Let’s settle the Trolley Problem debate

There is a philosophical distinction so simple it sounds almost boring when you first hear it. A positive duty tells you that you ought to do something. A negative duty tells you that you ought not to do something. That is it. That is the whole idea. And yet, sitting with that distinction long enough quietly dismantled positions I had held for years: on trolley problems, on ethics, on what I eat for dinner. Not because it gave me easy answers, but because it forced me to ask a harder question. Is there a difference between the harm I cause and the harm I fail to prevent? I think there is. And I think it matters enormously.

The Lever You Probably Think You Should Pull

If you have spent more than ten minutes near a philosophy class, a Reddit thread, or a late-night conversation with someone who just discovered Peter Singer, you have heard the trolley problem. A runaway train is heading toward five people trapped on a track. You are standing next to a lever. Pull it, and the train diverts onto another track where it will kill one person instead of five. Do you pull the lever? Most people say yes. For a long time, I said yes too. The math seems obvious. Five deaths versus one death. If morality is about minimizing suffering, you pull the lever. Hesitating feels almost like vanity, like you are protecting your own clean conscience at the expense of four human lives. But there is another way of describing exactly the same situation. If I pull the lever, I cause the train to hit the person on the other track. If I do nothing, I do not save the five people. Read those two sentences again. They describe the same physical scenario, but they are not morally identical, and the reason they are not is precisely what the positive and negative duty distinction is trying to capture.

Two Rules That Do Not Have Equal Weight

Consider two moral rules. You shall not kill. You shall save lives. Both principles exist in our moral and legal systems, but they do not have the same status. Think about what it would mean if the positive duty to save lives were truly as strong and binding as the negative duty not to kill. We would have an enormous obligation to intervene whenever someone was in danger. We would all have to become firefighters, doctors, police officers, or simply people constantly intervening in the lives of strangers. But that is not how we normally understand morality, and almost nobody actually lives as though these two duties are equal. Even people who intellectually endorse strong utilitarian principles, when pushed, admit the obligations feel different in kind and not just in degree. This is not hypocrisy. It is a genuine moral intuition pointing at something real. We have always understood, at some level, that there is a meaningful difference between killing someone and not saving someone. The language we use reflects it. The laws we write reflect it. A surgeon who operates and accidentally causes a death faces a very different legal and moral situation than one who refuses to operate and allows a patient to die, even if the outcome is the same. Now bring that back to the lever. If I pull it, I am not saving five people in some passive, bystander sense. I am redirecting lethal force toward a specific human being. I become an active link in the causal chain that ends their life. That person was not in danger before I intervened. If I do nothing, five people die, and that is terrible. But I have not killed anyone. I have failed to save them, which is a morally different thing. Under a framework that takes negative duties seriously, that difference is morally decisive. I would not pull the lever.

Why Consequentialism Gets Uncomfortable Very Quickly

At this point, a committed utilitarian is getting impatient. Five is more than one. Pull the lever and stop overthinking it. But this is exactly where I started becoming uncomfortable with purely consequentialist reasoning. Imagine the one person on the other track has twenty children who depend on them entirely. The five people on the main track are elderly with no surviving family. Does that change your answer? Now imagine one of the five is a surgeon who will go on to save hundreds of lives. The one person is terminally ill and has weeks to live. Different answer again? What if the five are strangers and the one is your child? The point is not to be cynical about consequentialism. The point is that the apparently simple calculation of five versus one is never actually that simple. The moment you commit to whatever produces the best outcome, you are committed to knowing the outcomes, all of them, cascading into the future, weighted against each other somehow. And you do not know them. You cannot. Real moral decisions almost never come with complete information. The clean arithmetic of the trolley problem only works because the problem artificially strips away everything that makes real life complicated. Negative duties do not have this problem. Do not kill does not require you to model the future. It does not depend on who has more dependents or whose life has more downstream value. It is a constraint on your actions, not a calculation about outcomes, and that is precisely its strength.

I Never Thought I Would Write This

I should be honest here. I used to be a fairly hardcore utilitarian. I found the idea of maximizing overall well-being intuitively compelling. If we can prevent five deaths at the cost of one, why would we not? But there comes a point in the life of anyone interested in philosophy when principles stop being abstract ideas and start becoming rules you have to apply to actual situations. And when I did that honestly, the implications of pure consequentialism led to conclusions that felt deeply wrong. Not just uncomfortable, but wrong in a way that made me question the framework rather than override the intuition. That does not mean I suddenly became a Kantian. I still find consequentialist reasoning useful. There are situations in which consequences clearly matter and I do not think morality can simply ignore them. But I no longer think that every moral problem can be reduced to a calculation of outcomes. Sometimes the difference between doing something and allowing something to happen matters. Sometimes refusing to cause harm is morally different from failing to prevent harm. And sometimes a rule like do not kill deserves to constrain what we are willing to do, even when breaking it might appear to produce a better numerical outcome.

And Then There Is Meat

This is also where my thinking about vegetarianism changed. I used to defend eating meat largely in consequentialist terms. I could ask whether eating meat actually produced more or less suffering, whether humans need animal products, whether animals have sufficiently complex experiences, or whether individual consumption makes any meaningful difference. But eventually I started looking at the question through the lens of negative duty. I do not necessarily have a positive duty to save every animal. But perhaps I have a negative duty not to kill an animal unnecessarily. That is a much simpler claim. I do not have to believe that eating meat is the greatest source of suffering in the world. I do not have to believe that becoming vegetarian will solve industrial animal agriculture. I do not even have to claim that humans have exactly the same moral status as animals. I only have to accept a more limited principle: if I can reasonably avoid killing an animal for food, and I do not need to do so to survive, why should I? For me, that question was more persuasive than any utilitarian calculation I had ever encountered. Not because it was louder, but because it was harder to argue away.

What the Trolley Problem Is Actually Asking

The trolley problem has become something of a cultural punchline, the philosophy class thought experiment that everyone has an opinion on and nobody agrees about. But it persists for a reason. It forces our moral principles out of the abstract and into the concrete. It makes you decide, under artificial but clarifying conditions, what you actually believe about the structure of moral obligation. And what I now believe is this: morality is not only about engineering the best possible world from the available inputs. It is also, and perhaps more fundamentally, about the kind of agent you are willing to be in the process of trying. It is easy to say that we should maximize well-being. It is harder to decide what counts as maximizing well-being when we do not know all the consequences. It is easy to say that saving five people is better than saving one. It is harder when saving the five requires you to actively kill the one. There are things that constrain what we may do in pursuit of good outcomes. Not because consequences do not matter, but because some actions are not merely suboptimal. They are violations. They cross a line that is not just about adding up costs and benefits, but about what we are prepared to do to another person, deliberately, with our own hands. Do not kill is one of those lines. It does not solve every problem, and it does not tell you what the right thing to do always is. But it tells you something real about where the moral floor is, the place below which you should not go, regardless of the math. Maybe that is where ethics has to begin. Not with the grandest calculation of overall welfare, but with the simplest and most serious question we can ask ourselves. Is this something I should be doing at all?