Corrective vs. Distributive Justice: the Case of Apologies

Abstract

This paper considers the relation of corrective to distributive justice. I discuss the shortfalls of one sort of account that holds these are independent domains of justice. To support a more modest claim that these are sometimes independent domains of justice, I focus instead on the case of apologies. Apologies are sometimes among the measures specified by corrective justice. I argue that the sorts of injustices that apologies can help to correct need not always be departures from ideals specified by distributive justice. Apologies and the moral relations they engage might thus be parts of a domain of justice that is neither distributive nor dependent on distributive justice.

1 Introduction

Corrective justice tells us, among other things, what justice permits or requires when someone has been denied a good that was her due. Broadly speaking, distributive justice tells us how and why people in some group may have certain benefits and responsibilities regarding various divisible goods.¹ How should we understand the relationship of corrective to distributive justice?

I approach this question by examining some of the claims and the reasons for action corrective justice offers, and comparing them to those of distributive justice. Exploring the structure, content, and scope of the reasons and claims of corrective justice can help us understand what role, if any, distributive justice has in justifying or limiting reparative measures. The results of such an inquiry should illuminate the role of corrective justice in a wider theory of justice.

One family of accounts holds that corrective justice is subordinate to distributive justice.² On this sort of view, corrective justice has nothing to say unless distributive justice first identifies what belongs to whom. Indeed, on some accounts, principles of reparation have no unique work to do except perhaps that of specification or to provide some rhetorical advantage in favor of repairing some maldistribution.³ Proponents of such views hold that remedies under corrective justice are justifiable only as devices to bring about some distributive improvement or to provide some remedy specifiable by distributive principles. From the standpoint of justification, however, corrective justice is then “parasitic,” “ancillary,” overdetermined by, or otherwise wholly dependent on distributive justice.⁴

One way to criticize the view that corrective justice is subordinate to distributive justice is to defend a position that I call the “autonomy thesis”. The autonomy thesis holds that corrective or reparative justice is sometimes not dependent on distributive justice.⁵ On this view, corrective justice has a different structure, function, or justification than distributive justice. Reparative principles may then uniquely establish some claims and duties.

In what follows I discuss the shortfalls of one sort of defense of an autonomy thesis. I then focus on the case of apologies to indicate how some principles of corrective justice need not be secondary to distributive justice. I argue that the sorts of injustices that apologies can help to correct need not be departures from ideals specified by distributive justice. Apologies and the moral relations they engage can thus be parts of a domain of justice that is neither distributive nor dependent on distributive justice.

Before proceeding, I should note a few limits to my discussion. First, I do not survey all arguments that discuss the relation of corrective to distributive justice. I focus on some leading accounts to illustrate their structure and limitations. Second, many discussions of the relation between corrective and distributive justice are in service of philosophical accounts of the bases of tort theory. While my arguments may have some implications for such discussions, my main concerns are with the conceptual and normative status of principles of reparative justice. Finally, there are obviously many competing theories of distributive justice and reparative justice. Though my discussion may sometimes draw on some features of particular substantive accounts, I hope not to commit to any one theory. My focus is on a type of moral repair whose justification need not be subordinate to distributive justice.

Section 2 presents an analysis of how corrective justice might be subordinate to distributive justice. Section 3 considers a related account by Stephen Perry to establish the autonomy of corrective justice and shows how it is vulnerable to challenges. Section 4 presents apologies as measures of moral repair. I argue that neither apologies nor the moral relations they cultivate are proper targets of distribution. Section 5 considers objections that reparative duties and claims are ultimately distributive. There I argue that for a certain range of wrongs, the view that corrective justice is secondary to distributive justice is either false or presupposes disputable formal or substantive accounts of distributive justice.

2 Unpacking the View that Corrective Justice is Subordinate to Distributive Justice

Before exploring the relation of corrective to distributive justice, consider each as a type of justice. Justice can be a feature of actions, persons, groups, institutions, or states of affairs. Many considerations are bound up with justice, such as fairness, equality, reciprocity, need, and desert. Justice often comes to us in the form of certain norms, rules, or laws. Theorists sometimes describe justice as a virtue; if it is, it is a (perhaps the) mandatory virtue. It is mandatory because, on many accounts, justice provides very stringent (perhaps peremptory) reasons. The reasons of justice thus take precedence over many if not all other reasons. Those reasons are at least sometimes other-regarding (Hume 1983, sec. 3; Slote 2014; Simmons 2007, chap. 4; Morris 2007; Schmidtz 2006; Raphael 2003; Rawls 2005).

We might dispute parts of this brief outline of justice, and various theories of justice will show stark contrasts in their details. However, I hope to have provided a sufficiently plausible sketch so that we can say this much: both corrective justice and distributive justice, as types of justice, provide other-regarding norms that generate very stringent reasons. For instance, corrective justice sometimes requires various forms of redress such as restitution of stolen goods, compensation for injuries, or an apology for some transgression. Corrective justice also provides reasons. For instance, it provides reasons for individuals it identifies as wrongdoers to take remedial measures by restoring stolen goods, offering compensation for injuries, or apologizing to victims. No doubt, this need not be all that corrective justice does. Corrective justice need not only provide reasons to or claims against wrongdoers. But, surely this is among what it claims to do.

Among the propositions that the norms of corrective justice purport to justify are those with the form “X has a claim that Y perform action Φ”, and those with the form “Y has a duty to X to perform some action Φ”.⁶ Call any such proposition “P”. Corrective justice would cast P as identifying a remedial duty or a remedial claim. The notion that corrective justice is subordinate to distributive justice can then be specified as either or both of two theses:

• (1) The dependence thesis ⁷: Some consideration(s) of distributive justice are among the truth conditions of any such proposition P. Here, distributive justice is necessary for the truth of such statements of corrective justice.⁸

• (2) The normative irrelevance thesis: For any such true statement P, appeals to distributive justice considerations, along with some relevant statements of fact, are sufficient to establish P. Distributive justice overdetermines corrective justice. Corrective justice then has no unique normative work to do.

I argue that the dependence thesis and the normative irrelevance thesis are both false because of at least one counterexample, hence supporting the autonomy thesis. Corrective justice then need not always be subordinate to distributive justice. The next section considers drawbacks to some arguments one might cite for the autonomy thesis. Later I offer my own defense of the thesis.

3 Corrective Justice, Harm, and Distributive Justice

In this section, I consider a defense of the autonomy thesis modeled on Steven Perry’s notion of corrective justice. For Perry, corrective justice has a different function than distributive justice (Perry 2000). On his account, corrective justice is autonomous because of what it takes as normatively significant. Perry’s subtle argument is worth discussing in some detail as it illustrates the drawbacks with one sort of approach to distinguishing corrective justice from distributive justice.

Perry writes that corrective justice is “independent of distributive justice” because it works with entitlements that are not properly “subject to distributive justice” (Perry 2000, 256). Corrective justice targets harm, not distributive shares (Perry 2000, 256–57). Even in cases involving damage to property, corrective justice does not protect entitlements as distributively just (Perry 2000, 256–57). “Corrective justice applies in such cases because the owner of damaged property has suffered harm” (Perry 2000, 257). In this sense, Perry claims, corrective justice is conceptually independent of distributive justice (Perry 2000, 237, 257).⁹

Perry does not think that corrective justice presupposes a set of entitlements about which distributive justice is silent. Indeed, on his view, distributive justice gives something the imprimatur of legitimate entitlement before it can be a candidate for repair in the breach. After that initial work, distributive justice fades from the picture. Corrective justice then focuses on repairing the harms done from the violation of those entitlements previously established as legitimate (Perry 2000, 254). Perry stresses that corrective justice protects those entitlements not in order to preserve some just distribution. It protects them because their violation constitutes harm.¹⁰

One problem with defending an autonomy thesis by using an account such as Perry’s is that it leaves corrective justice crucially dependent on distributive justice. Corrective justice protects entitlements because their violation is sometimes harmful. But, on Perry’s account, harm is a function of entitlement, and entitlement is a function of distributive justice. Corrective justice must then await the blessings of distributive justice.

For Perry, an act counts as a harm in virtue of distributive justice. Corrective justice might propose a remedy when Chris sets fire to Pat’s barn. It targets for repair not the violation of Pat’s distributive share. Indeed, it is unconcerned with entitlements as distributive shares. It targets the harm Pat suffers from Chris’s trespass. The problem is that what makes Chris’s action harmful is that Pat is entitled that Chris not set fire to the barn. Distributive justice, the advocate of the dependence thesis might say, must first confer upon Pat an entitlement that others not burn the barn in order for Chris’s burning the barn to count as a harm demanding redress. For Perry, it is only because of the conclusions of distributive justice that what Chris suffers changes from merely being misfortune to becoming harm. Because entitlement for Perry is distributive entitlement, corrective justice has a dependent role. Perry’s argument thus fails to refute the dependence thesis.¹¹

If some corrective justice claim did not depend on distributive justice, that would be a counterexample to the dependence thesis. Moreover, if distributive justice were unable to show the truth of some statement of corrective justice, we would have some counterexample to the normative irrelevance thesis. I next consider whether claims to apologies might provide such counterexamples.

4 Moral Relations and Nondistributive Entitlement

The previous section argued that we do not establish the autonomy of corrective justice by appealing to its normative focus on harm. Might there be another way to show corrective justice is independent? In the present section, I discuss that question while considering representative arguments that defend a normative irrelevance thesis. I highlight how it is difficult to justify duties and claims to apologies if we may only draw on the resources of distributive justice. I begin with accounts by some proponents of the normative irrelevance thesis.

4.1 Corrective Justice as Normatively Irrelevant

Kok-Chor Tan discusses how the genealogy of some contemporary unjust disadvantages in the historic depredations of colonialism is “normatively irrelevant in a sense” (Tan 2007, 285). That history is irrelevant for justifying reparative duties. Those duties are overdetermined. On Tan’s view, distributive justice identifies maldistributions needing repair. Focusing on historic injustice can help, but only to show a “reason of justice for ameliorating these existing disadvantages” (Tan 2007, 289, 288). Tan does not then propose eliminating appeals to reparative arguments. They “can be a useful argumentative strategy when arguments about global inequality and its impact on the cultural identities of some peoples fail to persuade” (Tan 2007, 289). On this sort of view, distributive justice overdetermines reparative justice. Reparative justice provides no unique reasons, but it might provide special motives. Reparative justice merely makes up for the rhetorical shortfalls of distributive justice.

Leif Wenar offers a related account. On his view, “reparative arguments… only get their power from forward-looking distributive considerations instead of from backward-looking reparative considerations of independent force” (Wenar 2006, 401).¹² In arguments for moral repair, there is a “primacy” to distributive justice (Wenar 2006, 401).

On accounts such as Wenar’s and Tan’s, corrective justice is normatively irrelevant. Distributive improprieties are what ultimately matter. Historic injustice merely flags which ones should grab our attention first.

One challenge for proponents of the normative irrelevance thesis is that if distributive justice does everything that corrective justice purports to do, we might elide the distinction between compensation and reparation. Compensation provides some resources or opportunities to someone who has suffered some shortfall. Reparations are measures transgressors provide to compensate their victims or otherwise make up for the wrongful injuries they inflicted.¹³ I next consider the importance of this distinction and the difficulties of sustaining it under the normative irrelevance thesis.

4.2 Moral Relations, Compensation, and Reparation

Consider a case where Sam takes Tom’s bicycle without Tom’s permission. On many understandings of corrective justice, Sam must return the bicycle. More than that, Sam owes it to Tom to return Tom’s bicycle. Corrective justice may also say that Sam owes it to Tom to provide an apology.

Many political moralities would underwrite such claims. For instance, various liberal egalitarian political theories would say that we should interact with one another only according to mutually acceptable reasons. Among those reasons would be a commitment to a sort of equal moral standing by which each of us has claims against others that they justify their actions according to principles we can all find mutually acceptable. The libertarian family of political moralities would treat persons as project pursuers, bearers of very stringent rights, or equally free autonomous agents, and in any case entitled to other persons’ forbearance in many respects. In many of these and other theoretical frameworks, wrongdoers owe repair to their victims. There can be directed obligations of repair that include, but might not be exhausted by, compensation.

Consider now whether distributive considerations can justify material compensation as part of reparations. When Sam steals Tom’s bicycle, anyone can compensate Tom for his loss. Some (or any) party’s doing so might be permitted and perhaps required by distributive justice. We can imagine, for instance, a system of distributive justice that compensates people for their misfortunes. Perhaps, as Boxill and Goodin separately suggest, distributive justice is concerned with guaranteeing that people have adequate opportunities to pursue their life plans (Boxill 2014; Goodin 1991). Distributive justice might even assign the burdens of compensation for misfortune to persons or parties meeting certain criteria, such as the state, or all persons above a certain level of wealth. The question is whether distributive justice assigns reparative burdens to Sam the bicycle thief and, to other transgressors of injustice, all of the appropriate burdens of reparation.

Certainly, reparations are sometimes impossible. Wrongdoers or their victims might be dead. The wrong may have irremediably altered irreplaceable goods and so make reparations impossible. Often, however, there can be some steps toward repairing injuries. This repair may include but is often not exhausted by compensation. What is key, though, is that for compensation to be (part of) moral repair, wrongdoers must provide it. Anyone can compensate someone, but only wrongdoers can repair the injuries they inflict. There are no vicarious reparations.¹⁴

To illustrate the challenge for rooting all reparative claims and duties in distributive justice, consider the moral relations injustices upset. Certain normative features of human interactions capture the moral relations I have in mind. They range over expectations, powers to lay claims, liability to claims, and the mutual acceptance of values regarding what we may do, must do, and must not do for and to one another. These relations provide reasons for action. The shared understanding that there are such reasons provides a framework for interaction and expectations. Obviously, the details depend on the substantive moral values shaping the relations, but the general idea is that moral relations are features of how people interact when governed by moral reasons.

Margaret Urban Walker understands moral relations as marked in part by shared dispositions and expectations (Walker 2006). Such dispositions and expectations take certain shared values as their subject matter. On her view, when moral relations are in place, people mutually acknowledge certain normative standards. They are disposed to support certain shared values and respond to each other as accountable claimants within a body of mutually acceptable moral reasons. Moral relations depend on hope and trust in such shared values.

Sustaining such relations is part of ordinary life. There are rituals, habits, and customs. Of special note, however, are moral relations in the breach. Repairing such relations is sometimes no easy matter. (See, for instance, the discussion in Walker 2006, chap. 6.) It might not involve simply returning a stolen bicycle. It might concern injuries that can never be fully repaired. The violations might be part of a pattern of depredations that spanned generations and involved entire peoples. Confronting the histories of such injuries can be excruciating.

There may be a corresponding right in the victim to moral repair. On Locke’s view, for instance, anyone in the state of nature has the right to punish criminals. Though magistrates might wield such rights in political communities, they “cannot remit the satisfaction due to any private man for the damage he has received” (Locke 1980, sec. 11). Locke does not specify what such “satisfaction” entails, but it certainly sometimes includes material compensation by the wrongdoer. Many victims of material depredations would not be satisfied when third parties compensate them for their unjustly caused misfortunes.¹⁵ Whether or not victims have rights to such repair, any compensation they might receive is not reparation until and unless the wrongdoer provides it.

The reason wrongdoers must provide reparations is not mainly a function of material conditions or levels of welfare. After all, many material goods are fungible. Stolen bicycles can be replaced and resources to pay for assault victims’ material losses can come from any pocket. Wrongdoers, however, are the only ones to repair some injuries. Their providing such repair may be necessary for undoing the degrading treatment the transgression entailed (Murphy and Hampton 1988). For example, many of the “comfort women” who were confined to sexual servitude at the hands of the WWII Japanese military initially refused compensation because the Japanese government did not provide it (Barkan 2001, chap. 3). Third parties cannot provide the repair that wrongdoers owe.

To be clear, specifying what compensation should be is often a difficult matter. Indeed, sometimes fixing compensation is inappropriate. What price is there for the years of sexual slavery of the “comfort women”? In this and other cases such as African slavery, Hitler’s genocidal gas chambers, Rwandan slaughter, and regrettably, and so on, it may even seem a further affront to offer compensation as (part of) reparations. However, it need not always be so. Sometimes, a condition of restoring moral relations is that transgressors provide some compensation.

4.3 Apologies as Moral Repair

Let us now return to the main question about the relation of corrective to distributive justice. Advocates of the normative irrelevance thesis might have no problem with the claim that transgressors ought to compensate their victims when possible. If part of the task of distributive justice is to assign benefits and burdens regarding various goods, then distributive justice might say this: in cases where X’s transgression wrongly harms Y, then X should at least be among those assigned the burden of compensating Y. Indeed, wrongdoers might be the only ones assigned the burdens of compensation since, sometimes, asking anyone else to bear such burdens might infringe on their entitlements.¹⁶ For proponents of the normative irrelevance thesis then to succeed in showing that corrective justice does no unique normative work, they must show that distributive justice can always justify the same moral repair corrective justice supports.

Perhaps distributive justice can do this with justifying amounts of material compensation. Material compensation is not, however, the whole of moral repair. Indeed, sometimes it has nothing to do with it. In order to restore or create moral relations among transgressors and their victims, transgressors might also (or instead) need to admit their wrongdoing, commemorate or memorialize history, and, most notably, provide an apology. The case of apology illustrates how difficult it is for distributive justice to require the full range of reparative practices prescribed by corrective justice.

Apologies serve many functions, but among them seem to be restoring and nurturing relationships as well as publicly acknowledging a shared commitment to some moral values.¹⁷ Consider how corrective justice might call for an apology for a transgression. We need not settle the specific form the apology might take nor what if any material compensation should accompany it. The substantive details of an account of corrective justice would specify such measures, or, at least, a range of possibilities. Whatever the details, the repair corrective justice might require could include an apology of some sort. However, justifying such moral repair does not seem to be the task of distributive justice.

Distributive justice seems mainly concerned with allocating the benefits and burdens regarding various goods that are divided or allocated among a group of persons. However, appeals to the norms of distributive justice (plus some statements of fact) cannot establish propositions for claims or duties regarding apologies. Before corrective justice identifies occasions to apologize, there is no correct distribution of apologies.¹⁸ This is partly because, even under full compliance, the correct distribution of apologies is not one where there are no apologies to distribute (because, say, there would be no occasions to apologize). Even under full compliance, it seems likely that there would be occasions for agents to apologize. Innocent misunderstandings may arise that call for apologies as measures to solidify moral relations. These moral relations would create claims and duties that distributive justice seems ill suited to specify. Corrective justice seems crucial for establishing claims or duties regarding apologies. This suggests the normative irrelevance thesis is false.

In this section, I sketched challenges for any defense of the normative irrelevance thesis. Distributive justice does not necessarily assign wrongdoers reparative burdens, and it does not require that wrongdoers provide all apologies needed to repair moral relations.

A critic might offer the following rejoinder: since distributive justice determines what we may claim of or owe to others, then it can govern apologies. In the next section, I consider some versions of this rejoinder. Their weaknesses open the door further for the autonomy thesis.

5 Objection: Reparative Duties and Claims are Distributive

The previous section outlined part of an argument for the autonomy thesis by suggesting some reparative claims and duties are not justifiable by distributive justice. Critics may object that anything required by corrective justice must be rooted in or specifiable by distributive justice. After all, this seems to be what distributive justice does: it assigns benefits and burdens. In this section, I canvass four versions of such an objection. They each say the autonomy thesis is false, either because (a) distributive justice constrains corrective justice, (b) distributive justice establishes primary social goods, (c) distributive justice is the framework for understanding what it means for there to be a right or a duty, or (d) distributive justice establishes all rights and duties. I discuss each in turn.

5.1 Distributive Justice as a Constraint

The autonomy thesis holds that some duty or claim regarding an apology is neither specifiable by nor dependent on distributive justice. A critic might object that distributive justice limits what corrective justice can prescribe. For instance, a person’s claim for some reparation is defeasible. Distributive justice might be among the considerations that constrain corrective justice.¹⁹ Even Locke notes that the right of reparation only protects repair of unjustly inflicted injuries, but never allows infringing on the rights of third parties such as “the innocent Wife and Children” of unjust combatants (Locke 1980, sec. 180). We might add as a further example that one may not satisfy a duty to deliver an apology at the expense of an innocent third party’s property rights.²⁰ Suppose, for instance, A owes B an apology. C is an uninvolved third party. A may not satisfy the duty to B by appropriating C’s pen to write a letter of apology. Nevertheless, these points do not show reparation is a distributive entitlement. A limiting condition is not a justifying condition. That distributive justice sometimes constrains corrective justice establishes neither the dependence thesis nor the normative irrelevance thesis.

5.2 Distributive Justice, Primary Goods, and the Basic Structure of Society

Another criticism of the autonomy thesis might draw on Rawls. On Rawls’s view, there are certain “primary goods” that any rational person “wants whatever else he wants” (Rawls 2005, 92). Among those goods are the social bases of self-respect. Self-respect “includes a person’s sense of his own value, his secure conviction that his conception of his good, his plan of life, is worth carrying out” (Rawls 2005, 440). Self-respect, Rawls adds, “implies a confidence in one’s ability, so far as it is within one’s power, to fulfill one’s intentions” (Rawls 2005, 440). Persons who have suffered unjustly inflicted misfortunes might consequently suffer shortfalls in Rawlsian self-respect. They might be insecure in their senses of self. They might lack confidence in their abilities to carry out their plans or doubt that their plans are worthwhile. Since parties to the original position decide on the bases of social life in a self-contained society, among the items to which they would agree are the conditions under which agents would be obligated to provide apologies and/or entitled to claim them. Consequently, the objection concludes, reparative justice is ultimately distributive justice. If so, then both the dependence thesis and the normative irrelevance thesis would be true.

In reply, we should note that not all occasions of corrective justice revolve around restoring or preserving self-respect or its social bases. Sometimes, for instance, victims of unjust depredations suffer no shortfalls in Rawlsian self-respect. They have gone on with their lives, and though they might have a great interest in receiving moral repair from their transgressors, such is not necessary for their self-respect. None of this would seem to undermine the positions that victims have claims to an apology or that transgressors owe apologies to their victims. Furthermore, parties to the original position agree to principles governing the basic structure of society assuming full compliance. They are unconcerned with principles to give “satisfaction” to the victims of injustice, stout or otherwise.

The Rawlsian account would not give to distributive justice the task of providing all (and perhaps any) reasons of moral repair. As Richard Lippke notes (Lippke 1999, 156, 165). on a Rawlsian account, distributive justice and corrective justice have different functions. Distributive justice focuses on the basic structure of society. It would be unconcerned, for instance, with justifying claims to an apology for some wrong.

All this might show a Rawlsian account is poor resource for advocates of the normative irrelevance thesis. However, Rawls can be a resource for another sort of objection to the autonomy thesis. On this objection, distributive justice sets the context under which corrective justice operates. For instance, distributive justice would specify a minimally acceptable distribution. If that were not in place, no remedial claim or duty under corrective justice could be true.

At the risk of an unfavorable acronym, call the objection that appeals to the significance of a minimally acceptable distribution the MAD objection. According to MAD, the dependence thesis is true: any true proposition about reparative duties or claims presupposes some fundamental distributive truths, such as, for instance, about the basic structure of society.²¹ However, MAD is problematic. Much depends on how we specify the minimally acceptable distribution. If we are Rawlsians, for instance, a political community that fails to secure certain core human rights might not be minimally acceptable (Rawls 1999, pt. II). A proponent of the MAD objection, however, is committed to denying that persons in such societies can ever have any claims or duties regarding apologies. This is implausible. Surely individuals might sometimes owe apologies to others even if certain distributive minima fail to obtain. If Jack lacks access to minimally acceptable healthcare, that does not mean Jill would not owe him an apology as part of moral repair for wrongly punching him in the nose.

The MAD proponent might refine the notion of “minimally acceptable”. Perhaps the problem is not so much the failure to secure certain core human rights. Perhaps the problem is that persons whose distributive context is not minimally acceptable fail to enjoy any core human rights. They are profoundly insecure in their persons. They have no property. Their lives are, for lack of a better description, “solitary, poor, nasty, brutish, and short” (Hobbes 1968, chap. 13).

Such a MAD proponent would then have gone too far. Appeals to a Hobbesian state of nature do not support MAD. An inadequate distribution is not the problem with the Hobbesian dystopia. The problem there is that any claims of justice—including those of distributive justice—would have no “place” in that Hobbesian state of war (Hobbes 1968, chap. 13). This would not show the dependence thesis is true. At best, it shows justice must be possible before there can be any valid claims or duties regarding apologies.

A critic of the autonomy thesis might here defend one sort of view in which all claims and all duties are distributive.²² Call this the imperial view of distributive justice.²³ According to the imperial account, entitlements and duties are distributive notions because it falls to distributive justice to specify the basis, conditions, or meaning for any juridical relations. There are two versions of the imperial view, which we can specify when considering two more objections to the autonomy thesis.

5.3 Metaethical Imperialism About Distributive Justice

One way to defend both the dependence and normative irrelevance theses is to appeal to a metaethical view about juridical terms. On this view, any statement that there a claim or duty regarding an apology is a statement about distributive considerations. The imperial metaethical view holds that statements about claims or duties simply are distributive claims. To be a claim or duty is to be specified as a claim or duty by distributive justice. This is a thesis about meaning. This imperial view rejects the autonomy thesis by defending the metaethical position that entitlement and duty are simply distributive notions. The normative irrelevance thesis and the dependence thesis would then both be analytically true.

This sort of view is implausible. It is not necessarily a function of the meaning of terms that statements of entitlement and duty are distributive statements. For instance, to say Jack owes Jill an apology for insulting her does not mean that Jack has been assigned such a duty by distributive justice. Of course, it might be a mistake to believe that corrective and distributive justice sometimes mean different things. Proponents of the metaethical imperialist view would then need to provide an error theory and perhaps revolutionize our understanding of justice. Pending that account, let us bracket this sort of objection to the autonomy thesis.

5.4 Substantive Imperialism About Distributive Justice

Critics of the autonomy thesis might defend a different sort of imperialist theory, this time a substantive one. They might argue that any reparative claim or duty must be designated by distributive justice. On this view, any reparative duties or entitlements are distributive duties because it falls to distributive justice to specify what duties and entitlements we have.

Consider the possibility that corrective justice says Jack owes Jill an apology for stepping on her toe. A proponent of this imperial version of distributive justice might insist that it is in virtue of the ideals and standards of distributive justice that Jill’s toe counts as her toe. Distributive justice justifies how anything, including a toe, can belong to Jill. Thus, the dependence thesis is supposedly true. More than that, if there were a remedial duty or claim regarding apologies for the toe incident, distributive justice would assign it. On this imperial view, distributive justice exhaustively determines which claims and duties we have by justice. Thus, the normative irrelevance thesis is supposedly true.

Suppose Jack insults Jill with a racist epithet. Corrective justice might say Jack owes Jill an apology (and perhaps more than that, and perhaps not merely to Jill). A critic of the autonomy thesis might say, however, that such an epithet fails to treat Jill as a moral equal. The epithet is thus a failure to uphold distributive justice. According to this view, distributive justice assigns us a moral standing such that others owe us certain respect and forbearance. That standing entitles all of us to an autonomy that racist epithets infringe. Distributive justice would then assign to Jack the obligation to apologize to Jill.²⁴

This view seems mistaken. It is not the task of distributive justice to assign benefits or burdens regarding apologies. Distributive justice does not specify all the conditions that corrective justice identifies as calling for an apology in the breach. The moral relations that an apology might help to repair or construct are not always distributive relations. Corrective justice may prescribe apologies, but not always for a violation of distributive norms.²⁵ Apologies may be required for violations of other norms, such as those that draw on appropriate moral relations.

Consider in particular the moral relations expressed by the mutual acceptance of values indicating what we owe to one another. Some of those values are distributive values, but some are not. Mutual respect and honesty, for instance, are not distributive relations. In light of these and other moral considerations that define moral relations, sometimes we might owe an apology, or someone might be due an apology from us, but not because of violation of some principle of distributive justice. A person’s standing as someone due equal concern and respect (for instance) is not something that must depend on or be specifiable by distributive justice. Distributive justice might presuppose and express such moral standards, but they need not be distributive ideals.²⁶

Corrective justice required various remedies of the US government for its complicity in the Tuskegee syphilis experiments. For four decades, the US Public Health Service lied to hundreds of African American men and allowed their syphilis to proceed unchecked (Reverby 2000). Among the moral costs to the experiments were betrayals of trust and betrayals of the experimental subjects’ dignity. The study also perpetuated group-based harms against African Americans who were not experimental subjects; it expressed the idea that their race is a justification for treating them as things. Advocates of the dependence thesis must argue that all the reasons for moral repair arising from this experiment are distributive. This seems mistaken. The US government denied the experimental subjects their due, but not all they were due is a function of distributive norms. For instance, the subjects’ immunities to betrayal of trust are an important features of their moral status, but such immunities seem ill-suited as targets of distribution, and so too the related claims to apologies in the breach. The US government’s duty to create or repair moral relations through an apology in response to Tuskegee is not merely a duty in light of the distributive moral relations it violated. The Tuskegee study was not merely a maldistribution. It violated other moral relations.

Moving away from institutional to interpersonal moral relations, suppose that Pat and Chris were in a committed and exclusive romantic relationship but that Chris was unfaithful. Corrective justice may say (at least) that Chris owes Pat an apology. The imperialist view implausibly supposes some distributive norms specify and are necessary conditions for all directed duties and claims of fidelity between Pat and Chris. However, the duty to apologize, and Pat’s claim to receive one, are neither dependent on nor specifiable as breaches of distributive relations. Chris owed fidelity to Pat. We need not appeal to distributive norms as a condition for the truth of remedial claims and duties regarding Chris’s infidelity. Hence, the dependence thesis is false. We do not establish remedial claims and duties regarding Chris’s infidelity merely by appealing to distributive norms. Hence, the normative irrelevance thesis is false.

A further reason to reject this imperial view of distributive justice is that apologies are not the sort of resources subject to transfer. However, they would need to be transferable, at least in principle, in order to be governed by distributive justice. Consider, for instance, how distributive justice governs certain political or civil liberties. We could transfer or trade such liberties. Of course, many political moralities (especially Rawlsian ones) would forbid some such transfers for substantive reasons. But the transfers are at least conceivable. Such transfers would simply not make sense, however, for claims and duties regarding apologies. Individuals cannot purchase or sell apologies. Since distributive justice seems to govern transferable resources, then it cannot cover apologies.²⁷

Before concluding, we should consider one further criticism of the autonomy thesis. The criticism begins by acknowledging that some claims or duties regarding apologies have nothing to do with distributive justice. For instance, you might owe it to a friend to provide emotional support at certain times and in certain ways; an insensitive failure to provide it would call for an apology regarding which there are duties and claims. However, the critic has a potential reply. Such claims and duties to apology in friendship are beyond the reach of justice. If they are beyond the reach of justice, then corrective justice does not prescribe them. Consequently, the critic concludes, appeals to claims and duties regarding apologies in friendship fail to establish the autonomy thesis.²⁸

This objection fails. It supposes there can be directed moral duties and directed moral claims neither governed by nor founded in justice. But such claims and duties are the province of justice. Instead, a more plausible view is that justice presents at least some duties and claims of friendship. Such claims and duties are not under the jurisdiction of distributive justice. For instance, justice might require acknowledging, “I owe you an apology for betraying your confidence as a friend”. The proposition can be true without appeal to distributive norms. Moreover, on many accounts, the proposition cannot be established by distributive norms. Of course, this is not a complete refutation of the objection; it merely increases its costs. The objection is committed to disputable substantive accounts of justice, distributive justice, and friendship.²⁹

An alternative to the imperial views is to say that distributive justice is one part of a wider theory of justice. That wider theory presents various mandatory and other-regarding norms. The norms determine what we morally owe to and may claim of each other. Among the things we might morally owe to each other (and to which we might sometimes lay claim) are apologies for past injustices. The moral reasons, claims, and duties regarding such apologies are sometimes neither specifiable by nor dependent on the findings of distributive justice. Moreover, sometimes, the moral relations such apologies repair or preserve are not targets for distribution.

6 Conclusion

It might be impossible to correct some injustices fully. A rapist cannot undo the violation. Governments cannot erase the damage they inflicted on generations of slaves. Insults can never be unsaid. Nevertheless, these injuries might be partly offset by a wrongdoer’s apology that acknowledges the injustice and takes steps toward restoring or creating moral relations. Corrective justice furnishes reasons for such apologies. They need not be reasons of distributive justice. The claims or duties regarding such apologies need not turn on distributive entitlements.

I have argued that corrective justice sometimes requires measures of moral repair. Those measures can include apologies that admit to wrongdoing and offer compensation as reparation for an injustice. Distributive justice might very well constrain the performance of any duties to provide such repair. It might also limit the reach of any supposed entitlements to reparations. However, not all reparative claims and duties are justified by appeal to distributive justice, nor does the truth of all statements about such reparative claims and duties depend on the findings of distributive justice.

Critics might insist on an imperial view of distributive justice. On such views, justice simply is distributive justice, either because justice means distributive justice, or because distributive justice exhaustively specifies our duties and claims. Such views are, at least, controversial.

I have shown that many leading objections to the autonomy thesis are committed to disputable substantive or metaethical views. If those and other objections fail, sometimes corrective justice generates reparative claims and reasons that are not and cannot be distributive. Corrective justice sometimes picks out a domain of justice that is not distributive.