Can You Use a Person Without Treating Them Like a Tool?
What If You Used Your Friend?
Imagine you are building a model rocket for a big contest. Your neighbor is great with glue and wire, so you ask her to help you assemble the fins. You never mention the contest — you just want her skill to boost your chances. You get what you want, but your neighbor never knows your real goal. Have you done something wrong?
The philosopher Immanuel Kant (1724–1804) would say that the answer depends on whether you treated your neighbor merely as a means. To use someone as a means is to treat them as a tool to get something you want. That is not automatically bad — we use people all the time, asking a bus driver for a ride or a teacher for an explanation. The problem, Kant thought, is using someone merely as a means: treating them as if they were nothing more than an instrument for your own purposes.
Kant put his idea into one powerful sentence, which he called the Formula of Humanity (or Humanity Formulation) of the categorical imperative, his supreme moral principle:
So act that you treat humanity, whether in your own person or in the person of any other, always at the same time as an end, never merely as a means.
Here “humanity” means the rational nature in each person — her ability to set and pursue her own goals. To treat someone as an end in herself is to recognize that she has a life and purposes of her own, just like you do. Kant believed that whenever you treat a person merely as a means, you act wrongly. But he admitted that figuring out exactly what “merely” means is tricky, and philosophers have been arguing about it ever since.
Two Ways You Might Just Use Someone
What separates an ordinary use of a person from using her merely as a means? Kant gave a famous example. Suppose you need money and, knowing you will never pay it back, you promise to repay a loan. You use the lender to get cash. The lender cannot share your end — the goal you are really pursuing. If he shared your real end (getting money without repayment), then he would not be making a loan at all; he would simply be giving you a gift. Some philosophers put it this way: it is logically impossible for the lender to be making a loan and at the same time to adopt your true goal of never returning the money. Because he cannot share your end, you are treating him merely as a means.
This end‑sharing idea suggests a test: if the person you are using cannot possibly aim for the same thing you are aiming for, then you are just using her. The test works well for lies like the false promise. But does it catch all the cases it should?
Consider a mugger who points a gun at a passerby. He uses her to get money. Could the passerby share his end? In theory, yes — she might secretly want to give him a hundred dollars, so it is not logically impossible for her to share his goal. Yet nearly everyone would say the mugger is treating her merely as a means. So the end‑sharing test, taken strictly, seems to miss this clear case.
Another way to think about “merely” looks not at shared ends but at consent. The philosopher Onora O’Neill suggested that you treat someone merely as a means if she cannot consent to your use of her. If you deceive someone, she does not know what you are really doing, so she cannot genuinely say yes or no — her dissent is “in principle ruled out.” If you coerce someone with a threat, again, she has no real chance to refuse. The mugger leaves the passerby no opportunity to avert being used, so he is using her merely as a means.
Yet the consent account also stumbles. If you save someone’s life by slipping a lie to a dangerous person, and you use an innocent bystander to pass on the lie without telling her the truth, you have made it impossible for her to consent. Is that really treating her merely as a means? Many people think it is not. The bystander might even share your ultimate goal — saving a life — even if she cannot consent to being used as a messenger. The debate shows that no single, simple test has won agreement.
When Using Is Not “Merely” Using
If the hard part is saying when you are treating someone merely as a means, it is equally hard to say when you are not. One natural thought is that you avoid being a “mere tool” user if the other person gives her voluntary, informed consent. On this view, if she understands what she is being used for and freely says yes, then you are not treating her merely as a means.
But some disturbing cases make philosophers worry that consent is not always enough. Imagine a poor mother who cannot afford a good school for her children. A rich person offers to pay for her children’s education — but only if she agrees to be his personal slave, serving him for life. The mother understands the arrangement and, desperate, she consents. The rich person uses her with her full, unforced agreement. Yet many feel strongly that he is still treating her merely as a means, because the choice is so crushingly unfair.
What if, instead of actual consent, we said that a use is not “merely” using someone if the person could rationally consent — if she has sufficient reasons to say yes, even if she never does? That idea runs into famous puzzles. Suppose a surgeon could save five dying patients by secretly killing one healthy person and using his organs. The healthy person would probably never consent, but maybe he has an impartial reason to agree (five lives saved). According to the rational‑consent idea, the surgeon might not be treating him merely as a means. That conclusion strikes many as horrifying. So the search for a clear rule continues.
Is It Always Wrong?
Kant held that treating someone merely as a means is always, without exception, morally wrong. But many people have trouble with that after looking at cases of self‑defense. Imagine someone is attacked by a mugger. To escape, the victim shoves the mugger hard into a wall. The victim uses the mugger as a means to get away, and the mugger certainly could not consent. Yet few would say the victim acted wrongly.
What about a situation where using one person without consent could prevent a nuclear explosion that would kill a million? You might think that using that person — say, by trapping him in a bunker to stop a chain reaction — would still be a bad thing, but the benefits are so enormous that the act might be justified all‑things‑considered. Some philosophers say that the rule against treating people merely as a means is a strong but not absolute moral reason. Others insist, with Kant, that the constraint is unbreakable.
This is live philosophy, not a settled argument. The line between what is morally out of bounds and what is harsh but justified remains blurry, and thoughtful people come down on different sides.
Back to That Model Rocket
Kant’s question follows you through ordinary life. When you ask a friend to distract a teacher while you finish a test, you are using him. Could he share your goal? Not if he would be ashamed to admit it. Does he have a real chance to say no? Maybe not if he feels pressured. Are you treating him as an end — as someone with his own goals and reputation that matter as much as yours? Or are you treating him like a convenient tool?
Philosophers still cannot agree on a flawless definition of “treating someone merely as a means.” But the question itself matters. Every time you involve another person in your plans, you face a choice: to see them as a separate, whole person, or to look past their humanity and see only what they can do for you. Wrestling with that choice is what Kant was really asking of us.
Think about it
- If you ask a friend to help you with a scheme that makes them uncomfortable, but they agree because they want you to like them, are you treating them merely as a means? Why or why not?
- Could a government ever be justified in using someone without their consent — say, a pilot — to prevent a disaster that would kill thousands? If you think it could be, where would you draw the line?
- Think of a time you used a person. Did you pause to wonder whether they could share your goal? Would it have changed your actions if you had?
