Selfishness Is Irrational

Why you should care equally about everyone

Nearly everyone cares significantly more about themselves than others. I think this is irrational. Of course, in practice it is almost impossible to care for others as we do ourselves, just as it is impossible to be rational across the board. But I think if we reasoned correctly—if we saw the world with perfect clarity—we would not be selfish. We’d care for others as we do for ourselves.

Here’s the argument in a nutshell: either God exists or God doesn’t. If God exists, then selfishness is irrational because it is bad for you. If God doesn’t exist, then probably reductionism about person-identity is true. The boundaries between people are nebulous and a matter of degree, and caring much more about yourself is arbitrary. Thus, in either case, you ought to care about others equal to yourself. I’ll also provide another argument against selfishness that doesn’t depend on either theism or atheism.

1 God

Suppose that God exists. Then it seems obviously irrational to mostly act in your own self interest. After all, if God exists, there is probably an afterlife (it would be a very capricious thing to snuff us out after a few decades, when we could live on in nice conditions forever).

If there is an afterlife, it is reasonably likely that we will be rewarded for moral behavior. This could be direct, with God simply giving us welfare proportional to how well we behaved. Or it could be indirect: moral behavior strengthens our connection with others. These connections last forever; for this reason they are of infinite value.

Now, maybe you’re a bit skeptical of these accounts. But the potential reward from them is infinite. So even on self-interested grounds, if God exists, you have strong moral reason to behave well. Thus, theism only needs to be minimally plausible for it to be irrational to be selfish.

2 Atheism

(Note: this entire argument comes from Parfit’s excellent book Reasons and Persons.)

What if there is no God?

If there is no God, it is fairly unlikely that there is an immaterial soul. Souls are complicated things. It is hard to see how, in a Godless world, there would be a precise correspondence between souls and bodies without highly complicated laws. This is one reason atheists generally deny the existence of souls.

But if there aren’t immaterial souls, then what could fix facts about personal identity? While you bear certain similarities to your four-year-old self, there is no highly robust sense in which you are the same person. Just as there may be no fact of the matter about whether a ship remains the same after changing out its planks, there may be no fact of the matter about whether a person remains the same after undergoing a bunch of changes.

To see this more clearly, imagine that there are a series of switches. Each switch will make some change to your brain and body. If they are all pressed, you will turn into a lady named Barbara. If just some of them are pressed, then you will become more like Barbara, proportional to the number of buttons pressed. At what point do you stop being the same person?

It seems that there cannot be such a point absent a soul. There are simply degrees of connectedness that dwindle as more levers are pressed.

Here’s one way to see this: identity is supposed to be transitive. If your four-year-old self is the same as your eight-year-old self, and your eight-year-old self is the same as you, then you are the same as your four-year-old self. Clearly if identity were objective, then every person would be the same as the person produced by the adjacent lever. But then this implies that after all the buttons were pressed and you became Barbara, you’d be the same person.

Or, for another case, imagine that your left brain goes in one skull and your right in another. Which one are you? Surely you’re not both (you can’t be two people at once). Are you just one? If so, which? Now, you might think you don’t survive the process, but we normally think people survive the destruction of one of their hemispheres.

I think this illustrates that identity facts, given atheism, are probably vague and frequently indeterminate. There’s no really robust sense in which you will be the same person in one year. But then why care about your future self? It can’t be because they’re you in some deep sense! There is no deep sense in which they’re you!

On this picture, caring more about your future self is like caring more about some stranger just because they resemble you more. It just seems clearly irrational. It involves caring about some nebulous, vague, and arbitrary property that isn’t the source of any genuine reasons. Without a non-physical soul to fix identity, there are just degrees of similarity.

Consider the case with the buttons. Do you have much more reason to care about the person who is the result of one button being pressed than two? If over time, you’d go from being yourself to being Barbara, would your interests matter less and less over time? Or would there be some discrete cutoff? Should we then care less about our future self than our present self because they bear fewer similarities to us? Surely not.

You might worry that this proves too much. If there’s no personal identity, how do we have any special obligations? How do we maintain that we have stronger duties to our friends and family than others? In response, I’d say:

1. You can either give up reductionism or belief in robust special obligations. As it happens, I reject them both!

2. Special obligations can generally be grounded in facts about the person as they are. For example, it is true of some person at a time that they are your friend. You need not think that there’s some highly robust sense in which they’re the same as their past self. Similarly, in many cases, like the case of obligations to parents or children, it doesn’t seem so bad to anchor obligations in the person being psychologically continuous with the person who raised you. If your parents were replaced with clones, you’d still have special obligations to them. In contrast, if you were replaced with a clone who was not you, you wouldn’t have self-interested reasons to care about them.

So then if atheism is true, personal identity is sufficiently unreal not to be the source of reasons. But then being selfish is irrational. It involves assigning significance to something that is not real in any crucial sense and drawing obviously arbitrary distinctions.

3 Seasons and persons

Parfit had another related argument in Reasons and Persons for the irrationality of being selfish. This one dovetails nicely with the last one, but it can be made on its own, and can work even given theism.

Imagine a person who only cared about their present self? Or perhaps one who cared about their welfare at all times except on future Tuesdays. Such a person would be irrational! That some period of pain or joy falls on a future Tuesday is no reason to care less about it. Similarly, we behave irrationally when we procrastinate and neglect our future welfare.

This illustrates that rationality isn’t just about getting whatever it is we want. The person who only wants current welfare behaves irrationally in neglecting their future welfare. But if it’s irrational not to care about your future self, why isn’t it irrational not to care about other people?

Here are some things that might be said about why it isn’t irrational to care about other people’s welfare.

“They’re not me—what happens to them doesn’t affect me.” But similarly, the person who cares only about his present self could say “my future self isn’t present me—what happens to them doesn’t affect present me.”

“I just don’t care about their welfare as much as my own.” But similarly, the person who cares only about his present self could describe that he doesn’t care about his future welfare.

“You’ll later regret ignoring your future welfare.” But the person who just cares about his present self could object that his present self will never regret it. In a parallel way, other people will regret you not taking seriously their interests.

In each case, the explanation of why one is permitted to care only about one’s own welfare can be mirrored by the view that one is permitted to care only about one’s own present welfare. So long as we recognize that caring only about your present self is irrational, by symmetry, we should recognize that not caring about other people’s welfare is irrational.

This claim is surprising but not extremely so. It seems that one has a reason to pull a child out of a pond even if one doesn’t want to. But if you have a reason to do something, then rationality would incline you to do it. So, it seems, rationality would incline people in the direction of caring about others. Moral realism goes naturally with the claim that there are stance independent reasons to care about others—but if there are, then rationality inclines you to care for others, not just for yourself.

I also think this position becomes more intuitive if you think that our reasons for acting don’t come from our desires. Even if one had a desire to cause themselves future agony, they wouldn’t have a reason to cause themselves future agony. But if this is right, then the fact that a person only cares about their own welfare tells us little about what they have reason to care about.

The alternative view is also subject to evolutionary debunking. Care for oneself is adaptive. Thus, we should expect to privilege our own welfare even if doing so is irrational.

Lastly, I find the position more intuitive when I reflect. When, for example, I think about loved ones and imagine things from their perspective, it seems like the rational thing to do is care about their interests as my own. If we saw more clearly about morality and rationality, then we’d see everyone this way.

4 Conclusion

Here I’ve advanced two main arguments for the irrationality of selfishness. The first takes the form of a dilemma: if theism is true, then God exists and selfishness is bad for you. If it’s false, then reductionism about personal identity is true, and caring about yourself is rather like caring about whether a ship remains, in some semantic sense, the same ship.

The second argument is an argument by analogy: it is hard to see why we have reason to care about our future self, even if we don’t care about it, but not reason to care about others. I have similarly argued that this position’s counterintuitiveness doesn’t tell us much against it because it is plausibly explained by evolution.

Maybe you’re uncertain about these arguments. In such a case, you’ll have some reason to behave selflessly, given that selfish behavior might be irrational. You have some reason to refrain from doing things that have reasonable probabilities of being in error. Uncertainty about these arguments might permit you to care somewhat more about yourself than others, but it means you ought to care significantly about others.

There is also self-interested reason to care about others: those who do tend to be happier. Doing good is good for us, both instrumentally and possibly intrinsically. Those who give more to charity are happier, even controlling for other factors. Thus, there are ample grounds for moral behavior. For this reason, and others, I’d suggest that at the very least you should take the morally low-hanging fruit and give some non-trivial slice of your money to highly effective charities that save lives cheaply.