conventional view in defense of its moderate claim of the relation between
justice and luck. The objection, however, does not seem fatal.
First of all, insofar as the objection voices a worry about the condition for
responsibility being too stringent, so that it is impossible for that condition to
ever be met, the objection is misdirected. The objection, that is, seems to
point to the fact that requiring the neutralization of background luck is
premised on the view that people should have full control over the factors
which affect their actions, and then holds that, since no one ever has full
control over all the factors on which their actions depend (for one thing, no
one has control over being born, which is one such factor), requiring that
kind of control undermines the very ideas of responsibility and desert. It is
worth noting here that this is how John Rawls’s critique of desert has
sometimes been interpreted (Rawls 1999 , 89 ; Nozick 1974 , 214 , 225 ; Scheffler
1992 ): Rawls has been taken to suggest that, in order to deserve, people would
have to be responsible, and deserve, ‘‘all the way down;’’ but since it is
impossible to have this kind of regressive control over the factors affecting
one’s achievements, no one can ever deserve anything. In fact, as some have
pointed out, this interpretation of Rawls is mistaken (Cohen 1989 , 914 ; see
also Moriarty 2003 ). At any rate, the fair opportunity view does not need to
hold that this kind of full and regressive control is necessary. 6 Rather, it holds
that only inequalities in luck be neutralized, and only insofar as these affect
different people’s ability to bring about a specified event (such as the event of
‘‘earning as much money as Jones’’). If both Smith and Jones have the same
talent and choose to exercise it to different degrees, thereby producing
different performances, the fact that Jones’ performance reflects his talent is
not a problem from the point of view of justice.
A second claim that might be made in reply to the objection is as follows: if
the objection is one about the difficulty in identifying the presence of unequal
luck and discounting for it (Rawls 1999 , 274 ; Cohen 1989 , 915 ; Moriarty 2003 ,
523 – 4 ), then, once again, the fair opportunity view can survive this objection.
This is because we may go some way toward realizing the demands of choice-
based desert by operating on the background conditions against which desert
claims arise (rather than discounting for unequal luck once it occurs).
Ensuring free and equal high-level primary and secondary education, for
example, goes some way towards ensuring that making access to universities
6 See Zimmerman ( 1987 ) for the distinction between full or unrestricted control of this type, and
restricted control. See also Susan Hurley’s discussion of the ‘‘regressive control’’ requirement of
responsibility; Hurley ( 2003 ).
justice, luck, and desert 447