Appendices - Killing Happy Animals: Explorations in Utilitarian Ethics - page 160

counterpart in W n , who is worse off than her counterpart in W Jane ,
Jane. 18
It follows from the Personal Dominance Assumption that Charlotte
ought to prefer each subsequent world above its predecessor, because in
each subsequent world Charlotte's baby is better off than its counterpart
in the previous world.
The final assumption is that preferences are transitive. It follows that
Charlotte rationally ought to prefer the world in which she has Jane
above the world in which she has James.
Transitivity:
If you prefer that W 1 rather than W 2 be actual, and you prefer that
W 2 rather than W 3 be actual, then you prefer that W 1 rather than W 3
be actual. 19
So, Hare has provided an argument based on the assumption that iden-
tity is vague and on four rather uncontroversial additional assumptions.
According to Hare, claiming that the 14-year-old girl in Parfit's case
might as well have the child immediately proves to be morally and/ or
rationally deficient.
Let us now see how the argument works for the alternative theory
about identity across states of affairs. Let us assume the 'real identity
across states of affairs theory'. This theory holds that there can be real
identity across states of affairs. Departing from this theory, the Personal
Dominance Assumption would read as follows:
Personal Dominance [assuming real identity across states of affairs]:
For states of affairs S 1 , S 2 , if Charlotte has one baby in both, and her
one baby in S 1 is her one baby in S 2 , and that baby is better off in S 1 ,
then, all other things being appropriately equal, she ought to prefer
that S 1 come about. 20
Now, it matters whether the baby that Charlotte would have in S 2 is
indeed the same baby that she would have in S 1 . This, again, depends on
one's view about personal essence. It is plausible to assume that essence
is somewhat fragile. That means that large enough differences of the
identity-determining property or set of properties result in different
babies. However, it is also plausible to assume that essence is not perfectly
fragile. So, some difference in the identity-determining property or set
of properties might be compatible with the baby in S 2 still being the
same baby as the baby in S 1 . That means: 'All actual people could have
been ever so slightly different along any given dimension.' 21