Appendices - Killing Happy Animals: Explorations in Utilitarian Ethics - page 157

possible outcomes, they are by definition not contingent people. They
are necessary people. If it were two different populations, x and y, then
the Narrow Person-Affecting Restriction would be indifferent between
those outcomes. However, the Wide Person-Affecting Restriction in
conjunction with utilitarianism would prefer outcome B, because the
y people are better off in B than the x-people are in A. Then, outcome
B would be better than outcome A. So, Arrhenius has not shown that
the Person-Affecting Restriction can be incompatible with the Prior
Existence View, and thus he has not shown that the Person-Affecting
Restriction does not imply it.
Interestingly, Arrhenius acknowledges that the Person-Affecting
Restriction, in conjunction with the assumption that existence cannot
harm or benefit a being, does imply a view called Strict Comparativism.
This is another view about the moral status of potential beings. According
to Strict Comparativism:
We should completely disregard the welfare of uniquely realizable
people, that is, people that only exist in one out of two compared
outcomes. We should only count the welfare of non-uniquely real-
izable people, that is, people who exist in both of the compared
outcomes.
Strict Comparativism is not the same as the Prior Existence View.
Arrhenius correctly points out that there are differences between both
views. 12 I wonder whether those differences are such that the Person-
Affecting Restriction in conjunction with the assumption that exist-
ence cannot harm or benefit a being does imply Strict Comparativism
but not the Prior Existence View. In many cases the contingent people
are also the uniquely realisable people and the necessary people are the
non-uniquely realisable people. However, there might be differences.
Arrhenius explains:
A person that is uniquely realizable relative to all pairs of outcomes
in the situation of which she is part is also a contingent person, but
a contingent person is not necessarily uniquely realizable in respect
to all pairs of outcomes in a choice situation since she can exist, for
instance, in two out of three outcomes.
Thus, a contingent person might exist in two or more of the possible
outcomes, while a uniquely realisable person exists only in one. I do
not see how this should make a difference for the issue at hand, i.e.
whether the Person-Affecting Restriction in combination with the