if responsibility for past injustice depends on the persistence of both the
wrongdoer and victim over time, what happens when one of them disap-
pears, or is absorbed into another (Posner and Vermeule 2003 , 738 ; Spinner-
Halev 2004 )? 3 It might be that descendants of the victims in these cases have
no remedy, because there is no agent to whom responsibility can be assigned
in any meaningful way. Sometimes a new state is advantaged by an injustice it
did not commit; and sometimes descendants of the victims have gone on to
live decent lives without ever receiving any remedy (or even asking for one).
Are they still owed some form of reparation? Not all historic injustices matter,
because not all of them feature in the shared history and collective memory of
people, or a state, in the same way (Margalit 2002 , 94 – 104 ). And collective
memory can be manipulated to dreadful eVects, as we know. However, that
does not mean the best strategy—moral or prudential—is simply to ignore
claims about historical injustice. Although they can be forms of interest-
group special pleading, they also often point to wider issues of structural
injustice and unfreedom.
This leads to a third problem which has also been mentioned above, but
bears repeating. Is it not the case that most claims for reparations that weWnd
compelling overlap with ongoing disadvantage? And if so, what real work is
thehistoricalnature of the injustices doing? Maybe it reminds us of how badly
people have been treated in the past, or the harm that states and peoples can
do. But that does not necessarily give us any additional reason for doing
justice now; our reason for helping people now has to do with their current
situation, not the fact that they were badly treated in the past. So when
historical injustice and contemporary disadvantage overlap, the case for
reparations is considerably stronger, but mainly because of forward-looking
reasons (distributive justice, reconciliation, non-humiliation) not backward-
looking ones. But I think our intuitions are pulled in diVerent directions here,
and we should acknowledge the tension as opposed to eliding it. The relative
wealth of present-day African-Americans or Jews does not lessen the wrongs
of slavery or the Holocaust, and so we should not, by deWnition, reduce all
claims for reparations to claims about contemporary disadvantage. Facts
about relative and absolute levels of well-being matter for making political
judgments in these cases, but the historical facts also matter. They matter
because historical injustice plays a crucial role in shaping how the rules and
3 I am grateful to JeVSpinner-Halev for pressing me on this point and for helpful conversations
about these issues in general.
522 duncan ivison