Would You Pull the Lever? The Trolley Problem Explained

 

A runaway trolley is barreling down a track toward five people tied up and unable to move. You're standing next to a lever. Pull it, and the trolley switches to a side track, saving the five people, but there's one person tied up on that track who will die instead. Do you pull the lever?

Most people say yes without much hesitation. Five lives outweigh one. It seems like straightforward math.

Now change the scenario slightly. Same runaway trolley, same five people about to die. But this time you're standing on a bridge above the track next to a very large stranger. The only way to stop the trolley is to push this stranger off the bridge, using his body to physically block the trolley before it reaches the five people. He'll die, but the five will live. Same math, five saved instead of one. Do you push him?

Most people say no, and often quite forcefully. But here's the uncomfortable part. The numbers are identical in both cases. One person dies, five people live. If you pulled the lever in the first version but wouldn't push the stranger in the second, you owe yourself an explanation for why, because it clearly isn't the arithmetic driving your answer.

This thought experiment, first introduced by the philosopher Philippa Foot and later developed further by Judith Jarvis Thomson, has become one of the most famous tools in moral philosophy specifically because it exposes a gap between two ways people reason about ethics, often without realizing they're using different systems depending on the situation.

Two Competing Ways to Judge Right and Wrong

Utilitarianism says the morally right action is whichever one produces the best overall outcome, typically measured by total wellbeing or minimized suffering across everyone affected. Under strict utilitarian logic, the lever case and the bridge case are morally identical. Five lives outweigh one either way, so if you should pull the lever, you should also push the stranger. The method of causing the death is irrelevant. Only the final tally of who lives and who dies matters.

Deontological ethics says the morality of an action depends on the nature of the act itself, not just its consequences. Certain things are wrong regardless of the good outcome they might produce, and using a person merely as a tool to achieve an outcome, rather than treating them as an end in themselves, is one of the clearest violations of this. This tradition traces back to Immanuel Kant, who argued that treating a person purely as a means to someone else's benefit, rather than respecting them as an autonomous being with their own worth, is a fundamental wrong no matter how good the resulting outcome is.

This is why the bridge case feels so different from the lever case, even though a strict utilitarian calculation says they shouldn't. In the lever version, the one person's death is an unfortunate side effect of redirecting an existing threat. In the bridge version, you're actively using a person's body as an instrument, a means to an end, to physically stop the trolley. Something about direct, physical use of another human being as equipment for saving others triggers a much stronger moral resistance than diverting a threat that was already in motion.

The Doctrine of Double Effect

Philosophers have proposed a more specific principle to explain this gap, sometimes called the doctrine of double effect. It distinguishes between harm that is intended as a means to an end, and harm that is merely a foreseen but unintended side effect of achieving that end through other means. Pulling the lever foresees that one person will die, but that death isn't the mechanism by which the five are saved. It's a side effect of redirecting the trolley. Pushing the stranger, by contrast, requires his death as the actual mechanism of salvation. Without his body stopping the trolley, the plan doesn't work at all. His death isn't a side effect. It's the method.

This distinction, between foreseeing harm and intending it as a means, has shown up in real ethical debates well beyond runaway trolleys, including discussions in medical ethics about the difference between giving a patient pain medication that hastens death as an unintended side effect versus directly ending a patient's life as the mechanism of relieving their suffering.

Why This Thought Experiment Refuses to Die

It would be easy to dismiss all this as an overly cute philosophy classroom game with zero real stakes, since nobody actually encounters runaway trolleys. But the underlying structure shows up constantly in situations most people will actually face.

Self driving car programming is probably the most direct modern version. If an autonomous vehicle's brakes fail and it must choose between staying its course and hitting several pedestrians, or swerving and killing its own passenger instead, engineers are quite literally writing code that encodes an answer to the trolley problem, whether they frame it that way or not. Military decisions about acceptable civilian casualties to prevent a larger attack run through the exact same tension between total outcomes and the method by which harm occurs. Even public health policy, choosing to allocate scarce resources in a way that saves more lives overall while explicitly sacrificing others, runs on this same fault line.

What Your Answer Actually Reveals

The real value of the trolley problem isn't landing on a final correct answer, since philosophers themselves remain split between committed utilitarians and committed deontologists after decades of debate. The value is in what your gut reactions to the different versions reveal about your own moral reasoning. Most people aren't consistent utilitarians or consistent deontologists. They're something messier, willing to accept certain deaths as acceptable side effects of a good outcome, but recoiling from the exact same math when it requires directly using a person's body as a tool.

That inconsistency isn't necessarily a flaw to be corrected. It might reflect something genuinely important that pure outcome counting misses, some instinct about human dignity that resists being reduced entirely to arithmetic. Or it might just be a bias worth examining rather than trusting blindly. Either way, the next time someone asks whether you'd pull the lever, the more interesting question sitting underneath it is why your answer to that version might not match your answer to the bridge.