how the commitment to impartiality is to be worked out: Should we, for
example, use utilitarian calculation, or should we take what people would
agree to as our touchstone? Clearly, if we take the scope of impartiality to
extend across individuals in their ordinary lives, then the requirements of
impartiality will be very diVerent from what they would be if we were to take
impartiality to be primarily a requirement of moral and legal rules. And
similarly, if we work out the commitment to equality via straightforward
utilitarian calculation we may arrive at diVerent practical conclusions from
the ones which would result were we to appeal to what people can, or could
reasonably, agree to.
To give an example, utilitarianism is an impartialist theory: classical
utilitarians are committed to treating everyone equally, and they claim that
the right way to do that is to count each person as one and no one as more
than one. However, utilitarianism, particularly in this straightforward form,
may dictate that great sacriWces be made by some people in order that overall
welfare be increased. It may require that a minority live in servitude, if that is
what is needed to maximize welfare. This, however, seems to some to be the
wrong kind of argument against slavery, and it is this thought that prompted
John Rawls to propose an alternative interpretation of impartiality, one which
rests upon agreement rather than maximization of utility. Rawls writes:
‘‘while there may be some excuse for slavery in special circumstances, it is
never an excuse for it that it is suYciently advantageous to the slaveholder to
outweigh the disadvantages to the slave and to society.... since slavery does
not accord with principles which they [the slaveholder and the slave] could
mutually acknowledge, they may each be supposed to agree that it is unjust’’
(Rawls 1958 , 190 ). In short, impartiality as eYciency is diVerent from impar-
tiality as agreement. So, even if the impartialist commitment to equality is
clear, the scope and character of that commitment—the contexts in which it
applies and the way in which it is to be worked out—are not.
Third, there is considerable disagreement about whether impartiality
(however understood) is a good thing, or even a possible thing. Thus,
Bernard Williams has insisted that ‘‘somewhere one reaches the necessity
that such things as deep attachments to other persons will express themselves
in the world in ways which cannot at the same time embody the impartial
view, and that they also run the risk of oVending against it’’ (Williams 1981 ,
18 ). He draws attention to the fact that the dictates of impartial morality may
conXict with personal ties and aVections that matter very much to us, and he
notes that, when such a conXict occurs, we may wonder why we should
424 susan mendus