A runaway trolley is heading toward five people tied to the tracks. You're standing next to a lever that can divert it onto a side track, where only one person is tied down. Do you pull the lever, killing one to save five? That's the trolley problem, and it's been tormenting philosophy students and dinner party guests since Philippa Foot introduced it in 1967. Most people say they'd pull the lever. But the variants of this thought experiment reveal that our moral instincts are far messier than simple math.

The Classic Setup

The original version is deceptively simple. Five lives versus one. Pure numbers say pull the lever. And about 85-90% of people agree when surveyed. The math checks out, the logic feels clean, and you walk away thinking moral philosophy isn't that complicated after all.

Then the variants start.

The Fat Man on the Bridge

Same trolley, same five people about to die. But this time there's no lever. You're standing on a bridge above the tracks next to a large man. If you push him off the bridge, his body will stop the trolley before it reaches the five. He dies, they live. Same outcome as before: one dead, five saved.

Suddenly, the numbers don't feel so clean. Only about 10-15% of people say they'd push the man. The math is identical, but the act of physically shoving someone to their death triggers something the lever scenario doesn't. Philosophers call this the difference between doing and allowing. Pulling a lever feels indirect. Pushing a person off a bridge feels like murder, even if the body count is lower.

This is where utilitarianism and deontological ethics crash into each other. A utilitarian says the outcome is all that matters: five alive beats four alive, period. A deontologist says certain actions are inherently wrong regardless of outcome, and using a person as a tool to stop a trolley crosses that line.

The Surgeon Variant

Here's where it gets truly uncomfortable. A surgeon has five patients who will die without organ transplants. A healthy patient walks in for a routine checkup. The surgeon could kill the healthy patient, harvest their organs, and save all five. Same math: one death, five lives saved.

Almost nobody endorses this one. If you pulled the lever in scenario one, you now have to explain why the surgeon scenario is different. The numbers are the same. The outcome is the same. But something about a doctor murdering a patient in a hospital feels categorically wrong in a way that diverting a trolley doesn't.

The discomfort you feel right now? That's the whole point. The trolley problem isn't trying to give you an answer. It's trying to show you that your moral reasoning isn't as consistent as you think it is.

When you play through trolley problem scenarios interactively, the pressure of making quick decisions reveals instincts that careful philosophical reasoning can hide. Your gut reaction and your reasoned answer are often different.

What Most People Actually Choose

Decades of research on the trolley problem have produced remarkably consistent results across cultures. The pattern holds whether you survey people in the US, China, Brazil, or India:

What's interesting is that the explanations vary wildly. Some people cite the physical directness of pushing. Others point to consent, or the difference between redirecting existing harm versus creating new harm. There's no single principle that cleanly explains why most people accept scenario one and reject scenario two. We just feel it. The moral dilemmas game lets you explore dozens of these scenarios and see how your choices compare to everyone else's.

Why Self-Driving Cars Made This Real

The trolley problem used to be purely academic. Then autonomous vehicles showed up and turned it into an engineering question. If a self-driving car's brakes fail, should it swerve into a wall (killing the passenger) or continue forward (killing a pedestrian)? Someone has to program that decision tree.

MIT's Moral Machine project surveyed millions of people worldwide on exactly these scenarios. The results were messy. Different cultures prioritized different things: some valued saving the young over the old, some valued saving more people over fewer, some valued passengers over pedestrians. There was no universal consensus.

This is the trolley problem's real legacy. It was never about trolleys. It's about forcing us to confront the fact that moral intuitions are inconsistent, culturally shaped, and deeply resistant to simple rules. When you're playing would-you-rather scenarios, you're exercising the same decision-making muscles.

Face the Trolley Problem Yourself

Work through escalating moral dilemmas and see how your choices compare to thousands of other players.

Play the Trolley Problem

Utilitarian vs. Deontological: The Real Divide

The utilitarian position is straightforward: maximize total well-being. If pulling the lever saves a net four lives, pull it. If pushing the man saves a net four lives, push him. The math doesn't care about your feelings.

The deontological position (associated with Kant) says that using people as mere means to an end is always wrong. The man on the bridge isn't a trolley-stopping tool. He's a person with rights that don't evaporate because five other people are in danger.

Most people are intuitive deontologists in practice. We follow rules ("don't push people off bridges") even when breaking them would produce better outcomes. But we're utilitarian enough to pull a lever when the harm feels sufficiently indirect. We live in the messy middle, and the trolley problem is the clearest mirror for that inconsistency.

If you want to stress-test your own moral framework further, the personality test can reveal patterns in how you approach decisions that you might not consciously recognize.

The trolley problem has no right answer. That's what makes it useful. It doesn't tell you what to think. It shows you how you already think, and dares you to explain why.