Reparations, Recognition, and the Restoration of Relational Equality

II.A. Harm and Inheritance

In the philosophical literature, two grounds for reparations for historical injustice have been the subject of extensive discussion. According to harm views, descendants of victims of injustice are harmed today because of the past injustice and so are entitled to compensation for that harm.² According to inheritance views, persons have natural rights the violation of which grounds a claim-right to reparation and if such a right is not fulfilled, then the descendants of victims inherit it.³ There is no statute of limitations on natural rights, so reparative claim rights are inherited by descendants of victims in perpetuity until they are fulfilled.

Harm and inheritance accounts of reparations are morally individualistic; they are based in principles that govern reparative duties within small-scale interpersonal relations. Interpersonal moral requirements are standards that we can often hold persons accountable to through our institutions. Paradigmatically, the institution of criminal law is meant (among other things) to hold persons accountable for complying with their duties to others. So, given that we have interpersonal reparative duties on harm or inheritance grounds, it appears that we may use institutions to hold person accountable for fulfilling these duties.

As appealing as these views may be, I believe that relying on them for a justification of reparative law and policy for injustice in the relatively distant past faces insurmountable epistemic obstacles. The basic difficulty is that in order to determine whether a person is a reparative claimholder on either of these accounts we need a great deal of information that we lack and that in all likelihood we will never have.

These epistemic difficulties arise because, on harm and inheritance views, reparative claims are grounded in part in facts about individuals. This is a general feature of individualistic moral principles; their requirements are sensitive to the details of an individual’s life. On harm views, it must be assessed whether a candidate claimholder is harmed by virtue of past injustice, or by virtue of a more contemporaneous cause, or perhaps is worse off only because of her own decision-making.⁴ Only in the first case will a harm view ground a reparative claim. Similarly, on an inheritance view it must be assessed whether one’s ancestors might have renounced their reparative claims, whether they received some partial compensation, and whether the claims would have been squandered or wasted in such a way that prevents the reparative compensation from reaching later descendants.⁵

It is unlikely that members of a political community will ever be able to access the requisite information on either view. Consider inheritance first. Persons will have a claim to reparations on inheritance views only when they are the latest link in an unbroken chain of inheritance. Any break in the chain will preempt reparative claims from reaching contemporaries. To see whether the chain is broken, we must trace the lineage of each candidate reparative claim holder and assess whether they or any of their ancestors acted in ways that would undercut their reparative claim. This requires detailed biographical and historical knowledge about one’s family and each member of it.

Harm views are even more epistemically demanding; they require that we trace all causal contributions of past wrongdoing. Anyone harmed by virtue of past wrongdoing holds a claim to reparative compensation. This includes persons who are not members of the groups to whom past injustice was done. Suppose, for instance, that some group G was treated unjustly, and Alfie is the adolescent great-grandson of a person who suffered the initial injustice. Suppose further that because of that past injustice, Alfie is raised in social conditions that make him angry and impulsive. Alfie robs a store, harming the shopkeeper in the process. If suffering harm because of past injustice is the ground for holding reparative compensation, then the shopkeeper would have a claim to reparations because of the injustice done to G, and this would be the case regardless of whether the shopkeeper herself is a member of G.⁶ On harm views, determining whether one holds reparative claims will require tracing such complex causal channels back to past wrongdoing.

These epistemic difficulties will only be exacerbated when we take into account the dynamic nature of political society. In intervening years between past wrongdoing and today, the population will have changed through immigration and emigration, territory and subjects may have been added or lost, family lines will have ended or mixed, among other events that will influence whether reparative claims obtain on morally individualistic views.

Two objections to harm and inheritance views arise because of these epistemic limitations. First, morally individualistic views cannot be relied on to assess what kinds of reparative laws or policies we have reason to institute. In lieu of information about the chain of inheritance over generations or the causal channels by which persons alive today are harmed because of past injustice, the reasons provided by harm and inheritance views will be indeterminate with respect to how reparations should be institutionalized.

Is this an objection, or merely an observation of a general feature of law and policy in large-scale, complex societies? Incomplete possession of morally significant information is hardly unique to reparations. We face such epistemic limitations in almost all law and policy, and this does not undermine the passage of law or policy as such. If we may justifiably pass policy on such significant matters as economics, environment, or individual rights while being limited in our possession of morally significant information, why should we not be justified in relying on our incomplete but best available evidence regarding reparations?

I agree that incomplete morally relevant information does not undermine the justification of instituting law or policy as such. When it comes to reparations for injustice done in the relatively distant past, the difference is one of degree, rather than kind. We are profoundly ignorant of morally relevant information, and this makes acting on reasons provided by harm and inheritance views morally hazardous, by which I mean that acting on those reasons carries a significant likelihood of wrongdoing.

Specifically, we are liable to commit two kinds of wrongs. Both turn on the fact that by relying on harm or inheritance views under a severe deficit of information, we are likely to provide reparative compensation to many persons who have no claim on grounds of harm or inheritance, and we are likely to fail to provide compensation to those who do have such claims. The first kind of wrong is done to those who bear the burden of contributing resources for reparative law and policy. To be forced to give resources by the coercive apparatus of law and state when one has no duty to do so is wrongful extraction. The second kind of wrong is done to those who have genuine reparative claims on grounds of harm or inheritance but do not receive compensation. Being denied compensation as a legitimate claimant while others with no claim receive compensation is wrong. These facts weigh against the justificatory force of harm and inheritance reasons for instituting reparations. In general terms, I claim that if one lacks information that is pertinent to acting on some reason R, and insofar as that lack of information makes one liable to commit wrongdoing by acting on R, then R is more easily overridden in one’s deliberation in proportion to the number and severity of the wrongs that one is liable to commit out of ignorance. Our profound ignorance about information pertinent to harm or inheritance views therefore diminishes the justificatory force of those reasons significantly. While this does not weigh against the truth of harm or inheritance views, it greatly reduces their appeal as a practical guide to decision-making about reparative law and policy.

I turn now to the second objection based on epistemic limitation, which is that because of such limitation, harm and inheritance views fail to satisfy a necessary condition of interpersonal accountability, namely, publicity. Accountability is a reciprocal social practice in which we hold one another to common standards.⁷ Standards are public when persons with typical cognitive capacities could, through reasonable effort in normal moral contexts,⁸ identify the evidence needed for assessing whether the standard is satisfied.⁹ Publicity is a requirement of social practices of interpersonal accountability among persons conceived of as moral equals.¹⁰ The authority of an equal is reciprocal—I cannot view you as an equal while treating your demands as somehow privileged, as though you were a priest or sage with special insight into moral truth.¹¹ For there to be uptake of a demand in our accountability relations, the audience of the demand must be able to appreciate the reason for the demand. The epistemic obstacles to information pertinent to reparative obligations on morally individualistic views therefore make the demands of such obligations non-public.¹² Being non-public, they are not the kind of thing that we can hold one another accountable to. So, morally individualistic principles do not provide us with a justification for using our institutions to hold persons accountable for providing reparations; the reasons they offer are undercut because non-public.

Let me be clear that this challenge is not an objection to harm and inheritance views as such. The epistemic obstacles facing them is not a deep feature of all reparative contexts, so these views may provide justifications for reparations where such obstacles do not obtain, among which will be small-scale interpersonal contexts. My objections only cast doubt on relying on these views to guide practical decision-making when deliberating about reparative law and policy for injustice done in the relatively distant past. This challenges only the justificatory implications of reasons arising from harm or inheritance views as they concern the use of law and policy.

II.B. Institutionalization and Political Principles

Why does it matter that a theory of reparations provides a justification for reparative law and policy? Call a theory of reparations that does so “institutionalizable.”¹³ A theory of reparations that is not institutionalizable would fail to address an important part of political debate about reparations. Reparations are thought to be an important element of achieving justice for members of historically oppressed groups, and political debate about reparations concerns whether there is a justification for using our institutions to redistribute benefits to contemporary members of such groups in part because of past injustice. Institutionalizability matters because the fulfillment of reparations is a collective action problem the resolution of which depends on the use of institutions. The kinds of injustice that call for repair generally occur at large scale and so implicate a great number of persons as prospective claimholders and duty-bearers. Organizing the contribution and distribution of benefits at this scale requires resolving numerous coordination and cooperation problems, which is the task of political institutions. A theory of reparations that can address political debate about reparative law and policy must therefore be institutionalizable.

The obstacles facing the institutionalization of morally individualistic views gives us reason to consider whether there are other principles that are institutionalizable. I propose that the right kind of principle is political, or a principle that concerns the organization of shared social and political structures, as opposed to individualistic principles which concern the appropriate way to act with respect to other individuals. I believe that reparations are justified on grounds of relational equality. The basic thought behind any account of reparations is that persons alive today are in some significant way connected to injustice in the past. The question is in what way are persons alive today connected to the past injustice, and why is it that this connection grounds a claim to reparations. On the relational egalitarian theory of reparations, the connection is that persons alive today stand in social relations that are unjustly hierarchical at least in part because of past injustice.

The relational egalitarian view of reparations will be institutionalizable because its core principle, the anti-inferiority principle,¹⁴ applies, in the first instance, to social positions within a particular kind of social structure (hierarchy), and so applies to individuals by virtue of their occupying such positions. A familiar example of structural principles is John Rawls’s two principles of justice. To apply the two principles on Rawls’s view is to take “the position of certain representative individuals and consider how the social system looks to them.”¹⁵ We identify relevant positions by taking a macro view of the social structure and then considering the kinds of relations that systematically obtain and constrain the conduct of members of that structure.¹⁶ So, for example, Rawls’s difference principle applies not to the specific individuals who are worst off in a given society, but to the position of the least advantaged, understood as those persons, whoever they may be, who are, roughly, worst off with respect to income and wealth.¹⁷

On the relational egalitarian view, persons hold a claim to reparations by virtue of occupying the social position of public inferior. This ground for reparative claims is not concerned, in the first instance, with individuals, and so does not require knowledge about the histories of particular individuals or their family lines in order to assess whether they hold reparative claims. Since political principles take our social structures as the primary object of evaluation, we do not face the deep epistemic barriers that arise when our primary evaluative subject matter is individuals and all pertinent facts about them.¹⁸

In addition to avoiding the difficulties facing harm and inheritance views discussed above, the political character of the relational egalitarian view carries implications concerning important details about reparative obligations. Such egalitarian reparative obligations are owed to persons in part because of their occupying the social position of inferior, rather than because of facts about their family line or the harms caused by past injustice. The latter considerations do not influence whether relational egalitarian reparative claims obtain. This results in two general categories of persons being included among reparative claim holders on the relational egalitarian view that are often excluded on individualistic views. First, members of groups that were subject to past injustice who are not descendants of victims of that injustice can come to hold reparative claims. For example, a recent migrant to the United States who is regarded as a social inferior in virtue of being Black could, on this view, hold a claim to reparations for such things as slavery or Jim Crow, despite her ancestors not suffering those injustices. Second, members of groups not defined by shared lineage can hold reparative claims because of past injustice done to members of their group. This means that members of groups such as women or the working class are candidate reparative claim holders.

Readers with intuitions that lead them to find harm or inheritance views appealing will likely find these implications puzzling, if not outright objectionable. I submit that when dealing with injustice of relatively greater temporal and social scale (compared to small-scale interpersonal interaction), these intuitions will be misleading. Reckoning with injustice in the relatively distant past requires that we turn our attention to structural social phenomena consideration of which will lead to judgments that do not align perfectly with judgments arrived at based on morally individualistic considerations.¹⁹ Those considerations become intractable when we turn to a greater social and temporal scale. I do not deny that morally individualistic principles such as those that ground harm or inheritance views are intuitively appealing, nor do I challenge their ability to serve as an account of reparations in small-scale interpersonal contexts. I maintain only that when dealing with injustice in the relatively distant past, our practical reasoning on these matters will be best served by relying on political principles whose primary object are social and political structures, rather than individuals. Reasoning on such political considerations will not fully accord with intuitions that are attuned to small-scale interpersonal settings.

These are only preliminaries that concern the form of the relational egalitarian view. To better understand its reparative implications, we must turn to substantive matters. Before proceeding, let me address what makes for a successful theory of reparations.

IV.A. Inferiority

According to the

Anti-inferiority principle: it is unjust for members of society to be publicly regarded as inferiors.

I appeal to the anti-inferiority principle for two reasons. First, it is based on a claim shared among relational egalitarian views, namely, that public distinctions of inferior or superior social worth among persons and accompanying disparities of treatment are a paradigm violation of relational equality.²² Secondly, whereas relational egalitarians view inferiority as intrinsically unjust, the anti-inferiority principle allows for the injustice of inferiority to be extrinsic. Public inferiority may be found objectionable because of its harmful or unfair effects. Decades of empirical work in sociology have revealed that “status beliefs create inequalities in assertive versus deferential behavior, actual task performance, attributions of ability, influence, and situational rewards between otherwise equal men and women, whites and non-whites, and middle-class and working-class people.”²³ Allowing for public inferiority to be extrinsically objectional makes the relational egalitarian view of reparations more appealing, as it can be relied on by views that do not share fundamental moral commitments with relational egalitarianism.

The anti-inferiority principle has been a source of justification for the institutionalization of morally important legal reform. A prominent example is the American Supreme Court’s reasoning for undoing the racial hierarchy imposed by Jim Crow laws. Those laws were notoriously defended in Plessy v. Ferguson, where the “separate but equal” doctrine was established, and even as of that ruling, the anti-inferiority principle was raised as a source of criticism. Justice John Marshall Harlan, in his dissenting opinion, contended that the

arbitrary separation of citizens on the basis of race while they are on a public highway is a badge of servitude wholly inconsistent with the civil freedom and the equality before the law established by the Constitution.²⁴

Justice Harlan’s opinion proved prescient of the eventual ruling made in Brown v. Board of Education of Topeka, where the unanimous opinion held that

Considerations [of social inferiority] apply with added force to children in grade and high schools. To separate them from others of similar age and qualifications solely because of their race generates a feeling of inferiority as to their status in the community that may affect their hearts and minds in a way unlikely ever to be undone.²⁵

The relational egalitarian view of reparations maintains that reasoning on these grounds will also justify reparative law and policy.

Among the grounds of the injustice of the racial hierarchy imposed by Jim Crow is that certain persons were publicly regarded as inferiors. To be a public social inferior is to occupy a lesser position in a social hierarchy of which most members of a given society are also a part. By “social hierarchy,” I mean a particular kind of social phenomenon, namely, a stratified social structure. A social structure consists of interconnected relations among a set of persons and institutions, with a paradigmatic constituent being social norms.²⁶ Positions are isomorphic sets of relations within a structure. So, for example, if person A is my boss, and person B yours, then you and I occupy the same kind of position, namely, that of professional subordinate within the structure of the workplace. A social structure is stratified when individuals within it come to widely share beliefs and attitudes about the relative value of persons within the structure.²⁷ Here, “valuing” is used in the descriptive sense to refer to interconnected positive attitudes and beliefs that persons hold about some entity. Sociologists refer to the phenomenon of relative valuing of persons as social status, and I follow that use here. “[S]ocial status” in other words, is “the prestige, respect, and esteem that a party has in the eyes of others…[s]tatus is an index of the social worth that observers ascribe to an individual.”²⁸ The basis on which some individual is seen as having greater worth than another is a function of the context of the social structure in question. A wide range of social structures are hierarchical, with professional athletes being seen as having greater or lesser worth within a sport on a different ground compared to the basis on which a philosopher is seen as having greater or lesser worth within academic philosophy. A hierarchy is public when most members of a given society occupy a position in it. The distinctive features of public hierarchies are that, first, the characteristics of individuals that make them more or less valuable in the eyes of others are things that most members of society can discern and identify, and second, most members of society share the beliefs and attitudes in virtue of which some persons are seen as having greater or lesser worth. Common examples of characteristics on the basis of which such widely shared beliefs and attitudes are formed are such things as race, gender, education, income, and wealth, among others. Public hierarchy is distinctive, then, in that it is not bound to a specific context, and the features in virtue of which persons are evaluated as being better or worse members of society are not based in competence or performance in specific tasks or skills, such as the quality of one’s tennis game or how convincing one’s arguments are. Bringing this together, we have:

Public inferiority: some person A is a public social inferior in society S when members of S converge on stable, shared beliefs and attitudes which hold A to have lesser value than some significant share of S.

Two further points are worth clarifying. First, the condition of stability is meant to capture that in order for hierarchy to organize social life over time, the evaluations and attitudes of persons that constitute it must be in equilibrium.²⁹ Second, the open-ended term “some significant share” captures that there are varying extents to which a person can be a social inferior, where the extent to which one is an inferior is proportionate to the number of people who are represented as being more valuable than one.

The role that hierarchy and public inferiority play in the relational egalitarian theory of reparations is that of explaining how contemporary social life is connected to injustice in the past such that there are duties of reparation. That connection obtains when past injustice brought about, exacerbated, or maintained public social status hierarchies which persist to the present day.

Importantly, public hierarchies can exist and persist in light of changes to legal and political institutions. While legal and political institutions have historically contributed to the stability of public social hierarchy, they are not themselves essential constituents of public inferiority. We can imagine, for instance, a society that gave equal political and legal status to all citizens, say by instituting democracy with rule of law, while it still being the case that citizens maintained the status relations that constitute public inferiority.

This is precisely the situation of contemporary India. Though caste discrimination was legally banned as of India’s independence in the middle 20^th century, and though several decades of affirmative action policy for members of lower castes have been in place, caste continues to be a significant influence on Indians’ lives. There remain pervasive inequalities in access to labor markets, choice of marriage partner, rates of child mortality, access to education, and income that obtain between descendants of different castes because of that difference in caste.³⁰ This shows that public inferiority can outlive the legal and political institutions that may at one time have been an important source of support for, if not outright cause of it. This detail is important, as it explains why reparative duties can obtain despite significant changes in legal and political institutions.

IV.B. Recognition and Assurance

In addition to the anti-inferiority principle, the idea of recognition respect, and assurance of recognition in particular, is central to the relational egalitarian view of reparations.

To treat something with recognition respect is to view it as a source of sui generis constraint on one’s practical reasoning.³¹ The kind of recognition respect that matters here is recognition respect for persons. To accord a person recognition respect is to view that person as having equal fundamental moral standing, which involves recognition of the person as counting, morally speaking, for their own sake, taking this fact into account in one’s deliberation, and constraining one’s conduct accordingly.

Recognition respect is related to the anti-inferiority principle in that it is taken to be the ground of relational egalitarian principles. Many accounts of relational equality share the foundational claim that “relations of social inferiority and superiority are inappropriate among moral equals.”³² To say that persons are moral equals is to make a claim about status, but of a kind that is fundamentally different from what I called social status above. “Status” in general refers to the relative worth of entities on the basis of some standard. Having equal moral status means that, ceteris paribus, persons count the same in terms of basic moral considerations (whether this is understood in terms of their interests, desires, rights, or whatever moral notions are taken to be basic). As discussed above, social status refers to a kind of social fact, rather than a normative one. Whereas moral equality is a fact inherent in persons, social equality is a contingent circumstance to be brought about which obtains when there is widespread convergence on those beliefs and attitudes that constitute valuing (in the descriptive sense) such that all members of a given society are equally valued qua members of society. The contingent nature of social equality makes recognition of one as a social equal distinct from recognition of one’s moral equality. Firstly, I view you as a public social equal just when I have those beliefs and attitudes in virtue of which I value you as a member of our shared social world to the same extent as I value myself as a member of the social world. Secondly, I can recognize your social status by assessing what the convergent evaluation of you in comparison to others in our society is, which, if you are valued equally relative to me, is recognition of our social equality. To bring these different notions of recognition together, I can, for example, appreciate that an unhoused person is my moral equal, value her equally as a member of society, believe that others ought to recognize her as an equally valuable member of society, but appreciate that other members of society do not do so, and that she is in fact valued less than most other members of society. Details about how we can bring about those beliefs and attitudes in virtue of which persons would be widely valued as equals will be taken up at the start of the next section. For now, I return to the importance of recognition respect (for moral equals), and the value of its assurance.

There are derivative moral considerations that bear on the goodness or badness of practices of recognition respect, by which I mean the pattern of actual experiences and expressions of recognition respect in a given society. One important consideration is publicity, which holds that individuals ought to be accorded recognition respect in such a way that they are in a position to appreciate the fact that they are recognized as a moral equal.³³ In addition to being a condition of interpersonal accountability relations, as discussed in Section II.A, publicity is also a good-making feature of practices of recognition respect.

A practice of recognition respect is better in one respect when it satisfies publicity because publicity is instrumental to providing persons with assurance of their recognition as moral equals, and such assurance is instrumentally valuable for securing stable cooperative social relations.³⁴ In the absence of assurance of recognition, beneficial cooperative social relations can fray. When a person lacks assurance of their recognition as a moral equal, she may come to worry that she is exposed to some risk of wrongdoing. Others may neglect her moral equality and engage in opportunistic wrongdoing for their own advantage. If a person has reason to worry about such risk, she may withdraw from social relations or keep her engagement limited. She might even engage in opportunistic or preemptive defensive wrongdoing herself. When a practice of recognition respect satisfies publicity, it will be a source of assurance of one’s recognition as a moral equal, and so a source of assurance that one will not be taken advantage of or made to suffer opportunistic wrongdoing.

Importantly, assurance offered to moral equals must be addressed to their reasoning. One’s need for assurance arises by virtue of a reason that one has to believe that others will act in ways that may result in threat or insecurity to something that one finds important. To appropriately provide assurance to one who needs it, you must address the reason by virtue of which one takes oneself to be insecure.³⁵ The provision of assurance should occur in such a way that one is able to appreciate new reasons that outweigh or silence the reasons by virtue of which one initially took oneself to be insecure.

Since the provision of assurance requires addressing the reasoning of those who need it, there are ways of removing the threat or insecurity against which assurance is needed that will fall short of an appropriate provision of assurance. Namely, one might compensate for the threat in question without addressing the reason by virtue of which the insecure party needs assurance. If I were to break a promise that I made to you, it would not suffice if I simply opened my wallet and asked what the damages were. Breaking a promise is not merely a material inconvenience that warrants some kind of compensation; it is a failure to abide by an especially weighty reason that making a promise establishes. My promises to you are not tentative appointments that I need only to pay the right price to cancel. If I treated my promises to you in this way, you would have reason to doubt that I respect your status as promisee. My giving you cash compensation, even if it is equal to the material cost that you took on in planning around my unkept promise, will fail to address the reason that you have to doubt that I will recognize the obligation that I had to you. If the good of assurance of recognition as an equal is to obtain, then assurance must be provided in a way that addresses the reason by virtue of which assurance is needed. In this case, sincere recognition of the wrong in the form of apology is owed.

In sum, practices of recognition respect are better in one respect when they satisfy publicity by virtue of the assurance of recognition as a moral equal that such publicity provides. The relevant kind of assurance will only obtain when public expressions of recognition respect are made in a way that addresses the reasons that are the source of the need for assurance. With the ideas of the anti-inferiority principle and assurance of recognition in hand, I now turn to the defense of the relational egalitarian view of reparations.

V.A. Reforming Status Relations

The relational egalitarian theory of reparations, recall, holds that 1) reparations are owed to persons who are public social inferiors in part because they are members of a group that has been subject to injustice in the past, and 2) reparations are to be such that a) they ameliorate and undo positions of public inferiority and b) members of the relevant group are assured of their recognition as moral equals.

A basic assumption of this view is that there are some means for reforming status relations and positions of public inferiority. The prevailing theory of social status in sociology is status construction theory, according to which some feature of persons becomes a status characteristic when it is correlated with things that are perceived as valuable.³⁶ When persons in some group G with some characteristic C, a particular skin color say, tend to have a greater share of some good, then members of G will come to conceive of C as itself being a good-making property of individuals. The value attached to the greater possession of goods “diffuses” and confers value on other features of persons who possess those goods.³⁷ This suggests that there is at least one way to transform status relations to be more equal, which is to provide greater access to things widely recognized as goods to those with low status such that the characteristics by virtue of which persons are appraised as being low status are no longer correlated with possession of a lesser share of goods.³⁸ So, throughout this section I will speak of reparative claims in terms of entitlements to some kinds of goods, though those entitlements are held on instrumental grounds because the receipt of goods by the relevant persons contributes to undoing positions of public inferiority.

The benefits owed are those that are most effective at providing, in an enduring way, the various advantages that are viewed as sources of social status and security against unjust treatment. This suggests two general claims about the kind of reparative benefits that the relational egalitarian account gives us reason to provide. First, guarantees of political power will be an appealing means of reparation. The specific form of such guarantees should be sensitive to the particular political institutions in a given society, some examples of which would be guaranteed political representatives, or special legislative committees, among other arrangements. Political power is both a means for and source of greater public social status and is a safeguard against unjust treatment such that it is especially effective as a means of assurance of recognition. Secondly, it is doubtful that reparations should be provided as one-off cash transfers. While resources are status-conferring in most every human social group, the problem with cash-transfer policy is that the contribution to ameliorating public inferiority will run out once the money is spent. More enduring sources of status may instead be found in conferral of benefits such as education, e.g., affirmative action policies, or opportunities for the cultivation of skilled labor and capital, e.g., business subsidies.³⁹

V.B. Burden Justification

The reason why persons alive today who are not responsible for past wrongdoing ought to bear the burden of providing reparations is because the anti-inferiority principle is a principle of justice and persons have a duty to do their fair share in promoting justice.⁴⁰ When sufficiently just political institutions exist, this duty takes the form of a duty to comply with the demands of those institutions at least when they promote justice. Accordingly, citizens have a duty to do their fair share in supporting the institutionalization of laws and policies that ameliorate and undo positions of public inferiority, which involves taking on a fair share of the costs of those laws and policies.

This account has the following implications for the distribution of reparative compensation and the burdens of providing that compensation. First, relational egalitarian reparative claims are owed to the relatively least advantaged among relevant groups, which, as discussed in Section II.B, need not be defined by any shared lineage. To see why the relatively least advantaged among such groups are owed compensation, consider the position of an especially advantaged member of a relevant kind of group. Given that this person already occupies a relatively advantageous social position, conferral of additional benefits will do little to change social status relations, whereas the conferral of benefits to the relatively disadvantaged member of the relevant group will make a greater contribution to undoing the social conditions that bring about positions of public inferiority. Since the content of reparative duties on this account is to undo relations of public inferiority (rather than provide benefits as such), the reparative claim of a relatively advantaged member of a relevant group is thus a claim to the provision of benefits for relatively disadvantaged members of that group.

Second, the burden of providing reparations should not be shared with members of the group to whom reparations are owed for two reasons. First, given that possession of wealth is, in just about every human society, a contributing factor of one’s public social status, requiring contributions from members of the groups to whom reparations are owed will often be self-defeating. The second reason turns on a general claim about duties to promote justice, which is that victims of injustice do not, in the first instance, bear the duty to undo the injustice from which they suffer. There can, however, be circumstances where the most advantaged members of groups that were subject to past injustice can come to have an obligation to contribute to reparations. This will be the case when, i) the individuals in question have such a great relative share of wealth and resources that making a contribution to reparations does not setback the undoing of relations of public inferiority that members of this group stand in, and ii) other members of the society who bear the duty to undo the positions of public inferiority in the first instance either neglect their duty or lack the means to provide adequate reparations.

Third, membership in historically dominant groups is not as such a basis for bearing a greater share of the burdens of providing reparations. There may be contingent reason to distribute a greater share of the burdens to members of such groups—for example, if members of historically dominant groups tend to have greater shares of resources, then their fair share of the costs of advancing justice may be greater for that reason. Moreover, insofar as members of those groups come to hold goods with particular symbolic connections to past injustice, they may have obligations to contribute or surrender those goods as part of assurance-conferring expressions of recognition. An inheritor of a plantation in the Southern United States, or an owner of historical artifacts used for enslavement, may on this basis, be required to relinquish or sell that property.

V.C. Sensitivity to Wrongdoing

The ease with which the relational egalitarian view satisfies burden justification appears to come at the steep cost of satisfying sensitivity to wrongdoing. Persons, on this view, are taken to have an obligation to bear the burden of providing reparations because it is a requirement of justice, and the provision of bare justice is not reparation.

I deny that the relational egalitarian view is unable to satisfy sensitivity to wrongdoing on grounds of the following argument:

1. When facts about past injustice to some group G are salient to contemporary members of G, members of G will have reason to regard their recognition as insecure.

2. Such insecurity warrants assurance, and assurance addressed to moral equals must speak to the reason by virtue of which such assurance is needed.

3. Two relevant reasons held by members of G are: a) a reason to distrust political institutions, and b) a reason to anticipate being, and continuing to be, neglected as a subject of justice.

4. To appropriately address (a), political institutions will need to make public expressions of recognition of the moral status of members of G.

5. Such public expressions are at risk of being cheap talk—they must be made credible as well.

6. Cheap talk is made credible by having the party making the expression act, at some cost to itself, on those considerations that warrant the expression in question.

7. The relevant considerations are (b), such that the sincere expression of recognition must be accompanied by acting to promote justice with respect to contemporary members of G.

The general idea behind this argument is that when contemporary members of G have reason to believe that their recognition as moral equals is insecure, public assurance of recognition must be provided to them and sensitivity to wrongdoing will obtain for this reason.

Premise 1 introduces the notion of salience to explain how contemporary members of G can come to be insecure of their recognition because of past injustice. Salience is a complex idea that concerns the numerous ways in which facts about the past of some society influence the way in which contemporary members of that society act.⁴¹ Attempting to provide a comprehensive account of how past injustices, or facts about a society’s history more generally, are salient in ongoing social relationships would take the present discussion too far off course. As such, I hope to explicate the idea by examining a series of examples.

First, facts about past injustice may be part of the best available explanations that persons rely on to understand why it is that members of some group are regularly disadvantaged. Many explanations of the relative disadvantage of Black Americans, for example, draw on evidence about past discrimination in order to explain disadvantages in housing, wealth, education, and health.⁴² Second, expressions and symbols that have the effect of humiliating or demeaning members of groups that were subject to past injustice (e.g., slurs) often have that effect because the meaning of such expressions or symbols is informed by facts about the relevant society’s past. Imagine a school principal who, solely for reasons of aesthetic preference, organizes a school photo such that Black students sit on the left and white students on the right.⁴³ Even if the intention of the principal is innocent, within the context of the United States and its history of segregation, her behavior is demeaning. Third, facts of past injustice can be salient because they are relied on as a hermeneutical tool by those who suffer contemporary injustice or those who seek to understand it. Injustice sometimes has the property of obscuring the fact that there is injustice. In such cases, the disadvantaged are subject to a hermeneutic injustice,⁴⁴ and drawing on facts about the past injustice can be a valuable tool for undoing that hermeneutic injustice. For example, prison reformers and prison abolitionists in the United States rely on the history of slavery to understand the ways in which Black persons are oppressed by prison practices.⁴⁵ In sum, we can see the salience of facts about past injustice at least when such facts are relied on to explain disadvantage to some groups, when such facts explain why certain forms of expression are demeaning or humiliating, or when such facts are relied on as a hermeneutic resource for understanding injustice.

Premise 2 states the requirements of appropriate provision of assurance discussed in Section IV.B. Premise 3 identifies two reasons that contemporary members of G may have for believing that their moral equality is not recognized, which are: a) a reason to distrust their political institutions, and b) a reason to expect being neglected as a subject of justice. These reasons will generally but not necessarily follow from the fact that there has been injustice in the past that remains salient. They do not necessarily follow because it is possible that despite the salience of past injustice to G, contemporary members of G are in a position where reasons that undercut or preempt their reason to be insecure obtain. This would occur when, for example, members of G have control of government, or access to a significant share of political power. So, there may be countervailing reasons that make moot the need for assurance on the part of G, in which case the claim to reparations will be weakened if not defeated (claims of bare justice may persist). When such countervailing reasons do not obtain, members of G have reasons to distrust their political institutions and anticipate being treated unjustly that are not defeated or undercut. (a) obtains because political institutions are the principal means of coordinating and effecting large-scale social structural change through collective action, such that occupants of offices within political institutions take on responsibilities of justice. These responsibilities include duties not to commit injustice against subjects and duties to protect subjects from injustice. Political institutions may thereby have been implicated in historical injustice when they were directly relied on to perpetrate the injustice (e.g., through legal enforcement) or when they tolerated pervasive wrongdoing against some subjects while possessing sufficient capacity to protect against such wrongdoing. When these institutional responsibilities are breached, contemporary members of G will have reason to distrust their political institutions.⁴⁶ (b) obtains because when past injustice to one’s group is salient, and when one’s group is subject to ongoing injustice, one has reason to believe that injustices against members of this group are at least regularly tolerated, if not deliberately inflicted.

Premise 4 gives the appropriate way to address reason (a). Part of the insecurity of members of G arises because of their reasons to distrust political institutions. Accordingly, recognition must be publicly expressed by their political institutions in order to suitably address the reasons by virtue of which assurance of recognition is needed.⁴⁷ The content of such expressions can include: the fact that the compensation in question is warranted because of the past injustice; the fact that the occurrence of the past injustice is regrettable and worthy of public commemoration; or (inclusively) the fact that perpetrators of the injustice are contemptable and dishonorable. Assurance-conferring public expressions of recognition can take the form of public expression of regret that accompanies reparative compensation, or it may come in the form of monuments or memorials that commemorate the victims of past injustice, among other modes of expression.⁴⁸

There is an important problem facing public expressions of recognition by political institutions, which is stated in premise 5. In order for one’s expression to be assurance-conferring, it must be seen as credible by the intended recipient. Given that members of the relevant kinds of groups have reason to doubt that their moral equality is recognized, they will have reason to doubt the sincerity of public expressions of recognition. There is a risk that such expressions are taken to be cheap talk.

Problems of cheap talk are resolved by making the relevant signal costly. We arrive at premise 6, which states that expressions of recognition are shown to be credible when the party making the expression shows itself to be acting on the relevant considerations at some cost to itself, rather than for the sake of mere appearances. However, our concern here is not with mere credibility, but with credibility for the right reasons. The insecurity of recognition faced by some persons is not a mere inconvenience for which society must pay a price. For an assurance-conferring expression to be credible for the right reasons, it must address the reasons by virtue of which assurance is needed in the first place. Assurance is needed because of i) a salient history of injustice against G, and ii) the fact that the contemporary members of G continue to occupy unjust positions of public social inferiority. It is the failure to abide by requirements of justice in the past (assuming salience) and the failure to achieve justice in the social relations that persist into the present that provide contemporary members of G with reason to be insecure about being recognized as moral equals. Sincerity of expression of recognition of moral equality is to be achieved, then, by having the party making the expression act, at some cost to itself, on considerations of justice with respect to those to whom the expression of recognition is meant to provide assurance. Those considerations of justice are given by the anti-inferiority principle. Therefore, the provision of various benefits meant to undo positions of public inferiority are related to assurance-conferring public expressions of recognition in that the costs of undertaking the former serve to make the expression of assurance sincere and therefore credible for the right reasons.

We can see now that the bare justice test is satisfied. Suppose past members of group G were subject to past injustice that remains salient, and contemporary members of G are public social inferiors. The public inferiority of G grounds claims to benefits that would undo that inferiority, and their insecurity of recognition grounds a claim to assurance of recognition, which needs to be rendered credible for the right reasons by taking on the costs of undoing the inferiority of G. Contemporary members of G are therefore owed inferiority-undoing compensation in tandem with assurance-conferring expressions of recognition. Necessarily, discharging this duty will require that the past injustice is appropriately acknowledged. Compare the counterfactual case of G*, which is the same in all respects to G except that G* has not been subject to past injustice. There are no salient historical facts to ground a claim to assurance-conferring recognition on the part of G*. In this case, members of G* have claims that their inferiority be undone and discharging the correlative duties does not necessarily require acknowledging past injustice. What was a reparative claim becomes a claim of bare justice in the absence of past injustice.⁴⁹ Hence, the relational egalitarian theory of reparations passes the bare justice test and therefore provides a ground for genuine reparative claims.

V.D. Relational Egalitarian Reparative Justice

In sum, relational egalitarian reparative claims obtain if, and only if, 1) there is some group G that has been subject to past injustice that remains salient, 2) contemporary members of G occupy positions of public inferiority, and 3) members of G have reason to distrust their political institutions as a consequence of (1) or (2) or both. In virtue of (1), members of G are owed assurance of recognition as moral equals; in virtue of (2), members of G are owed inferiority-undoing compensation; and, in virtue of (3), discharging duties to undo the inferiority of G serves as a cost that vindicates the credibility of assurance-conferring expressions of recognition for the right reasons. For the sake of illustration, let us consider some cases where the reparative demands of relational equality are unsuccessfully fulfilled or do not obtain.⁵⁰

On the relational egalitarian view, the provision of reparative compensation to G serves two purposes: i) it undoes the public inferiority of members of G and ii) it serves as a cost that makes credible for the right reasons assurance-conferring expressions of recognition to G. Compensation that does neither (i) nor (ii) does not contribute to discharging relational egalitarian reparative obligations in any respect. For example, if the U.S. government provided cash payments to Black citizens (assuming they are owed reparations), but those payments were not accompanied by assurance-conferring expressions of recognition and the payments themselves did not ameliorate or undo the public inferiority of Black citizens because, say, of widespread recalcitrant racist attitudes and beliefs, then those cash payments would not contribute to discharging reparative duties to Black citizens whatsoever.⁵¹ This case illustrates how acts which fail to discharge the duties arising from conditions (1) and (2) above thereby fail to contribute to fulfilling relational egalitarian reparative duties.

The general idea underlying (3) is that there must be an appropriate relationship between the duty to undo relations of inferiority and the duty to provide assurance of recognition in order for a relational egalitarian reparative duty to obtain. Not just any conjunction of these two duties constitutes a reparative claim—the duties must be borne by the same party, and need for assurance of recognition must stem from having suffered, and continuing to suffer, injustice in which the party making the expression of assurance is implicated, such that discharging one’s duties of justice grounded by the anti-inferiority principle renders expressions of assurance credible for the right reason. Consider a case of a group the members of which are both in need of assurance of recognition and regarded as public inferiors, but not for any reason to do with past injustice. Imagine that the unemployed are seen as public inferiors and that they are also disproportionately vulnerable to wrongdoing. This case appears to be a counterexample to the relational egalitarian view. It seems that the unemployed are owed benefits that will undo their inferiority in addition to assurance of their recognition, which means that they qualify for reparations on the relational egalitarian account. But this case seems to only be a matter of contemporary bare justice, which suggests that the present account is not distinctly about reparations.

Whether reparative obligations arise in this case will depend on political institutions’ relationship to the injustice, and where those obligations do obtain the case is no counterexample. Let us first set aside one set of reparative duties that will obtain in a contemporaneous case, namely, those owed by perpetrators of wrongdoing to their victims. As mentioned throughout, the relational egalitarian account is not exhaustive of the grounds for reparative obligations, and in cases of small-scale interpersonal wrongdoing, reparations may obtain on other grounds.

The key question is: were political institutions perpetrators of, or complicit in, injustice suffered by the unemployed? If the answer is no, then the unemployed have no reason to distrust their political institutions because of this injustice, from which it follows that those institutions have no duty to provide assurance-conferring expressions of recognition to the unemployed.⁵² Certainly, the unemployed are owed assurance of recognition from perpetrators of wrongdoing against them, for the same general reason that I owe you such assurance when I break a promise to you. Moreover, political institutions do bear an obligation of bare justice to undo the relations of public inferiority that the unemployed are subject to. But political institutions fulfilling their duty of justice cannot serve as a costly signal to vindicate the assurance-conferring expressions of recognition that perpetrators of wrongdoing owe to the unemployed. Consider a comparison to an interpersonal case. If I break my promise to you, such that you need assurance of recognition from me, and independently Alfie owed you some compensation (say, he bought something from you), this will not entail that you have a claim to reparations, for the two claims are not related in the right way.

If, on the other hand, political institutions were implicated in the wrongdoing, say by being delinquent in its duties of justice to the unemployed, then those institutions would owe assurance of recognition in tandem with duties to undo public inferiority. In this case, however, I do not think it is objectionable to say that there is a duty to provide reparations. As discussed in Section III, fundamental to reparative duties is a temporal orientation, not a temporal magnitude. The relational egalitarian view is particularly appealing because it provides a sound basis for instituting reparative law and policy for injustice in the relatively distant past, but this does not confine the view to only cases of relatively great temporal magnitudes. There may still be occasions where reparations on grounds of relational equality obtain for relatively recent injustice. Specifically, reparations for recent injustice will be justified on relational egalitarian grounds in just those cases where political institutions are at least complicit in, if not an outright perpetrator of, the enforcement and maintenance of relations of public inferiority. When the United States enforced the enslavement of Black persons, it was a perpetrator of the injustice of racial caste, and as a result of perpetrating that injustice owed reparations to its victims. It is the endurance of public hierarchy over time that connects those of us alive today to past injustice so as to ground reparative duties on the relational egalitarian account, and it does not follow from this that the view cannot ever justify reparations for relatively recent injustice. So, where the political institutions are at least complicit in the wrongdoing in virtue of which members of a group who are public inferiors need assurance, then reparations will obtain on grounds of relational equality, even for relatively recent wrongdoing.⁵³