We are in favor of making people happy, not making happy people, said Jan Narveson. The slogan captures the core idea of person-affecting ethics. On these views, ethics is fundamentally about what is good or bad for people, rather than about making the world impersonally better.

This view might look attractive. Suppose we can either cure malaria (preventing enormous suffering among people alive today) or bring into existence arbitrarily many happy beings. You might think we should do the former. Maybe it would also be good to bring many happy beings into existence, but perhaps this seems far less important. After all, no one is harmed if we don’t bring many happy beings into existence, but many will continue to suffer if we don’t cure malaria.

This line of thought resembles a common criticism of some longtermist views. Whatever value there is in creating additional happy lives, many believe we ought to prioritize the problems facing ourselves and our descendents. We should not, in short, turn our backs on people to benefit the world.

Should We Care About Merely Possible Beings?

One straightforward version of a person-affecting view is the “narrow” one. On this view, an outcome can be better or worse than another only if it is better or worse for someone who exists in both outcomes.

In one outcome, Mike has a good life; in the other, he has a much worse life. We can straightforwardly say that the first outcome is better for Mike. And so, on the narrow person-affecting view, it’s better overall. But now compare an outcome in which Mike has a good life with one in which Mike never exists.

Now it’s harder to say that the first outcome is better on the person-affecting view, because it’s hard to say that the first outcome is better for Mike. There is no Mike in the second outcome for whom anything is better or worse, so the comparison looks strange.

The narrow person-affecting view tells us to direct our moral concern towards those whose existence does not depend on what we do, whether they exist today or in a million years. This enables clean comparisons where the same people exist in both outcomes we are considering. It doesn’t care about “merely possible” people—those whose existence is up to us.

This might sound plausible, but it quickly runs into a problem. Suppose that instead of saying, We are in favor of making people happy, not making happy people, we said, We are against making people miserable, not against making miserable people. Whatever we think of Narveson’s slogan, this one seems far less plausible. Surely we are against bringing miserable beings into the world!

Yet on the narrow person-affecting view, merely possible people (happy or miserable) don’t count. The view therefore struggles to explain why we should not bring miserable beings into existence.

Accommodating the Asymmetry

Narveson agrees that we should not bring miserable beings into existence, even though he denies that we have a corresponding reason to create happy beings. He and others have argued that person-affecting views can accommodate our intuitions about population ethics, including the procreation asymmetry. Johann Frick argues, for instance, that there is no unconditional reason to create someone merely so that they can enjoy a good life. But there is a conditional requirement to ensure that, if we create someone, their life is worth living.

This is a coherent way of preserving the asymmetry, but it may strike us as oddly indirect. The claim that we should not create a miserable being now depends on further ideas about conditional reasons and welfare thresholds, rather than following cleanly from person-affecting ethics. Perhaps an impersonal view would more straightforwardly explain why it’s wrong to bring miserable beings into the world.

Also, talk of a “life worth living” may seem to require those strange comparisons between Mike’s existence and nonexistence. How else could we know whether a life is worth living except by comparing it with not living at all?

And if we allow such comparisons—and allow that bringing someone into existence can be bad for or harmful to them—then it seems we should also allow that bringing someone into existence can be good for or beneficial to them. If so, we need a deeper explanation of the asymmetry than the person-affecting view provides on its own.

The Non-Identity Problem

These considerations should give us some pause about person-affecting views. But the more decisive challenge comes from Parfit’s non-identity problem.

Suppose a pregnant woman can take a harmless drug that would prevent her future child from developing a terrible disease. Clearly, she should. But now imagine the case was like this. A woman who is planning to get pregnant learns that her child will develop a terrible disease unless she waits a short amount of time to conceive. If she waits, she will conceive a different child (since a different sperm and egg will meet), who will not have the disease.

Perhaps you think she should wait. After all, the cost is small, and it prevents great suffering—just like taking the harmless drug. The crucial difference between the cases, though, is which child exists. Unlike the drug case, where the same child can be made better off, the waiting case involves comparing the fates of two different possible children, one with the disease and one without.

You might think she ought to wait. Yet if she chooses not to, person-affecting views have trouble finding issue with her decision. This is because the child who develops the disease cannot say that they would have been better off had she waited: had she waited, that particular child would never have existed. Assuming their life is still worth living, her decision has not made them worse off than they otherwise would have been.

Person-Affecting or Suffering-Focused?

A person-affecting view may initially seem attractive, but perhaps some of that appeal is a mirage. The narrow person-affecting view, perhaps the most straightforward version, struggles to condemn bringing miserable beings into existence. Attempts to accommodate this intuition, we saw, may seem a bit ad hoc. And all person-affecting views stumble over the non-identity problem.

Perhaps our intuitions in these cases are better explained not by whether we benefit or harm particular people, but by a view that treats reducing suffering as especially important. Perhaps what matters to us is not which beings might exist or the modal status of their existence, but rather the kinds of experiences they undergo.

The reason we should eradicate malaria rather than create many happy beings might primarily be that reducing suffering is more important than promoting happiness. Similarly, perhaps the reason we have strong reason not to create miserable beings but no analogous reason to create happy beings is the priority of reducing suffering over promoting happiness. Finally, perhaps the reason the woman should wait to conceive is simply that doing so reduces suffering—even if it benefits no existing being.

Of course, there are other possible explanations for these intuitions. We might think there is a general priority to preventing harms over providing benefits, or appeal to prioritarianism, sufficientarianism, special duties, or other considerations. But suffering-focused views are a serious candidate worth considering. Such views may have some counterintuitive implications, but all moral views do! On close examination, we might find that these implications are more tolerable than those of other views.