Skip to content

Impersonality and Non-identity: A Response to Bramble

by Roger Crisp

Consider the following case, from David Boonin:

Wilma. Wilma has decided to have a baby. She goes to her doctor for a checkup and the doctor tells her that…as things now stand, if she conceives, her child will have a disability. . . that clearly has a substantially negative impact on a person’s quality of life. . . [but is not] so serious as to render the child’s life worse than no life at all. . . .[But] Wilma can prevent this from happening. If she takes a tiny pill once a day for two months before conceiving, her child will be perfectly healthy. The pill is easy to take, has no side effects, and will be paid for by her health insurance. . . .Wilma decides that having to take a pill once a day for two months before conceiving is a bit too inconvenient and so chooses to throw the pills away and conceive at once. As a result of this choice, her child is born [with the disability].

In a recent, ingenious article, Ben Bramble cites Boonin’s case as a ‘beautifully succinct and evocative’ example of what Derek Parfit famously called the non-identity problem.

The problem arises for those many people who think not only that Wilma’s action is wrong, but that any wrong action must wrong someone (that is, that ethics is ‘person-affecting’ rather than impersonal). In Boonin’s case, it’s not at all clear who has been wronged. After all, Wilma’s child seems to have nothing to complain about, given that her life is better than nothing. This is why – as Bramble notes – it is hard to accept alleged solutions to the non-identity problem according to which Wilma’s child has been harmed, or wronged in that, say, her rights have been violated.

Imagine that Wilma herself understands the non-identity problem, and thinks that ethics is person-affecting. As Bramble points out, many of us will still find her decision somewhat unsettling, and this, he plausibly suggests, is that she strikes us as cold and impersonal. We might expect her to put herself into the shoes of someone with a non-trivial disability, and choose to take the pill.

But why should she, if ethics is person-affecting? Because, Bramble suggests, her coldness will be entrenched by her acting on it and may well, through emotional contagion, makes others colder, and hence make it more likely that she and these others will act in ways that will affect particular people for the worse. And, Bramble thinks, if there are cases relevantly analogous to Wilma’s in which these broadly consequentialist considerations are absent, the action in question isn’t wrong.

I accept that Bramble has identified a property of Wilma’s action that makes it, in one respect, wrong. I suspect, however, that many of those puzzled by the non-identity problem will remain unsatisfied with Bramble’s explanation of why they are inclined to think Wilma’s action wrong. The doctor might indeed say: ‘Aren’t you being a bit cold?’ But they might also say: ‘Look, you can have a child with a much better life, and you should’.

Consider the following case, to which Bramble’s explanation doesn’t apply. You are the last person on the planet, and about to die. At no cost, you can bring about any of three outcomes: (a) No one lives after you; (b) One person lives an overall very happy life after you; (c) 1016 different people live overall very happy lives after you, over many generations. Here many will think that you have a strong moral reason to choose (c), even if choosing either (a) or (b) wrongs no one.

Is there any reason we might hesitate to accept that benevolence can be impersonal, and that one can act wrongly without wronging anyone in particular? Bramble considers an option like this, suggesting (506n8) that the most serious objection to such a principle is that it runs into Parfit’s ‘Repugnant Conclusion’. In this case, the objection would be that, if you think beneficence requires (c), then you have to accept that, if there were a further option (d), containing an even larger number of people living lives that are barely worth living, you would be required to choose (d) rather than (c).

But this follows only for act utilitarians. The impersonal beneficence principle may be one principle among others, and could be stated as what Ross called a ‘prima facie duty’, by adding to it an ‘other things equal’ clause. Other things equal, it could be claimed, you should produce the most happiness overall; but, perhaps, in the Repugnant Conclusion case, there are other principles – concerning the guaranteeing of certain minimum standards of welfare, say, or the value of certain activities, such as the creation of great art – and these principles outweigh the impersonal beneficence principle.

Note also that impersonal beneficence, understood as one duty among others, is quite consistent with person-affecting principles of beneficence concerning those who do or will exist. In other words, impersonal beneficence is perfectly reasonable, and indeed – as cases like that of Wilma seem to show – hard to deny.


(I am grateful to Ben Bramble, Brian Earp, and John Skorupski for advice and discussion.)

Share on

2 Comment on this post

    1. Thanks, Richard. Excellent blog, and I’m sorry to have missed it — I should have credited with the Repugnant Conclusion point! In general, it seems difficult for people to see that accepting impersonal goodness and badness leaves everything else where it was (person-affecting goodness and badness, compassion, love, partiality, etc). In the C18, people like Butler understood that common-sense morality includes the utilitarian principle, and that what is special about utilitarians is that they throw everything else out. This was forgotten in the C20, and led to a lot of confusion in philosophy. It’s also had the effect of making many people think that you have to be a utilitarian to take effective altruism seriously.

Comments are closed.