussed today.

Judith Jarvis Thomson2008"Turning the Trolley": Thomson changes her mind. The bystander may not turn the trolley (only the driver may).

In short

"Turning the trolley: one death rather than five" is an argument by Philippa Foot (Foot, "The Problem of Abortion and the Doctrine of the Double Effect" (1967); Thomson (1976, 1985), 1967) on the question "What makes something right?" (answer: consequences). Its claim: The driver may, and even must, turn the tram onto the track where it will kill only one person. According to AskPhilosophy's logic check (Z3 solver), the argument is valid only if one grants a premise the author leaves unstated (P3). AskPhilosophy presents 3 objections, 3 of them with a documented reply, and the original text behind each premise, checked word for word.

Foot, "The Problem of Abortion and the Doctrine of the Double Effect" (1967); Thomson (1976, 1985)

Valid with an implicit premise

A runaway tram is hurtling down a slope. On the track, five workmen will not be able to get out of the way. The driver can still throw the switch and take another track, where a single workman is working. Philippa Foot invented this story in 1967 to test our moral rules, and almost everyone gives the same answer: turn. But why? The simplest answer is the one from consequences: when every option causes deaths, you must choose the one that causes fewest, and five lives count for more than one. The case looks easy; it is by comparing it with others (the judge, the surgeon, the fat man on the footbridge) that it becomes one of the most famous puzzles in philosophy. Note: Foot herself does not defend the idea that we must always count lives. For her, the driver must turn because he is going to kill someone anyway, and must then do as little harm as possible.

click a premise to reject it

If you accept every premise, logic forces you to accept the conclusion. So the debate is about the premises.

Checked by a logic solver: the premises cannot all be true while the conclusion is false.
Without the implicit premise: Without P3, the computer finds a world where the driver follows the lesser-harm rule, but where the death of five is not a greater harm than the death of one. Then nothing requires turning. That is John Taurek's position: for him, each person loses just as much, and we should flip a coin.

Valid with P3. P1 describes the situation, P2 gives the lesser-harm rule, P3 says which of the two harms is the lesser. Foot does not state P3: it goes without saying for almost everyone. The debate turns on two points: P2 (does the lesser-harm rule hold everywhere? the fat man, the judge, the surgeon) and P3 (do lives really add up? Taurek).

Analogy. A firefighter can open only one door before the building collapses: behind one there are five people, behind the other just one. Almost everyone thinks he must open the first. The trolley asks the same question, except that here you are not choosing whom to save: you are choosing whom to kill.

Objections

Each objection is a conversation: click to see who replies and where the debate stands.

Judith Jarvis Thomson (the fat man) 1985 · attacks premise 2

You are on a footbridge over the track. The trolley is heading for five workmen. Next to you a very fat man is leaning over the railing: if you push him, his body will stop the trolley. One death instead of five, exactly as at the switch. If the lesser-harm rule held everywhere, you would have to push him. Yet everyone Thomson put the case to says it is forbidden. The same goes for the surgeon who would cut up a healthy patient to give his organs to five sick people. So we need to explain why the number decides at the switch and not elsewhere.

Foot replies

Philippa Foot 1967 · Foot's anticipated reply (the judge)

The driver has two duties of the same kind: not to injure five people, not to injure one. He cannot keep both, so he does the least harm he can. The judge who would have an innocent executed to stop a riot (like the person who pushes the fat man) is weighing something else: not harming someone, and helping others. And the duty to help is not as strong as the duty not to harm. So we cannot argue from the driver's case to the judge's.

Judith Jarvis Thomson 1985 · The bystander at the switch

Change one detail: the driver has fainted, and a mere bystander happens to be by the switch. If he does nothing, he kills nobody, he lets the five die. On Foot's view, he should therefore hold back, since killing is worse than letting die. Yet almost everyone thinks he may turn, and so does Thomson. The idea that "killing one is worse than letting five die" is therefore false, and it cannot explain the fat man case.

See the rest of the debate (5)

Judith Jarvis Thomson 2008 · Thomson changes her mind

Imagine the bystander has a third option: sending the trolley onto himself. Most agree he is not required to sacrifice himself. But then by what right would he impose that sacrifice on the lone workman? Thomson concludes that the bystander may not turn. Only the driver, who would otherwise kill the five himself, may.

Joshua Greene 2001 · The neuroscience reply (Greene)

By observing the brain while people answer these dilemmas, Greene shows that they engage the emotions to different degrees, and that this influences judgement. Agreeing to throw the switch is a simple piece of reasoning; refusing to push the fat man comes from an automatic emotional response, triggered above all by pushing someone with one's own hands ("personal force"). Yet that difference should not change what is right. So the fat man intuition does not refute P2: it is the intuition that must be corrected.

Selim Berker 2009

The brain images do no work in this argument. That an intuition comes from an emotion does not show it is false. Everything rests on a purely moral premise: "personal force does not matter morally". And that premise has to be defended by philosophy, not by science.

Joshua Greene 2010

Greene grants the principle: a moral conclusion cannot be deduced from scientific facts. But science still does some work: it shows what our intuitions are really responding to. If the reaction to the fat man tracks "personal force", and that force should make no difference, the intuition loses its authority.

Awad, Dsouza, Shariff, Rahwan and Bonnefon (Moral Machine) 2020 · The data: 70,000 people in 42 countries

MIT's Moral Machine website put the three cases to 70,000 people in 42 countries. Everywhere the order is the same: people accept sacrificing one for five more at the switch (81% on average) than in the loop (72%), and more in the loop than on the footbridge (51%). So the fat man divides people, it does not command unanimity; but the gap with the switch exists in every country.

The debate goes on (our summary) The debate goes on. For consequentialists, the fat man mainly shows that our intuitions are fragile. For their opponents, it shows that the number of deaths is not everything: how a person is treated also matters (pushing him, using his body). Another famous explanation is the doctrine of double effect: at the switch, the workman's death is only a foreseen effect, on the footbridge it is the means (see the argument "The doctrine of double effect"). Thomson's 2008 change of mind remains a minority view: most philosophers, like most people, still think the bystander may turn.

John Taurek 1977 · attacks premise 3

Deaths do not add up. If five people die, each loses his life, but nobody suffers a loss five times greater than the lone man would. So there is no reason to prefer the five to the one. To give everyone the same chance, Taurek proposes flipping a coin, and he would do so even if it were fifty people against one.

Foot replies

Derek Parfit 1978

Nobody suffers a loss five times greater, true. But that proves nothing: if the lone man dies, he loses as much as any one of the five; if the five die, together they lose more. Refusing to add up the losses of different people leads to absurd conclusions, for example drawing lots between five lives and one man's umbrella.

The debate goes on (our summary) Taurek has few followers, but his objection had consequences. It forced people to say why numbers count without treating the five as one big victim. Some, like Scanlon, look for an answer that does not rely on addition: each person in the larger group can complain if treated as though they were not there. Taurek's coin toss also keeps an appeal: it treats everyone equally.

Critique of laboratory dilemmas 1967 · attacks premise 1

In real life one is never sure of anything: the lone workman might get out of the way, the trolley might derail, the fat man might not be enough. Foot admits it herself: the certainties philosophers postulate hardly ever exist. P1 ("there is no third option") is true only in a made-up story. So our answers to the dilemma might depend on what we imagine beyond the story, not on moral principles.

Foot replies

Joshua Greene 2009

It has been checked: participants were asked what they really believed (would the fat man be enough?) and this was taken into account. These beliefs play a small role, but they explain only about 10% of the gap between the switch and the footbridge.

The debate goes on (our summary) Defenders of the dilemmas reply that they are made unreal on purpose, like a laboratory experiment: everything is fixed but one detail to see whether that detail matters. Critics reply that a morality built on impossible stories may guide us badly in real decisions, in hospital or when programming a self-driving car, where nothing is certain.

What today's voices say

Defend it

Sam HarrisThe same 2025 episode with David Edmonds covers 'trolleyology'. The page does not give his verdict.

Peter SingerIn "Ethics and Intuitions" (2005) he argues that gut reactions in cases like the trolley problem may be evolutionary leftovers, so they should not decide what is right.

Michael SandelLecture 1 opens with the choice of killing one person to save five, to show how moral intuitions clash. In Bangalore (2017) he asks why people who accept one death for five in the first case drop the same principle when it means pushing a man.

Criticize it

Yuval Noah HarariOf trolley-style dilemmas: philosophers had to think very hard to invent such extreme scenarios, which almost never happen in reality; you always have more options.