Show Summary Details

Page of

Printed from Oxford Research Encyclopedias, Politics. Under the terms of the licence agreement, an individual user may print out a single article for personal use (for details see Privacy Policy and Legal Notice).

date: 23 February 2024

Historical Injusticefree

Historical Injusticefree

  • Robert HusebyRobert HusebyDepartment of Political Science, University of Oslo


Historical injustice generally refers to theories that claim that past wrongs may have implications for current distribution. While it is not self-evident how to define historical injustice, many theorists reserve the term for wrongs whose perpetrators and victims are now both dead. Historical injustice thus concerns the potential claims that the descendants of the victims have against the descendants of the perpetrators. An example is the potential claims descendants of slaves in the United States have against the descendants of slave owners (or perhaps descendants of all non-slaves). While many find claimsbased on historical injustice intuitively compelling, such claims have also been subject to criticism on a wide range of theoretical grounds.


  • History and Politics
  • Political Philosophy

What Is Historical Injustice?

There are various ways of understanding historical injustice. Injustice as such, however, will not be defined in this context, since historical injustice might be relevant to many (but probably not all) conceptions of justice. Thus, it is the historical part that is of main interest. The term historical gives associations to the passage of time, but it is not the passage of time that is the most important aspect. Typically, historical injustice refers to cases in which both the victim and the perpetrator of the injustice are dead (Blomfield, 2021; Ivison, 2009; Perez, 2011; Spinner-Halev, 2012),1 and the question is whether and how the fact that an injustice is committed remains relevant in such circumstances.

Historical injustice, then, usually concerns the claims and duties of the descendants of the original perpetrators and victims (Perez, 2011). Furthermore, historical injustice tends to refer to cases concerning the claims of the descendants of historically wronged groups, such as African Americans, women, or Indigenous peoples (Nuti, 2019), whereas the duties to compensate are typically assumed to lie with other (majority or sometimes dominant minority) groups, because these in most cases comprise the descendants of the wrongdoers (Spinner-Halev, 2012).2 It is worth noting, however, that even if the most politically charged cases of historical injustice concern groups, the same questions can easily arise on the individual level as well. Because of this, examples involving individuals rather than groups will sometimes be used.

Note also that some cases of historical injustice could be likened to mere theft. If A steals B’s car, and they both die, it seems plausible that A’s descendants should give the car back to B’s descendants. At least this is the case if the theft occurs within a jurisdiction with an established, fair, and well-justified system of property rights. Such clear cases of theft will not be addressed here as they seem resolvable without reference to historical injustice as such. Internationally, of course, there is no established, fair, and well-justified system of property rights. At least that is the case if one does not subscribe tointernational libertarianism, according to which states may own or steal territory in a theft-like manner. Colonialism, for instance, might be seen as theft, and if so, it can be argued that the colonizers should simply return the stolen property. International libertarianism is not, however, a generally plausible theory of international distributive justice (see the section “Theories of Historical Injustice”), and the right answer in cases of historical injustice that includes stolen property or, specifically, territory, might be more complex, and involve a fair distribution in light of present needs or life plans (Waldron, 1992).

The purpose of this article is to assess arguments for and against the normative relevance of historical injustice. First, some skeptical arguments are presented. These rely on epistemic difficulties, challenges related to collective responsibility, supersession, and the non-identity problem. Even though these arguments do not necessarily strike at the normative core of claims from historical injustice, they are clearly problematic in most actual, complex, large-scale cases. Next, some prominent accounts that are assumed to counter (at least some of) these skeptical views, are examined. However, these accounts do not seem to succeed as general defenses of the relevance of historical injustice (even though they may be successful in more specific respects). Furthermore, the so-called redundancy thesis is assessed. According to this thesis, claims from historical injustice either conflict with claims from (more general theories of) distributive justice, in which case they risk becoming irrelevant, or else they coincide with claims from (more general theories of) distributive justice, in which case they risk becoming redundant. Clearly, the redundancy thesis relies on the assumption that claims from distributive justice are typically weightier than claims from historical injustice. One reason for this might be that historical justice, as noted, refers to injustices toward the forebears of the claimants, committed by the forebears of the presumed duty bearers. Injustice between currently living individuals seems intuitively more pressing.

A possible defense against the redundancy thesis exploits the fact that not all theories of distributive justice are complete and that claims from historical injustice could still be relevant (Blomfield, 2021). That they could be relevant, however, does not mean that they are relevant. And the lack of convincing positive arguments in favor of the normative relevance of historical injustice seems to lead toward a skeptical conclusion. This notwithstanding, a more modest claim is also examined. This is the claim that even if historical injustice is not generally normatively relevant, it may be relevant as a tie breaker in cases where different justice claims compete for attention. This suggestion too, however, seems unconvincing upon further scrutiny.

It is worth saying a bit more about what normatively relevant means. It is conceivable that claims from historical injustice have some weight in some situations, without ever being decisive. Suppose, hypothetically, that claims from historical injustice did, in fact, have some normative weight but that they always conflicted with claims from distributive justice and that, moreover, claims from distributive justice always won out. In that case, claims of historical injustice would never have any influence on what should be done, and they would in that sense be normatively irrelevant not because they do not exist or have no weight but because they never play any decisive role in the deliberations about how to act and what to do. Of course, claims that are normatively irrelevant may also be untrue or baseless, but that requires further argument. This article, then, does not ask whether claims from historical injustice are, in general, simply baseless, but rather whether they ever play any role in deciding what one normatively ought to do. Note, however, that if claims from historical injustice never play any role in the practical deliberation, one possible reason for this might, of course, be that they are baseless. But that is not the argument made in this article.

Furthermore, it is argued that symbolic acts, such as public apologies in response to historical injustice may nevertheless be appropriate under some circumstances, that is, circumstances under which such symbolic acts have beneficial consequences, for instance in virtue of building trust in divided societies. Before concluding, a brief analysis is made of the relationship between historical injustice and the so-called beneficiary-pays principle (BPP).

Critical Views

Historical injustice often seems intuitively relevant (Blomfield, 2021; Ivison, 2009; Nuti, 2019). In many cases, the descendants of the original victims are still marginalized and disadvantaged minorities with apparently very plausible claims for compensation and redistribution. It is sufficient to think of many contemporary African Americans, Indigenous Americans, and Indigenous Australians. These minorities are disadvantaged relative to the majority population, and it is very hard to avoid the thought that this state of affairs is directly related to the injustices that have historically been committed against these groups (McCarthy, 2004; Valls, 2018).3 When contemporary African Americans demand reparation for the historical injustices committed against their forebears, this appears to be a very reasonable claim, given what we know about history and contemporary American society. Nonetheless, claims from historical injustice have been met with substantial theoretical (and political) resistance.

First, some theorists have argued that many such claims fail because of the non-identity problem (Ivison, 2009; Parfit, 1984; Roberts, 2022; Thompson, 2001). When some time has passed since an injustice, it is very likely that the descendants of the original victims would not have been born had it not been for the injustice itself. The injustice is one of the (infinitely) many necessary conditions for their existence.4 Consider contemporary African Americans who descend from the original slaves in the Antebellum South. It is highly unlikely that they would have existed had it not been for slavery, that is, if their forebears had not been forcefully abducted from Africa. If these descendants nevertheless now have lives that are worth living, even if they are worse off than, say, the descendants of the slave owners, they cannot be said to have been relevantly harmed by slavery because without it, as noted, they would not have existed in the first place, and existing, at least with a life worth living, is arguably preferable to not existing (Parfit, 1984).

Another challenge is that there are massive epistemic problems connected with the attempt to figure out what people are owed due to historical injustices (Ivison, 2009). Even though there are many instances of historical injustice that are relatively recent and well documented, we obviously lack information about most of the injustices that have occurred throughout history. A related problem is that it is quite likely that (close to) all current holdings result from injustice in one way or the other. After all, history is long, and injustice is commonplace. This might of course be used as grounds for some sort of onetime global redistribution, as Nozick (1974) suggested, but epistemic problems at least make it hard to single out particular claims from historical injustice as valid. While some such claims could be well documented, it might be hard to argue for compensation in particular cases, when there are bound to be countless other claims that are valid in principle but for which evidence is impossible to produce (see Nuti, 2019, p. 3).5

A third problem is that collective responsibility is hard to establish. Most claims from historical injustice are directed at specific collectives, for instance (former) colonial powers or majority groups. For such claims to make sense, some sort of responsibility must be assigned to these collectives.6 According to some of the most sophisticated accounts of collective agency (List & Pettit, 2011), collective agents must be organized in particular ways in order to achieve agency at a level sufficient to assign moral responsibility to them. This includes decision mechanisms that ensure collective rationality over time. Also, collective intention at some level is necessary. It is unlikely that groups such as (former) colonial powers, White Americans, men, or majority groups in societies that encompass Indigenous peoples are typically organized among themselves in the required way. It is therefore challenging to assign collective responsibility for compensation to such groups (which would be true even if there turned out to be some exceptions; see also Ivison, 2009; Thompson, 2001, pp. 13–14).

Fourth, it has been suggested that injustice may be superseded if enough time passes. In cases of colonialism, for instance, the descendants of colonialists may over time become as attached to the colonized land as the original inhabitants, and it could arguably be wrong to ask the descendants of the colonialists, who may themselves be (relevantly) innocent, to give back (all) the land (Waldron, 1992). If this is the case, it would serve to undermine many claims of historical injustice. On the other hand, supersession, if justifiable, might help blunt the force of the second problem noted earlier, because it seems easier to gain knowledge about recent injustice, all else equal. This is no real consolation to defenders of the normative relevance of historical justice, however, since one of the problems is likely to remain.

All these issues seem quite severe, but it is worth noting that none of them comprise general objections to claims of historical injustice. They mainly point to various ways in which the application of a (potentially justifiable) principle of historical injustice might be limited. Historical injustice can very well have occurred (relatively) recently, on the individual level (as noted), in cases where information is ample,7 and in which the identity of the victim’s descendants does not depend on the injustice (for instance if the descendants were conceived prior to the injustice). Even if these critical perspectives would indeed be quite problematic and rule out many of the most contentious and discussed claims of historical injustice, it could still be true that claims based on historical injustice could be perfectly valid in some cases, which would be interesting enough, even if less practically relevant.

Theories of Historical Injustice

Several attempts have been made to circumvent these problems even in larger, collective settings. Bernard Boxill, for instance, in a discussion of Black reparations in the United States, claims that descendants of slaves can be said to have been relevantly harmed even if their identity depends on slavery having obtained. The reason is that such a descendant will also suffer harm after they are conceived. The harm will consist in the current government’s failure to bring their parents out of the poverty and ignorance that has unfairly been imposed on them, leading the descendant to grow up under disadvantaged conditions. And so on for that descendant’s children (Boxill, 2022). This argument, basically, is an argument to the effect that descendants of slaves suffer current harm. This argument seems perfectly sound and may well support the case for Black reparations, which is Boxill’s aim. Note also that the argument explicitly avoids the non-identity problem. The epistemic problem is partly circumvented in that the horrors of slavery are well known and documented. However, a challenge might still be that there are likely to be many other, undocumented, and (at least possibly) equally gruesome injustices that must go unmet. The problem of supersession does probably not arise in this case, since slavery is a relevantly recent wrong. It is less clear that Boxill’s account avoids the problems associated with collective agency, however. The reason is that descendants of slave owners (or, more generally, the descendants of non-slaves), hardly compose a collective agent in the required sense. The group that these people constitute is presumably not conscious of being a group and does not have collective intentions or any kind of decision mechanisms.

More importantly, however, the argument would not work as a general justification of the normative relevance of historical injustice, exactly because it relies to a large extent on current harm. While current harm to descendants of victims of historical injustice is sadly not uncommon, general claims from historical justice are not constrained in this way. Such claims should be valid, at least to some extent, even in cases in which the descendants of the victims do not suffer current harm. If not, one is left wondering exactly what work the historical injustice is doing. If the government fails to bring a person’s parents out of the poverty and ignorance that has unfairly been imposed on them, then clearly the government unjustly harms both the person in question and their parents. But that would be true of all persons who fit that general description, regardless of whether the parents and offspring descended from slaves or other victims of historical injustice. Anyone who suffers from injustice should be compensated, and hence, historical injustice makes no difference to the claims in question.

Boxill also has another, inheritance-based and Locke-inspired argument, according to which the original slaves had moral claims for compensation that were never honored, and their descendants have thus inherited these unmet claims. Furthermore, this claim can be pressed against the state, not only the descendants of the slave owners. The reason is that the general population did not express dissent, to a sufficient extent, to the institution of slavery.8

Since the slaves had titles to reparations against the estates of those who assisted, concurred, or consented to their enslavement, they would have titles to reparations against practically the entire white population, not just against the estates of the slave holders.

And just as the descendants of the slaves have inherited claims for compensation, descendants of non-dissenting members of the White population have inherited the duty to pay compensation.9

It seems that this argument too, steers clear of the non-identity problem and the supersession problem. There are still some reasons to doubt the inheritance argument, however. First, as was the case with Boxill’s first argument, it could be difficult to single out even very clear cases of historical injustice for compensation, when we have strong reason to think that history is replete with injustice. It is hard to rule out that the forebears of the original slaves were non-dissenters to other wrongful harms. And many of the White settlers in the United States fled from religious persecution in other countries. So, while the claim as such could be sound, it would be less sound to single it out for unique treatment, despite the fact slavery was a truly horrific wrong.10

Furthermore, there is a problem with the account of responsibility underlying the account. It does not seem plausible to hold that a person’s forebears’ non-dissent straddles one with heavy compensatory duties. Non-dissent to crimes and wrongful harm is certainly blameworthy but not in a way that travels across generations. If Alice’s great-grandfather did not dissent to Bert’s great-grandmother’s wrongful harm toward Cynthia’s great-grandfather, it seems unconvincing that Alice should now have compensatory duties toward Cynthia (assuming her great-grandfather’s claims were never met).11 As for the non-identity problem and the supersession problem, they do not seem problematic in this case to the extent that argument relies on inheritance and relatively recent injustice. Finally, and generally, the inheritance argument, even if sound, would not amount to a general defense of claims from historical injustice, because not all such claims are reasonably seen as cases of lost inheritance. (It also seems that cases that can reasonably be seen as cases of lost inheritance would be resolvable within a fair system of property rights, at least within a specific jurisdiction.)

Alasia Nuti’s (2019) theory of historical justice also rests on current injustice:

The chief argument . . . is that the unjust history that should normatively matter in justice-based considerations is present because it has been reproduced over time through different means, and it still is so. The reproduction of unjust history is pervasive as it shapes the background conditions in which some present wrongs occur and relations between agents are established. At the same time it is dynamic because it is enabled also through the agent’s actions and interactions. In this respect, like return-of-the-past narratives, the book does conceptualise history as present; however, it does so in a more complex way. (p. 4)

This theory avoids the epistemic problem, since it singles out a distinct subset of injustices as relevant, namely, those that are present, because they have been reproduced over time. It also, for the same reason, avoids the non-identity problem precisely because it singles out injustices that are present, and a person’s identity does not (necessarily) depend on the injustice they are currently subject to. By the same token, the supersession problem is resolved. Most likely, the account also avoids problems to do with collective responsibility, since it is reasonable to think that governments of modern societies are responsible for eradicating current injustices.

In Nuti’s view, moreover, history is vital to being able to recognize the many and subtle ways in which injustice occurs in the early 21st century. This includes injustices related to prejudice and stereotyping of groups such as women, Afro-American men, and Indigenous Australians (Butt, 2021; Nuti, 2019). Nuti’s theory is in an important sense structural, in that it highlights the historically conditioned structures within which current injustices take place (Butt, 2021; Nuti, 2019).

As Butt (2021) has pointed out, however, this view does not (necessarily) concern the claims of descendants of victims of injustice and their claims toward descendants of the perpetrators of injustice. Rather, it concerns the claims of current victims of structural injustice and is historical in the sense that it insists that history is vital to uncover (and eventually rectify) such injustices. This seems very plausible, and one would be hard-pressed to deny that history can be important in this respect. Nevertheless, in many ways, this is a theory, like Boxill’s, of (current) injustice, rather than a theory of historical injustice. Thus, the main tenets of Nuti’s theory can be accepted and, at the same time, the relevance of historical injustice in the sense in which it is understood in this article can be denied. Victims of structural injustice are victims of injustice here and now, even if that injustice has historical roots. Thus, it seems that Nuti fails to provide a convincing general defense of claims from historical justice, plausible as her account may otherwise be.

Janna Thompson’s theory of historical injustice conceives of nations (and some other groups) as intergenerational communities. Each generation in such communities, she claims, thinks that they are entitled to make some moral demands on succeeding generations, for instance the demand that future generations honor some of the commitments made by the present members. By the same token, present members should honor at least some of the commitments made by past people. One example of such commitments may be the signing of treaties (Thompson, 2001, p. xvii). Crucially, Thompson (2001, p. xix) holds that this duty to honor commitments made by prior generations includes the duty to make reparations for past injustices made by these prior generations.

This account might well avoid the problem of collective responsibility. Since Thompson is concerned with national communities, it is likely that these communities see themselves as a group with a certain level of collective intentions and relevant decision procedures.12 If so, the non-identity problem is also bypassed, since the identity of collectives is not dependent on the precise identity of its constituent members. The supersession problem might or might not arise depending on when a particular injustice occurred. However, the epistemic problem seems to remain.

Furthermore, while it makes sense, at least in many cases, to say that present generations are obligated to honor treaties signed by their forebears, this can be adequately explained in light of the general benefits of predictable and stable (international) cooperation. It is, however, hard to see that this implies a duty on the part of current people to make reparations for the injustice committed by people who are dead, just because the two groups happen to belong to the same nation. In light of this semi-reciprocal theory, it seems that present generations can now impose moral duties on future generations to repair their current injustices. And that does not seem reasonable. However, Thompson’s account of reparation is largely reconciliatory and forward-looking. As such, it is in agreement with the main thrust of this article, even if the view of reconciliation defended here is different from Thompson’s account.

Thompson also advances two other arguments. The first supports the idea that individuals can make reparative claims if their inheritance is unjustly taken from them (Thompson, 2001, pp. 114–117). While she is (rightly) skeptical of inheritance in general, she thinks that respect for what she calls lifetime-transcending interests might ground a duty to respect at least some of the wishes that people have concerning what should happen to their possessions after their death (Thompson, 2001, pp. 114–117). (Note that there are many concerns, for instance equality, that might limit the right of bequest.)

As noted (in the section “What is Historical Injustice”), fair systems of property rights within a well-functioning jurisdiction might ground certain inheritance-based claims of reparation. If so, however, it is the system of property rights that does the work and not an (independent) account of historical justice.13

In addition, Thompson holds that individuals can make reparative claims on the basis of belonging to family lines that have been subject to injustices. Slavery is an example. It seems, however, that a necessary condition for claiming reparation on this basis is that the (current) claimants are also now harmed by ongoing justice. It is thus not clear what actually justifies the claim. If it is current harm, then it is not a case of historical harm as understood in the present context, even though the roots are historical (Thompson, 2001, pp. 143–145). Apart from that, the other problems, except, again, the epistemic problem, would probably not arise.

Daniel Butt (2009) defends a theory of historical injustice that has many interesting aspects. The theory depends, however, on a quite controversial assumption (which Butt himself, for what it is worth, denies): “My argument has a conditional character—if one accepts the international libertarian account of distributive justice, then one should consider the . . . morally relevant forms of potential connection with past wrongdoing which the book outlines” (Butt, 2009, pp. 13–14).

International libertarianism is, however, not a very plausible proposition. It will lead too far to canvass the view in a comprehensive fashion, but it is worth pointing out that the distribution of natural resources among the states of the world is so random as to make it odd to just assume that it is morally all right at the outset, before considering questions of historical injustice. Furthermore, as Thompson (2001) notes,

in every area of the world, wave after wave of invaders have expropriated land and political control from existing occupants, As a result, virtually no nation can establish that it has a historical title to the territory it occupies. In fact, we know that most cannot make this claim. (p. 57)

Thus, if one accepts international libertarianism, oneshould, really, be preoccupied with figuring out how to distribute the world’s resources in such a way that it could serve as a morally sound starting point. Of course, such a redistribution would take care, presumably, of a lot of historical injustice. And historical injustice would still be a relevant and valid concern in such a view. The point here is just that international libertarians have more general issues to handle other than the specific issue of historic injustice.

More importantly, however, international libertarianism, in contrast to most versions of domestic libertarianism, typically rests on some account of collective agency. States, nations, or peoples on such views, can act in ways that make them responsible, as collectives, over time. Rawls (1999) and Miller (2007) are clear examples of this. Such an account of responsibility, however, seems very unfair on the individual level. To take Rawls’s illustration: Two peoples make different choices regarding industrialization and economic development. Consequently, a few generations down the line, their respective levels of welfare differ significantly (but both peoples are sufficiently well off, we can assume). In this case, Rawls holds, the wealthier society has no duties of distributive justice toward the less wealthy society (Rawls, 1999).

This might seem intuitively plausible at first blush, but it is hard to sustain the view on closer examination. The current individuals in the less wealthy society had no influence whatsoever on the choices made generations back, and many of them might also regret that they were made. And it seems wrong to straddle individuals with responsibility for what their forebears did. The same is true of the current generation in the wealthier society. They, too, have had no influence over the decisions made generations ago, and it is unclear that they now have moral claims to the wealth acquired by their forebears. Note also that on international libertarianism, it is seemingly no problem that some states just happen to have access to very valuable resources, whereas others are less lucky. The Norwegian state, for instance, has invested large parts of its oil revenue in a fund that is worth close to US$1,400 billion, or more than three times the country’s Gross Domestic Product in the early 2020s. It is hardly plausible to think that all that wealth rightly belongs to Norwegians, just because they were lucky enough to have access to a valuable resource. Thus, regardless of Butt’s internal arguments, it seems that international libertarianism should not be accepted.14

Finally, it is worth noting that this view would evade the non-identity problem. The epistemic problem could very well arise, as could the supersession problem (depending on when the injustice took place). The problem of collective agency could also arise.

The Redundancy Thesis

Even if some general account of historical injustice could be defended and could avoid the problems noted in the section “Critical Views,”15 there are further challenges. A broader criticism of claims of historical injustice is that such claims are redundant (Blomfield, 2021; Nuti, 2019). The reason is that most general theories of distributive justice (Blomfield uses egalitarianism as her example) will either already encompass claims from historical justice, or else deny them, explicitly or implicitly. Also, most such theories already entail, more or less automatically, their own (at least partial) accounts of corrective justice, in the straightforward sense that any breaches of justice should be rectified.

Egalitarians, for instance, hold that just distribution requires equality in some relevant respect. For most egalitarians, this implies that whenever inequality obtains, it should be rectified or corrected. And this will be true about many egalitarian theories, regardless of whether the inequality is caused by injustice, historical injustice, or natural events, such as floods or earthquakes. More generally, many theories of distributive justice will be of this kind. This does not mean that claims from historical injustice have no force, but it raises the question of what place such claims should have in the overall theories of justice. And that is not an easy question to answer.16

In response to critical views along these lines, Blomfield points out that not all accounts of distributive justice are complete in the sense that they specify one correct place for all relevant goods or advantages (Blomfield, 2021; see also Cohen, 2016; Huseby, 2015). Relational egalitarianism is a case in point. Relational egalitarians are not preoccupied primarily with the distribution of goods as such (Anderson, 1999; Blomfield, 2021; Lippert-Rasmussen, 2018) but with the quality of the relations between individuals. Thus, if people can interact in society as relevantly equal, other inequalities (in resources or welfare, say) may not matter as such.17 To be sure, relational egalitarianism can be specified in different ways, but as a general characterization, this will suffice.

While Blomfield is primarily concerned with forms of egalitarianism, her point can be generalized. Several forms of sufficientarianism, for instance, will also be incomplete in this sense and fail to specify a correct place for all goods (Axelsen & Nielsen, 2015; Huseby, 2010). Moreover, such theories may be compatible with at least some level of inequality in resources and welfare. Other views, such as utilitarianism (Singer, 2016), overall outcome welfare prioritarianism (Holtug, 2010), and luck egalitarianism (Arneson, 1989; Cohen, 1989) will most likely specify the holdings of all relevant goods.18 The same will also be true of some historically oriented theories, such as Nozick’s.19

To the extent that theories of distributive justice are incomplete, however, there may well be room for corrective claims, based on historical injustice (or other considerations) that are not encompassed by, nor contradicted by the distributive theory itself (Blomfield, 2021).20 Some distributions, for instance certain inequalities, may be unjust for historical reasons, even if they are not unjust from the point of view of some incomplete theory of distributive justice, such as for instance relational egalitarianism.

Blomfield (2021), drawing on Nuti (2019) also notes two other ways in which historical injustice may have a role to play even within the context of more general theories of distributive justice. First, if a theory, such as luck egalitarianism, for instance, deems two inequalities to be unjust and one of these inequalities results from historical injustice, then this inequality could possibly be prioritized over inequalities that result from chance (see Duus-Otterström & Page, 2023). Third, current inequalities that result from historical injustice may have to be addressed in particular ways, for instance through symbolic acts such as apologies from the majority to the historically unjustly treated minority (Blomfield, 2021). These three suggestions are discussed in turn in the next three sections.

Historical Injustice as Complement to Incomplete Theories of Distributive Justice

According to Blomfield’s (and Nuti’s) first suggestion, a given distribution of goods may be unjust in light of some conception of historical injustice even though it may not be unjust from the point of view of an incomplete theory of distributive justice, such as, for instance, relational egalitarianism (or sufficientarianism). Suppose that a society is perfectly fine as far as social relations are concerned. Everyone can interact as equals in all relevant ways, even though there are material differences among them. However, society is nevertheless marked by historical injustice, but these injustices are not such that they interfere with equal (and therefore just) relations.

Now, if equal relations is what justice requires, it should be asked why exactly historical injustices matter as well. They cannot matter in virtue of upsetting equal relations, since it is assumed that they are indeed equal. So they must matter in virtue of something else. Perhaps historical injustices matter in virtue of an unequal distribution of goods of some sort. It is conceivable (although not very likely) that a society could consist of two groups that enjoy equal relations but not equal access to some relevant goods. If it is supposed that the forebears of the materially least advantaged group have been victims of historical justice, people might start to think that something is nevertheless morally wrong. If so, however, it must further be asked what the wrong consists of, exactly. If it consists of the systematic material disadvantage along socially salient group lines, this just means that the distribution of material goods should also matter and that relational egalitarianism is not all there is to say about distributive justice. But then, of course, any systematic material inequality among group lines should be problematic. Or else some explanation must be given for why material disadvantage along group lines is problematic only when it is caused by (or correlates with?) historical injustice. But it is hard to see what kind of explanation that could be.

Again, relational egalitarianism is compatible with claims from historical justice, but it is unclear why such claims should be acknowledged by that theory. Historical injustice, as noted, is most naturally seen as a part of corrective justice. And corrective justice is the answer to how to rectify injustice. On relational egalitarianism, injustice does not cover inequalities that do not upset or endanger equal relations. Thus, in order to accept claims from historical injustice, it seems that some other theory of justice has to be affirmed, according to which historical injustice leads to inequalities or patterns of distribution that are now unjust. As noted, this is not to deny the possibility. However, there are reasons to doubt that historical injustice, as such, is morally relevant from the standpoint of most theories of distributive justice, even incomplete ones. In addition, as argued (in the section “Theories of Historical Injustice”), there does not seem to be any persuasive positive arguments in favor of the moral relevance of historical injustice in the literature. In light of this, it seems reasonable to be skeptical of claims of historical injustice.

Historical Injustice as Tie Breaker

This does not suffice to rule out the possibility that the elimination of historical injustice should (always or sometimes) be prioritized over the elimination of other kinds of injustice. This would amount to a quite weak, but still significant, difference between claims from historical injustice and other justice claims. It could be held that historical injustice is relevant only in the sense that it could serve as a tie breaker in cases of conflicting claims.

Duus-Otterström and Page (2023) have presented an argument along such lines. They claim that people have impersonal reasons to rectify injustice that are such that they should prioritize assisting the victims of injustice over victims of mere harm, all else equal. These reasons are general and do not only apply to those who have caused harm themselves or their descendants. Even though the descendants of victims of injustice are not (necessarily) victims of injustice themselves, it could be the case that Duus-Otterström and Page’s claim could cover such cases as well. However, their arguments do not seem to succeed even in cases of non-historical injustice (and, presumably, if this is right, it follows that their claims also fail in cases of historical injustice, since current injustice would seem to provide stronger reasons for moral agents than historical injustice). Suppose that A is a victim of injustice and B is a victim of some random harm and that A and B are otherwise relevantly equal and that they are harmed to the same extent. Now, assume that A’s wrongdoer has fled, that you are a bystander, and that you are willing and able to assist but that for some reason you can only assist either A or B. Should you choose to assist A without further ado, just because they are a victim of injustice, whereas B is not?21

To the extent that such prioritizations are allowed, it seems that some new form of unfair disadvantage or injustice will be allowed. If A is to be prioritized over B, even though their current situation is similar (barring the origin of these situations), then A gains an important advantage over B for reasons that are completely outside the control of both A and B. And that seems unjust (Arneson, 1989; Cohen, 1989). B is then deprived of a chance of being assisted and thus suffers an additional wrong.

Consider this in a situation in which very much is at stake. Suppose A and B are both in equally life-threatening situations but for different reasons. A is the victim of injustice and B is the victim of bad luck. If you can only rescue one of them, it seems disrespectful not to give B even a sliver of a chance at being rescued. Prioritizing in this way would also mean that B would have strong reasons to wish that she was the victim of injustice rather than bad luck. The reason is that if she was a victim of injustice, her chances of survival would increase dramatically from 0% to 50%. In short, it would be much better for her to be a victim of injustice, which seems odd. Note that this is not to deny the possibility of impersonal value. It is just to deny that impersonal value should play this role in a decision concerning whom to save or assist.

To put the point in more general terms (and in terms more closely related to historical injustice), assume that group A is unfairly disadvantaged, that it comprises a socially salient minority group (Lippert-Rasmussen, 2014), that A comprises the descendants of A* who were the victims of injustice inflicted by group Z*, the descendants of which, group Z, comprise the current majority group. Group B is also unfairly disadvantaged, B is also a socially salient minority group, and comprises the descendants of B* who were not victims of historical injustice at the hand of Z*. Together, A, B, and Z comprise society 1.

Compare A’s claims with B’s claims. B is also unfairly disadvantaged, but their disadvantage is not causally linked to any injustices committed by Z’s forebears, Z* (or Z, for that matter), and Z’s unfair advantage is similarly not linked to any such injustice toward B*. Suppose that B’s unfair disadvantage is caused by some natural disaster, such as an earthquake, a drought, or a flood.

In this simplified example, it can further be assumed that Z is the only group that is even potentially willing and able to meet A’s and B’s respective claims and that distributive justice requires some form of equality within societies such as 1. The main question, then, is whether there are morally relevant differences between A’s claim and B’s (alternative) claim. If claims from historical injustice have any force at all in virtue of being historical, then A’s claims must be more forceful than B’s claims.22 This seems implausible. Both groups are equally badly off and equally unfairly disadvantaged. They should thus both have the same claims based on justice.

Again, some people might accept a historical account of justice (Nozick, 1974), according to which rightfully held property can only change hands through voluntary transactions and any property that is wrongfully transferred should be given back to its rightful owner. On such views (barring non-identity, collective responsibility, supersession, and epistemic problems), A has claims on Z that B does not have. However, there are reasons to reject such theories of justice, not least because they over time (not least through inheritance) will lead to situations in which people are born with radically different life prospects for reasons that are completely out of their control. That seems unfair (and much more unfair than taxing the extremely rich in order to level the playing field for the young).

More generally, any society that allows some groups to be unfairly disadvantaged is for that reason unjust.23 And that is sufficient to conclude that redistribution is called for regardless of the historical roots of the unfairness. Similar claims, moreover, have been made by several theorists (Tan, 2007; Wenar, 2006).

Apologies and Symbolic Acts

All that has been said so far appears to undermine the plausibility of claims for compensation based on historical justice. This does not rule out that symbolic acts, such as public apologies from the majority to the minority for historical injustice can be plausible or required (Blomfield, 2021; Ivison, 2009). There is, of course, dispute over apologies issued by descendants of the perpetrators of injustice (or more typically a representative of the state that was at some point responsible for the injustice) toward the descendants of the victims. Since the injustice (insofar as it is merely historic and not [also] current) cannot have been committed by the agent who apologizes, and those to whom the apology is directed are not the actual victims of injustice, and because it is an apology typically issued by a representative of a collective toward another collective, there is a sense in which such apologies are fictions, and (merely) symbolic. However, symbols and fictions may have actual consequences, and it seems plausible that such apologies may sometimes work to improve strained relationships between groups, ease polarization, and so on. To the extent that this is the case, which of course is an empirical and contingent question (see Thompson, 2002, p. xv), such apologies may be both permitted and required.

Historical Injustice and the BPP

Claims of historical injustice appear to bear some resemblance to the BPP. According to BPP, innocent beneficiaries of injustice may acquire duties to compensate the victims of the injustice that they have benefited from (Butt, 2009; Goodin, 2013). Now, the descendants of perpetrators of historical injustice, for instance the descendants of slave owners, will often have benefited from that injustice, and there thus seems to be some connection between the two principles. On closer scrutiny, however, this is not the case.

The BPP, in most renditions, claims that agents that have benefited from an injustice have duties to compensate the victims of that injustice. Thus, although it covers historical injustice (to the extent that someone is still the victim of such injustices) it is not limited to historical injustice (which accounts of historical injustice arguably are). In addition, the BPP concerns only the victims and the beneficiaries of injustice, whereas accounts of historical injustice concern agents that are related to the original perpetrators and victims in different ways, including being their descendants. Furthermore, accounts of historical injustice most often (although see Butt, 2009) imply that being a beneficiary is neither necessary nor sufficient for gaining duties toward the victims of (historical) injustice.

The two principles might well overlap, because sometimes those who have benefited from historical injustice will be (otherwise) relevantly related to the perpetrators. For instance, descendants of slave owners will often have benefited from their forebears’ slave owning. Nevertheless, the two principles seem altogether distinct.

This comparison, however, might serve to suggest that BPP, if anything, seems more plausible than claims from historical injustice. After all, BPP requires that the beneficiary, that is, the agent with compensatory duties, actually benefits from the injustice. Claims from historical injustice do not (even though, as noted, this is sometimes implicitly suggested by the cases that are typically discussed). The former seems more plausible than the first (which is not to suggest that the BPP is ultimately justifiable; Knight, 2013; Huseby, 2015, 2016; Lippert-Rasmussen, 2017).


This article has assessed arguments for and against the normative relevance of historical injustice. The conclusion is that historical injustice as such does not seem normatively relevant and will not generally give rise to claims for compensation. In spite of this, some symbolic acts, such as public apologies, might be appropriate as a means to address a potential lack of trust and societal rifts that are caused by historical injustice. It should be emphasized that all victims of current injustices, many of whom are descendants of victims of historical injustices, should be compensated. Not because they are descendants of such victims, but because they are themselves victims of injustice.


  • Anderson, E. (1999). What is the point of equality? Ethics, 109(2), 287–337.
  • Arneson, R. J. (1989). Equality and equal opportunity for welfare. Philosophical Studies, 56(1), 77–93.
  • Axelsen, D., & Nielsen, L. (2015). Sufficiency as freedom from duress. Journal of Political Philosophy, 23(4), 406–426.
  • Blomfield, M. (2021). Reparations and egalitarianism. Ethical Theory and Moral Practice, 24(5), 1177–1195.
  • Boxill, B. (2022). Black reparations. In E. N. Zalta (Ed.), The Stanford Encyclopedia of Philosophy.
  • Butt, D. (2009). Rectifying international injustice—Principles of compensation and restitution between nations. Oxford University Press.
  • Butt, D. (2021). What structural injustice theory leaves out. Ethical Theory and Moral Practice, 24(5), 1161–1175.
  • Cohen, A. (2016). Corrective vs. distributive justice: The case of apologies. Ethical Theory and Moral Practice, 19, 663–677.
  • Cohen, A. I. (2009). Compensation for historic injustices: Completing the Boxill and Sher argument. Philosophy and Public Affairs, 37(1), 81–102.
  • Cohen, G. A. (1989). On the currency of egalitarian justice. Ethics, 99(4), 906–944.
  • Duus-Otterström, G., & Page, E. (2023). Defeating wrongdoing: Why victims of unjust harm should take priority over victims of bad luck. Social Theory and Practice, 49(1), 57–77.
  • Goodin, R. E. (2013). Disgorging the fruits of historical wrongdoing. American Political Science Review, 107(3), 478–491.
  • Holtug, N. (2010). Persons, interests, and justice. Oxford University Press.
  • Huseby, R. (2010). Sufficiency—Restated and defended. The Journal of Political Philosophy, 18, 178–197.
  • Huseby, R. (2015). Should the beneficiaries pay? Politics, Philosophy & Economics, 14, 209–225.
  • Huseby, R. (2016). The beneficiary pays principle and luck egalitarianism. Journal of Social Philosophy, 47, 332–349.
  • Ivison, D. (2009). Historical injustice. In J. S. Dryzek, B. Honig, & A. Philips (Eds.), The Oxford handbook of political theory (pp. 507–526). Oxford University Press.
  • Knight, C. (2013). Benefiting from injustice and brute luck. Social Theory and Practice, 39(4), 581–598.
  • Lippert-Rasmussen, K. (2014). Born free and equal? A philosophical inquiry into the nature of discrimination. Oxford University Press.
  • Lippert‐Rasmussen, K. (2017). Affirmative action, historical injustice, and the concept of beneficiaries. Journal of Political Philosophy, 25(1), 72–90.
  • Lippert-Rasmussen, K. (2018). Relational egalitarianism: Living as equals. Cambridge University Press.
  • List, C., & Pettit, P. (2011). Group agency: The possibility, design, and status of corporate agents. Oxford University Press.
  • McCarthy, T. (2004). Coming to terms with our past, part II: On the morality and politics of reparations for slavery. Political Theory, 30, 750–772.
  • Miller, D. (2001). Distributing responsibilities. Journal of Political Philosophy, 9(4), 453–471.
  • Miller, D. (2007). National responsibility and global justice. Oxford University Press.
  • Nozick, R. (1974). Anarchy, state, and utopia. Basic Books.
  • Nuti, A. (2019). Injustice and the reproduction of history: Structural inequalities, gender and redress. Cambridge University Press.
  • Parfit, D. (1984). Reasons and persons. Clarendon Press.
  • Perez, N. (2011). On compensation and return: Can the “continuing injustice argument” for compensating historical injustices justify compensation for such injustices or the return of property? Journal of Applied Philosophy, 28, 151–168.
  • Rawls, J. (1999). The law of peoples. Harvard University Press.
  • Roberts, M. (2022). The non-identity problem. Stanford Encyclopedia of Philosophy.
  • Sher, G. (2005). Transgenerational compensation. Philosophy and Public Affairs, 10(1), 3–17.
  • Simmons, A. J. (1995). Historical rights and fair shares. Law and Philosophy, 14(2), 149–184.
  • Singer, P. (2016). One world now: The ethics of globalization. Yale University Press.
  • Spinner-Halev, J. (2012). Historical injustice. In D. Estlund (Ed.), The Oxford handbook of political philosophy (pp. 319–335). Oxford University Press.
  • Tan, K. C. (2007). Colonialism, reparations and global justice. In J. Miller & R. Kumar (Eds.), Reparations: Interdisciplinary inquiries (pp. 208–306). Oxford University Press.
  • Thompson, J. (2001). Historical injustice and reparations: Justifying claims of descendants. Ethics, 112, 114–135.
  • Valls, A. (2018). Rethinking racial justice. Oxford University Press.
  • Waldron, J. (1992). Superseding historic injustice. Ethics, 103, 4–28.
  • Wenar, L. (2006). Reparations for the future. Journal of Social Philosophy, 37(3), 396–405.


  • 1. This is not to suggest that the passage of time is never relevant. If it can be imagined that people can live for a very long time, some injustices may have been committed so long ago that they could reasonably be labeled historical, even if both the victim and the perpetrator are still alive (see Thompson, 2001). Also, if an injustice was committed yesterday and the victim and perpetrator both die, the injustice might not reasonably be labeled historic. In addition, there could be cases in which only the perpetrator or the victim is dead. These different possibilities, however, fall outside the scope of the present article.

  • 2. In this article, the general term compensation is used, rather than reparation or restitution, which are sometimes used in the literature. This term is not, however, intended to cover symbolic acts, such as public apologies. See Boxill (2022) for a clarifying discussion of the relationship between these terms

  • 3. For further discussion of such views, see Blomfield (2021) and Nuti (2019).

  • 4. Note that “the injustice” is intended to cover the sometimes vast web of connected acts of injustice. Slavery and colonialism are, of course, not single acts of injustice.

  • 5. Some writers claim that not all instances of historical injustice are morally salient and offer reasons for why only some are. To the extent that they succeed, they avoid this particular problem.

  • 6. At least this seems to be the case. Arguably, however, such responsibility is not required in the context of claims from distributive justice. If distributive justice demands equality of resources, say, then A should distribute to B if A has more resources than B. Even though A in one sense has a moral responsibility to distribute the resources, A need not be held morally responsible for (having brought about) the unjust distribution in the first place. But corrective justice, of which claims from historical injustice is a subset, concerns wrongs. It is in the first instance the inflictors of wrongs that are the proper addressees of claims for compensation. Thus, it could be argued that to the extent that the descendants of the inflictors of wrongs should compensate in lieu of the original inflictors, this must be because the descendants also have some moral responsibility toward the descendants of the victims. Note, however, that there are other possible ways of grounding a duty to compensate, such as special relationships between groups (for instance, the descendants of the perpetrators and victims), or benefiting from injustice (Butt, 2009; Miller, 2001). Neither of these grounds presupposes that those who should compensate bear any moral responsibility for the original injustice or wrong in the first place. Thus, claims from historical injustice do not have to presuppose collective moral responsibility for the injustice. However, without such moral responsibility, it is more difficult to explain why some agents have (individual or collective) responsibility to compensate in the first place.

  • 7. An aspect of the epistemic problem might arise even in individual cases, however, to the extent that the case is known to be but one of countless other undocumented historical injustices.

  • 8. Of course, there were, as Boxill clearly states, many exceptions. But the general claim still holds, or so he argues. Note also that dissent, at some point was expressed, sufficient at least to abolish slavery.

  • 9. This duty is constrained, again in a Lockean fashion, by the duty holders’ right to subsistence.

  • 10. Note that Boxill is concerned only with establishing the claim, not to argue that it is unique.

  • 11. That leaves open the possibility that only Bert has compensatory duties toward Cynthia, but that is not Boxill’s argument. As he argues, “even if many descendants of the slave holders can be identified, proving that they inherited their property from their slave holding ancestors seems hopeless” (Boxill, 2022).

  • 12. Note that the intention here is not to claim that one particular view of collective agency is the correct one but, rather, suggest to what extent a given account of historical injustice is in line with the leading accounts of collective agency.

  • 13. Apart from that, this argument would avoid most of the noted problems, apart from (aspects) of the epistemic problem. In this respect, it is similar to Boxill’s inheritance argument.

  • 14. There are several other defenses of historical injustice, including Sher (2005), and Cohen (2009). It seems, however, that these, too, tend to rely on current harm or else fall prey to the non-identity problem. See Spinner-Halev (2012) for an overview.

  • 15. The former is arguably less clear.

  • 16. See, however, Simmons (1995). Note that historical accounts of justice, such as Nozick’s (1974) most likely entail claims from historical injustice. If changes in holdings do not flow from voluntary and permissible transactions, they are unjust and should be undone.

  • 17. However, clearly, resources, and welfare might be important to secure such relations in the first place.

  • 18. Luck egalitarianism can come in different forms, some of which will be incomplete. The same goes for some prioritarian accounts.

  • 19. With the exception of any yet unowned goods.

  • 20. Incomplete theories must also be pluralistic in a sense, in order to allow further justice claims. But to the extent that these theories have room for further claims without contradiction, pluralism is possible even without allowing conflict with the distributive theory. (This is not to deny the possibility that some claims from corrective justice will conflict with the demands from distributive justice, as suggested in the main text.)

  • 21. Note that if the wrongdoer is available, they might have particular duties toward A, because they are morally responsible for A’s plight but not for B’s. Bystanders, and even potential innocent beneficiaries, however, lack this morally salient connection with A.

  • 22. Assuming that A’s and B’s claims are otherwise equivalent, such that the potential bases of B’s claim are not especially forceful for some other reason.

  • 23. Note that unfair disadvantage is here intended in a loser sense than a straightforward luck egalitarian understanding.