The repugnant conclusion is the proposition that for any well-off population A, and for any quality of life Q which exceeds the minimal quality of a life worth creating (no matter how small the margin), there exists a possible population B with quality of life Q such that it is better from a utilitarian standpoint for B to exist than it is for A to exist. I find the repugnant conclusion convincing, for reasons which have been discussed elsewhere and I will not repeat here.
Consider the path from population A to population A+ to population B described in the linked wikipedia article from the perspectives of members of the above populations. To a member of population A who values the creation of lives worth living, the transition from A to A+ by creating lives worth living is an improvement. From the perspective of an average member of population A+, the transition from A+ to B is neutral; you might gain and you might lose, but on average you break even, and from an altruistic standpoint, it’s also break-even. But from the perspective of a member of population A, the composite transition all the way from A to B doesn’t look so great. If you accept the repugnant conclusion, then it’s a win from an altruistic perspective, but you personally lose out. It may or may not still seem like a good idea overall. (Similarly, even if you personally are not affected, people also care about their friends and family more than about arbitrary strangers, and these people that actually-existing people especially care about are disproportionately likely to also actually exist, so this could be another reason not to transition from A to B, but for simplicity, I’ll ignore such considerations for the rest of this post.)
Here we have a variant of the repugnant conclusion: the proposition that for any well-off population A, and for any quality of life Q which exceeds the minimal quality of a life worth creating (no matter how small the margin), there exists a possible population B with quality of life Q such that, if you are a member of A, you should, if given the chance, replace population A with population B and become an arbitrary member of B. Since decisions that affect populations tend to be made by people who already exist at the time, it makes sense to frame population ethics questions from the perspective of members of the population like this.
Let P be a variable measuring your personal well-being, T be the sum of variables measuring the the well-beings of everyone in the population, c be a positive constant, and f be a bounded, monotonically increasing function. Most people are at least somewhat altruistic, but also care about themselves much more than about others, so let’s assume that your utility function increases monotonically with P and T. Consider the possible utility functions U1(P,T)=f(PcT) and U2(P,T)=Pf(T). For scenarios in which the total population is fixed, both of these utility functions can be made arbitrarily selfish or altruistic by adjusting f and c, and for scenarios in which your choices only affect T (P is constant) and all options are deterministic, both utility functions will say to act like total utilitarians. In particular, both of them accept the repugnant conclusion. But while U1 accepts the variant, U2 does not. That is, if you accept the abstract repugnant conclusion, whether or not you would want to implement it in practice depends on not only how selfish or altruistic you are, but also the manner in which you are selfish and altruistic. I don’t like the idea of caring about myself arbitrarily little as the population grows to infinity, so it seems intuitive to me that my utility function would be more like U2 than like U1.
Selfish reasons to reject the repugnant conclusion in practice
Prerequisite: http://en.wikipedia.org/wiki/Mere_addition_paradox
The repugnant conclusion is the proposition that for any well-off population A, and for any quality of life Q which exceeds the minimal quality of a life worth creating (no matter how small the margin), there exists a possible population B with quality of life Q such that it is better from a utilitarian standpoint for B to exist than it is for A to exist. I find the repugnant conclusion convincing, for reasons which have been discussed elsewhere and I will not repeat here.
Consider the path from population A to population A+ to population B described in the linked wikipedia article from the perspectives of members of the above populations. To a member of population A who values the creation of lives worth living, the transition from A to A+ by creating lives worth living is an improvement. From the perspective of an average member of population A+, the transition from A+ to B is neutral; you might gain and you might lose, but on average you break even, and from an altruistic standpoint, it’s also break-even. But from the perspective of a member of population A, the composite transition all the way from A to B doesn’t look so great. If you accept the repugnant conclusion, then it’s a win from an altruistic perspective, but you personally lose out. It may or may not still seem like a good idea overall. (Similarly, even if you personally are not affected, people also care about their friends and family more than about arbitrary strangers, and these people that actually-existing people especially care about are disproportionately likely to also actually exist, so this could be another reason not to transition from A to B, but for simplicity, I’ll ignore such considerations for the rest of this post.)
Here we have a variant of the repugnant conclusion: the proposition that for any well-off population A, and for any quality of life Q which exceeds the minimal quality of a life worth creating (no matter how small the margin), there exists a possible population B with quality of life Q such that, if you are a member of A, you should, if given the chance, replace population A with population B and become an arbitrary member of B. Since decisions that affect populations tend to be made by people who already exist at the time, it makes sense to frame population ethics questions from the perspective of members of the population like this.
Let P be a variable measuring your personal well-being, T be the sum of variables measuring the the well-beings of everyone in the population, c be a positive constant, and f be a bounded, monotonically increasing function. Most people are at least somewhat altruistic, but also care about themselves much more than about others, so let’s assume that your utility function increases monotonically with P and T. Consider the possible utility functions U1(P,T)=f(PcT) and U2(P,T)=Pf(T). For scenarios in which the total population is fixed, both of these utility functions can be made arbitrarily selfish or altruistic by adjusting f and c, and for scenarios in which your choices only affect T (P is constant) and all options are deterministic, both utility functions will say to act like total utilitarians. In particular, both of them accept the repugnant conclusion. But while U1 accepts the variant, U2 does not. That is, if you accept the abstract repugnant conclusion, whether or not you would want to implement it in practice depends on not only how selfish or altruistic you are, but also the manner in which you are selfish and altruistic. I don’t like the idea of caring about myself arbitrarily little as the population grows to infinity, so it seems intuitive to me that my utility function would be more like U2 than like U1.