The Trolley Problem: Testing Moral Intuitions

A Trolley, a Switch, and Sixty Years of Argument

In 1967, British philosopher Philippa Foot introduced a trolley to academic philosophy — and it has not left since. Foot was working on the doctrine of double effect, a principle in Catholic moral theology that distinguishes between harm intended as a means and harm foreseen but not intended as a consequence. To illustrate the distinction, she described a runaway tram heading toward five workers on a track. A bystander could divert it to a side track, killing one worker instead. Most people's intuition: pull the switch. Five lives saved at the cost of one seems mathematically and morally justified.

Then Judith Jarvis Thomson added a variation in 1985. Same trolley, same five workers — but now a large man stands on a footbridge overhead. The only way to stop the trolley is to push him off the bridge, using his body as an obstacle. Most people's intuition: don't push. And here is the puzzle: the arithmetic is identical in both cases. One death prevents five. Why does the action feel profoundly different?

What the Trolley Problem Reveals About Moral Frameworks

The divergence in intuitions maps onto two of the oldest frameworks in moral philosophy. Consequentialism — particularly in its utilitarian form developed by Jeremy Bentham and John Stuart Mill — holds that the morality of an action is determined entirely by its outcomes. More welfare produced equals more moral goodness. By this standard, pulling the switch and pushing the man are equivalent: both save five lives at the cost of one.

Deontological ethics, most associated with Immanuel Kant, holds that some actions are intrinsically wrong regardless of consequences. Kant's categorical imperative demands that we act only according to principles we could universalize — and using a person merely as a means to an end violates the principle of treating rational beings as ends in themselves. Pushing the man instrumentalizes him as a trolley stopper, which Kant's framework prohibits even if the consequences are better.

The Neuroscience of Moral Intuition

In 2001, philosopher Joshua Greene and neuroscientist Jonathan Cohen used fMRI scanning to look inside the brain during trolley-style dilemmas. Their findings were striking. The footbridge scenario — pushing the man — activated regions associated with emotional processing, including the medial prefrontal cortex and posterior cingulate cortex. The switch scenario activated regions more associated with deliberate calculation. The brain, it seemed, was running two different systems depending on the personal versus impersonal nature of the harm.

Greene proposed a dual-process theory of moral judgment: fast, emotional intuitions operate in parallel with slower, more deliberate utilitarian calculations, and they frequently conflict. The footbridge scenario triggers a visceral "don't touch a person to harm them" response that overrides the cooler consequentialist arithmetic. This doesn't settle whether the emotional or the rational response is morally correct — but it explains why smart people can disagree.

Variants That Complicate the Picture Further

Philosophers and psychologists have proliferated trolley variants to probe the boundaries of moral intuition. Each new scenario reveals new inconsistencies and suggests the underlying principles are more complex than any single framework captures.

• The loop track: The side track loops back to the main track. The only way the one worker's body stops the trolley is if it hits them. Does this change things? Many who endorsed the switch variant resist this one, because the death is now mechanically necessary rather than incidental.
• The transplant surgeon: A doctor can save five patients by killing one healthy person and harvesting their organs. Nearly everyone rejects this — yet it is structurally similar to pushing the man. Why? The institutional context of medicine and trust carries enormous moral weight that pure arithmetic ignores.
• The fat villain: The person on the bridge is the villain who set the trolley in motion. Willingness to push increases sharply, suggesting desert and moral responsibility alter intuitions.
• The remote button: Instead of physically pushing, pressing a button drops the man through a trapdoor. Willingness to act increases as physical distance grows, suggesting our moral intuitions are partly tracking physical intimacy of harm, not just causal structure.

Cross-Cultural Studies and Universal Intuitions

A landmark cross-cultural study by Awad and colleagues, published in PNAS in 2018 as part of the Moral Machine experiment, gathered 40 million moral decisions from 233 countries via an online platform. The study used autonomous vehicle scenarios — whom should a self-driving car sacrifice in an unavoidable crash? The results showed both universals and cultural variations.

Why Philosophy Needs Thought Experiments

Critiques of the trolley problem are legitimate. Real ethical situations are messier — we rarely have certainty about outcomes, we have relationships with the people involved, and institutional and legal structures constrain choices in ways the thought experiment strips away. Some philosophers argue that the very artificiality of trolley cases makes them poor guides to practical ethics.

But that artificiality is also the point. By stripping away extraneous details, thought experiments isolate specific variables and force us to confront whether our moral principles are consistent. The trolley problem doesn't tell us what to do when an autonomous vehicle's brakes fail — but it does reveal that most of us hold moral intuitions that cannot all be simultaneously satisfied by any single framework. Living with that tension, and reasoning carefully within it, is what moral philosophy is for.