If humanity survives for a long time, then people today could have an extremely large number of descendants. Imagine that there are 100 drowning people, all of which happen to be infertile and one drowning young woman who is not infertile.

It is plausible that the woman could have a million descendants, where as the 100 people will obviously not have any descendants. Assuming that those people will live normal lives, clearly her life is far more valuable than the lives of the 100 people. This seems like a problem for utilitarianism.

Even ignoring the far future, it seems like there is an incredibly strong obligation under utilitarianism to have children. Having a child will presumable lead to about 80 normal years of life, a very good consequence. Murdering some random 40-year-old adult will only lead to a loss of about 40 life-years (plus maybe it might lead to some instrumental loss of utility, but this shouldn't be enough to get rid of the problem). Do utilitarians really think that having a child is a better action than murdering someone is a bad action, so it would be OK to murder a random person in order to have a child?

I think this is slightly different than the repugnant conclusion, though it is in the same vein.

  • Taking indirect consequences into ethical consideration is idle speculation. Always. In every framework.
    – Philip Klöcking
    May 22, 2016 at 22:20

2 Answers 2


The problems you see are typical issues that one arrives at when using too simple of a utility function. For example, it is well recognized that life-years is a terrible metric for utilitarians, with all sorts of known pitfalls (such as quality of life arguments). A utilitarian would argue that ones metric must be more developed than that.

The problem of future people is actually a subset of the real challenge utilitarians face: the problem of the unknown. At the moment a utilitarian makes a judgement call, such as to save one fertile woman, he or she does not know what will happen in the future. For all he or she knows, this woman will become the mother of a very evil dictator who will bring the world into war.

This problem is tackled in many ways, but the one which is most accessible is probability. Potential outcomes are assigned a probability, or at least a likelihood, and then are weighed accordingly. Different utilitarians may assign value in different ways. One utilitarian may value the fertile woman's ability to bear potential children more. One might value the lives of the other 100, because they know that all of these people can be saved, but the woman's bearing of children is just a probability. Others might even be more nuanced, and start comparing the ability to produce offspring against the sum wisdom of the 100, and their ability to enlighten other offspring (presuming this 1 woman is not the only fertile woman in the world).

  • Ultimately, it seems like the expected utility from saving the one woman is probably far higher than from saving the 100 people. While it is true that saving this woman might have very bad results, the "normal" result that she has a lot of descendants has reasonably high probability. Even if it is unknowable what will actually happen, one be reasonably confident that (whatever the metric) the significant change of one million more people existing throughout the future will outweigh anything the other hundred do.
    – vukov
    Feb 21, 2016 at 0:39
  • 1
    @vukov Then you now know something about what you perceive utility functions to look like: "change is good." That particular utility metric weighs heavily for the volume of change, and not the direction that change goes in. Might that utility function get slighted if the woman is a devout believer in a religion whose teachings you do not appreciate?
    – Cort Ammon
    Feb 21, 2016 at 1:06
  • I don't understand what you mean by change.
    – vukov
    Feb 21, 2016 at 1:26
  • @vukov "... the significant change of one million more people existing..."
    – Cort Ammon
    Feb 21, 2016 at 2:03
  • 1
    @vukov One can say option A is a good thing, but yet choose option B because it is a better thing. Utility functions have to be more nuanced. For example, those one-million ofspring consume resources, so there is an opportunity cost to them. If you do not save the woman, those resources may be instead given to one-million offspring of other individuals that are not part of the drowning group. The utility of the wisdom of the 100 transfered to the million, in this case, could potentially outweigh the mere reproductive abilities of the woman (theoretically many women have such abilities)
    – Cort Ammon
    Feb 21, 2016 at 2:50

I would argue the opposite and save the 100 people, even in the eyes of the utilitarian these people are worth more than one fertile women. These people will increase the live span of many fertile people. What if these people were doctors or police officers. These men and women would then be able to save life even though they can't produce any. This is furthered bu the fact that even though we know the 100 infertile people cant produce children, we have no guarantee that the fertile women will either. We also have to bring into account the fact that utilitarianism, by detention, is maximizing happiness throughout the population. The people would be much more upset if we let 100 people die instead of one even though she might be infertile

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .