The Suitcase Trolley Problem

Why Your Conscience Might be Killing People. 

There is a question that cuts to the heart of moral philosophy: when we decide what is right and wrong, are we ultimately answerable to outcomes, to how things actually turn out for real people, or to unbreakable principles that hold no matter the consequences?

The view that principles come first, that some actions are simply forbidden even if they produce better results, has enormous intuitive appeal. Most of us feel in our bones that there are things you just shouldn’t do to people, even to save others. Consider the sheriff who knows the man in his jail is innocent, but convicting him is the only way to prevent a riot that is likely to lead to the deaths of several people. Framing an innocent person feels deeply wrong in a way the prospective body count cannot justify.

This instinct powers a whole tradition in moral philosophy (often called deontology). But a thought experiment, sharpened by philosopher Kacper Kowalczyk in his 2022 paper “People in Suitcases”, puts it under extraordinary pressure.

The Classic Trolley vs. The Suitcase Twist

Imagine a runaway trolley heading towards two people tied to the track. You stand on a bridge above them. Beside you is a stranger. If you push him off, his body will stop the trolley and save the two below. One dies instead of two. This is a version of the classic “trolley problem”. Most people say: don’t push. There’s a real moral difference between redirecting a harm that was already coming and actively using a person as a tool, even a life-saving one. To push is to treat the stranger merely as a means. That violates his dignity in a way arithmetic can’t erase.

Now the twist: All three people are inside identical suitcases. You have no idea who is in which, and neither do they have any way of knowing which suitcase they are in. One suitcase sits balanced on the bridge; two are on the track.

Should you push?

Think carefully. Whoever is in the bridge suitcase has a 1-in-3 chance of dying if you push, because any given person has a 1-in-3 chance of being on the bridge. If you do nothing, that same person has a 2-in-3 chance of dying (the odds they’re on the track). The same arithmetic applies to every single person involved. Each one has a better chance of survival if you push. From behind a veil of ignorance, every individual might reasonably want you to push. Yet the physical act is identical: you’re still pushing a human body off a bridge to stop the trolley and save two others. The only difference is your knowledge of who is who, and their knowledge of where they are.

The Knowledge Problem

This is where deontology struggles. If pushing is wrong because it uses a person as a mere means, why does ignorance suddenly make it permissible? The suitcases don’t change what happens to the people, only what you know.

One response is to bite the bullet: pushing is wrong either way. But that means you should let the trolley run, even though doing so gives every person involved a worse chance of survival, and even though each of them, reasoning about their own fate from ignorance, would call it the worse bet.

The alternative is to say pushing is right with suitcases (behind the veil of ignorance) but wrong when the veil is removed. This draws the moral line at knowledge alone.

That leads to strange places. Suppose it takes you an hour to push the heavy suitcase, but after thirty minutes a crack opens and you see exactly who is inside. On this view, the act flips from permissible to impermissible, purely because of new information. Nothing physical has changed.

Or imagine you could push now (while ignorant) or wait until everyone climbs out and becomes identifiable. You might feel driven to push immediately, precisely to circumvent the moral constraint that would kick in once you can see their faces. A principle meant to protect human dignity ends up being gamed by the very person trying to follow it.

Now consider another classic version of the Trolley Problem: You can pull a lever and redirect the trolley from killing two people to killing one. Does it make a difference now if the people are anonymous, trapped behind a veil of ignorance inside their very own suitcases? Is your answer the same as before? Does the anonymity of the suitcases make a difference to your decision?

Why This Matters

There’s an old criticism that outcome-focused ethics treats people as vessels for welfare rather than ends in themselves. Kowalczyk’s suitcase scenario suggests the criticism may have the wrong target. A framework that prohibits the very action each individual would recognise (from their own perspective) as giving them the best odds of survival isn’t obviously the one showing deeper respect for persons.

Rules have their place. The gut feeling that some things simply aren’t done — you don’t frame the innocent, you don’t treat people as raw material — makes genuine moral sense. In the real world, routinely overriding it for “good consequences” could well make society worse. But when the rules demand the option that leaves everyone with worse survival chances, we should ask: Are we honouring the spirit of those rules, or are we just performing moral theatre?