Appendices - Killing Happy Animals: Explorations in Utilitarian Ethics - page 155

the only options. When A and B are the only options, A is clearly better
than B. So, why should they become equally good when a third option,
C, is added? This, according to Bradley (2013, p. 41), would 'violate a
plausible principle of independence of irrelevant alternatives'.
Bradley's argument, however, is misguided, to the extent that it ought
not to be directed against deontic orderings. If we are concerned with
the relative rightness of options, as opposed to their relative goodness, it
clearly matters what the available alternatives are. Broome (2004, p. 147)
explicitly acknowledges this (see also Roberts 1998), and restricts his
criticism to the axiological domain. Bradley, in contrast, fails to distin-
guish the axiological from the deontic domain in this regard. He speaks
about axiology and reasons for action as if it where the same thing. But
axiology is not about providing reasons. It is about goodness. We cannot
jump immediately to conclusions about obligations, not even according
to utilitarianism.
For instance, the version of prior existence utilitarianism that I have
brought forward here would indeed imply that we do not need to bring
any individual into existence. If, however, we decide to bring an indi-
vidual into existence, we should make sure that the individual we are
bringing into existence (whoever it will be) is as well off as possible. There
is nothing inconsistent or intransitive about that view.
D. Arrhenius on the Relation between the Prior Existence
View and the Person-Affecting Restriction
In Chapter 5 I have argued that the Prior Existence View needs two
assumptions in order to be a coherent utilitarian view on whose welfare
to count in the aggregation of welfare. It needs the assumption that in
the evaluation of outcomes the focus should be on harms and benefits to
sentient beings rather than on the quantity of welfare as such. This view
on the evaluation of outcomes is called the Person-Affecting Restriction.
In order to be a coherent utilitarian view, the Prior Existence View also
needs the assumption that causing a being to exist cannot benefit or
harm that being. I have argued that those two assumptions together
imply the Prior Existence View.
Contrary to this line of thought, Arrhenius has argued that the Person-
Affecting Restriction does not imply the Prior Existence View, or what he
calls 'Necessitarianism', even if it is true that bringing a being into exist-
ence cannot harm or benefit that being. Arrhenius attempts to show
that, irrespective of whether or not coming into existence can make
a being better or worse off, the Person-Affecting Restriction does not
imply any version of Necessitarianism, nor Actualism, nor Presentism. 11