Picture a runaway trolley hurtling down a track. Five people are tied to the rails ahead. On a side track, there is one person — a child. You are standing at the lever. Pull it, and the child dies. Leave it, and five adults are killed. What do you do?
This is the Trolley Problem — a thought experiment introduced by philosopher Philippa Foot in 1967 and endlessly debated ever since. It seems, at first glance, like a maths question dressed up in moral clothing. But the more honestly you sit with it, the more it reveals about something far deeper: what we actually believe about human life, responsibility, and the ethics of inaction… and perhaps the need for substantial more information in this hypothetical. We may also need to ask questions about the assumptions of the question, such as, why do we assume five lives are worth more than one when no measurable value can be placed on a life?
Can we really put a number on a life?
The utilitarian answer is clean and confident: pull the lever. Five lives outweigh one. The arithmetic is straightforward. But this assumes something that many of us instinctively resist — that human lives can be compared, ranked, or traded against one another like items on a balance sheet.
Philosophers call this the problem of incommensurability: the idea that some values simply cannot be reduced to a common scale. Five dollars is worth more than one dollar. But is five lives worth more than one life? The moment we accept that framing, we have already conceded something profound — that people are, in some sense, fungible. Interchangeable. Countable.
Most of us feel, in our bones, that this is wrong. And yet we struggle to articulate why.
“The moment we start counting lives,
we have already made a philosophical choice —
one with consequences far beyond any trolley.”
Context is not a distraction — it is the point
Consider what changes when we add detail to the dilemma. The five adults are elderly, estranged from family, and living rough. The one person is a child with parents who adore them, siblings, a whole life ahead. Does this change the calculus? Should it?
Many philosophers would say we are introducing emotional noise — that the exercise demands we strip away context to test our principles in their purest form. But there is a compelling counter-argument: stripping away context does not purify the dilemma. It destroys it. Because in the real world, a human life does not exist in isolation. It exists within a web of relationships, responsibilities, histories, and futures. The loss of one child reverberates through a family — through parents, siblings, grandparents, friends — in ways that may echo for generations. The loss of five people who have drifted from the world still has weight, still has meaning, but the ripples spread differently.
This is not sentimentality. It is, arguably, a more honest and sophisticated form of moral reasoning than the cold arithmetic of classical utilitarianism.
The Epistemic Objection is a well-recognized challenge. If consequentialism tells you to produce the best outcomes, but outcomes are radically unknowable, then the theory may be practically action-guiding in name only. You can never actually know you’re doing the right thing, which some philosophers argue renders it useless as a decision procedure even if it’s correct as a moral theory.
This connects to what philosophers call the “cluelessness problem”, articulated rigorously by philosopher William MacAskill and others. The argument runs roughly:
- Consequentialism requires you to consider all consequences
- Long-run consequences of any action are deeply uncertain and potentially vast
- Therefore we are systematically clueless about what consequentialism actually requires of us
- This is not a minor inconvenience — it may be a fundamental flaw
The Responses Consequentialists Make
To be fair, consequentialists have replies:
- Expected value theory — you act on probabilities, not certainties. You use the best available estimate of outcomes
- Rule consequentialism — instead of calculating act by act, you follow rules that generally produce good outcomes, sidestepping some epistemic chaos
- Satisficing (good enough) consequentialism — you don’t need to maximise, just produce outcomes good enough
Is doing nothing really doing nothing?
Here is where the thought experiment takes its sharpest turn. Many people, when confronted with the trolley problem, feel that not pulling the lever is somehow morally safer — that inaction absolves them of responsibility. After all, they did not cause the trolley to exist. They did not tie anyone to the tracks.
But this reasoning deserves serious scrutiny. If you are standing at the lever, you have agency. You have full knowledge of the situation. You have the physical ability to intervene. At the moment you choose not to act, you are not opting out of the moral situation — you are making a deliberate choice within it. The mental process is identical to pulling the lever: you weigh your options and select one. The only difference is whether your hand moves.
The philosopher Peter Singer pushed this point hard with a simpler scenario: if you walked past a child drowning in a shallow pond and chose not to help because you didn’t want to ruin your clothes, almost everyone would consider you morally culpable. The water caused the drowning. But your character is revealed by what you were willing — and unwilling — to do.
“Inaction, when chosen consciously by someone with agency and the ability to intervene, is not a neutral act. It is a moral choice — and it tells us something real about who we are.”
This connects to the concept of moral cowardice: avoiding a difficult choice not because inaction is right, but because acting feels uncomfortable, costly, or risky. Virtue ethics — the tradition stretching back to Aristotle — holds that our character is defined not just by what we do, but by what we are willing to do, and what we are prepared to let happen when we could have stopped it.
Where we draw the line — and why it matters
There is another dimension to this worth sitting with: how we decide which living things deserve moral consideration in the first place. If we accept that all life has inherent value, where does that obligation end? Plants are alive. So are insects. So are fish, pigs, chimpanzees.
One principled place to draw the line is at sentience — the capacity to suffer, to experience fear, to feel pain. A heart, a brain, a nervous system suggest an inner life that a plant, however alive, does not possess. This is not a perfect boundary. But it is a reasoned one. And it connects back to the trolley problem in a meaningful way: if what matters morally is the capacity for suffering and the existence of relational bonds, then the texture of each life on those tracks — not just the number — is morally relevant information.
What the trolley problem is really asking
The Trolley Problem endures not because it has a correct answer, but because it refuses to let us hide. It exposes the gap between what we say we believe and what we are actually prepared to do. It forces us to confront whether our moral intuitions are consistent, and whether the frameworks we use — utilitarian calculation, duty-based ethics, virtue and character — can survive contact with a real dilemma.
The most honest response is not to solve the problem, but to feel its full weight. To resist the urge to reduce it to arithmetic. To acknowledge that human lives are not units of currency, that context is not a distraction, and that the decision to do nothing is itself a decision — one that reflects, for better or worse, something true about the person standing at the lever.
There’s always a “trolley” coming.
The trolley problem doesn’t tell us what to do; it tells you something about who you are and forces genuine reflection. That’s closer to virtue ethics territory, ironically — where the question shifts from what produces the best outcome to what does this choice reveal about, and do to, my character.