Ethical Egoism: Do You Have a Duty to Be Selfish?

When an airplane loses cabin pressure, you're supposed to put your own oxygen mask on first. Before helping your child. Before helping the elderly person next to you. Yourself first. The reasoning is sound: if you pass out trying to help someone else, you've helped no one.

What if all of life is like that? What if we are supposed to think of ourselves first, not as a concession to human weakness, but as a genuine moral obligation?

That's the claim of ethical egoism: the view that each person has a moral duty to act in her own self-interest. Not that people do act selfishly (that's psychological egoism, which we explored in an earlier post), but that they should. Selfishness isn't a vice to overcome. It's a moral requirement.

The Plank of Carneades

To see why ethical egoism is worth taking seriously, consider an ancient thought experiment known as the Plank of Carneades. Two people are shipwrecked. There's one plank of wood floating in the water, and it can only support one of them. They're the same age, the same gender, equally healthy. Neither has a family depending on them. Neither is about to cure cancer. There's no tiebreaker. One of them will survive; the other will drown.

Should you fight for the plank? Most people's gut says yes. And if you think about it, it's hard to articulate why you shouldn't. What moral principle would require you to drown so that a stranger can live, when neither of you has a stronger claim to survival than the other? The usual reasons for self-sacrifice don't apply here. There's no greater good being served. There's no asymmetry of need. It's just you and another person, and one plank.

Maybe, the ethical egoist suggests, this scenario reveals something we're reluctant to admit: you are morally permitted, perhaps even morally obligated, to save yourself.

The Practicality Argument

The strongest argument for ethical egoism is about practicality. A moral theory needs to be one that people can actually follow. And you can't realistically expect people to follow a theory that systematically requires them to sacrifice their own interests.

Here's a thought experiment that makes this vivid. Imagine self-driving cars become legally required. A car manufacturer is programming the decision-making algorithm. You're driving on a one-way bridge over a canyon that's 100 miles deep. There's a sidewalk for pedestrians. A hole suddenly opens in your lane. On the sidewalk is a pedestrian who is your exact age, gender, and circumstances. The car has two options: swerve onto the sidewalk and kill the pedestrian, or drive into the hole and kill you.

If the manufacturer programs the car to drive into the hole, they're asking you to buy and drive a vehicle that you know will choose to kill you in order to save a stranger. Can we really expect anyone to agree to that? Would you buy that car?

The ethical egoist says this reveals something important: a moral theory that demands you accept your own death to save a stranger of equal moral standing isn't just idealistic. It's unlivable. No one would consent to it in practice. A moral theory has to be one that rational agents can actually adopt, and ethical egoism is the only theory that never asks you to act against your most fundamental interest.

The objection to this argument is that moral theories don't have to describe what people will do. They describe what people should do. We might appeal to our ideal selves rather than our actual selves. Can we really expect a heroin addict not to steal? No. But stealing is still wrong. The fact that people can't live up to a moral standard doesn't invalidate the standard.

The Cooperation Problem

The most obvious objection to ethical egoism is that it seems to permit all kinds of behavior we consider clearly immoral. If self-interest is the only moral criterion, then lying, cheating, and stealing are all fine whenever you can get away with them.

The standard egoist reply is that cooperation is usually in your best interest. Dacher Keltner's research on the Power Paradox supports this: people gain influence and social power through generosity, empathy, and cooperation, not through ruthlessness. Being good to others is good for you, most of the time. So ethical egoism, the argument goes, doesn't actually endorse the selfish behavior we're worried about. It endorses cooperation, because cooperation pays.

This is plausible as far as it goes, but it has a critical weakness. It works only as long as you need other people. Once someone reaches a position of power where cooperation is no longer necessary, the calculus changes. At that point, it may genuinely be in their self-interest to lie, cheat, and exploit others. Keltner's own research documents this: the same people who gained power through generosity often become selfish and abusive once they have it. Ethical egoism, applied consistently, endorses that transition. The powerful person who exploits the weak is doing exactly what the theory says she should, as long as she can get away with it.

The cooperation reply only works in conditions of mutual dependence. As a general moral theory, it fails precisely where it matters most: when power is unequal.

The Wrong Reasons Problem

But the deepest objection to ethical egoism isn't about what it permits. It's about why it says things are wrong. Even when ethical egoism gets the right answer, it gets there for the wrong reasons.

Consider a scenario. You are a sadistic person who takes delight in causing suffering. You have captured a four-year-old child who is completely alone in the world. No one will ever find out. You are absolutely certain you could get away with it. Should you torture the child?

Ethical egoism might be able to answer "no" by pointing to guilt, psychological damage, risk of discovery, or some other cost to you. But notice: when we say it is wrong to torture a child, that's not what we mean. We don't mean "don't do it because you might feel guilty." We mean "don't do it because it is wrong, because the child's suffering matters intrinsically, because there is something about the child that demands moral consideration regardless of what it does to you."

Ethical egoism can't capture this. It has no way to say that the child's suffering matters in itself. Every moral consideration has to pass through the filter of your self-interest. And in a case where self-interest genuinely points toward cruelty (a true sadist who won't feel guilty, who won't get caught), the theory has no resources to condemn it. That's not a marginal failure. It's a failure at the core of what morality is supposed to do.

What Ethical Egoism Gets Right

Despite these problems, ethical egoism captures something real. There's a version of selflessness that's genuinely unhealthy: the person who gives and gives until they're completely used up, who neglects their own wellbeing in the name of serving others. Telling that person to think about themselves isn't immoral. It might be the most important advice they could hear.

The airplane oxygen mask analogy isn't wrong. You do need to take care of yourself before you can take care of others. The question is whether that practical wisdom should be elevated into a complete moral theory, one that says self-interest is the only thing that matters. And the answer, I think, is no. Self-care is important. A duty to be selfish is something else entirely.