The distinctiveness of relational equality

Abstract

In recent years, a distinction between two concepts of equality has been much discussed: “distributive” equality involves people having equal amounts of a good such as welfare or resources, and “social” or “relational” equality involves the absence of (certain kinds of) social hierarchy and the presence of (certain kinds of) equal social relations. This contrast is commonly thought to have important implications for our understanding of the relationship between equality and justice. But the nature and significance of the distinction is far from clear. I examine several accounts of this issue and argue none are entirely satisfactory. In turn, I offer an alternative proposal. Relational equality, on my account, involves a concern with each person having an equal “civic status.” I characterize this concern and show it has distinctive and normatively significant positional and temporal aspects.

Introduction

In recent years, a distinction has emerged between two substantive concepts of equality. ¹ In rough terms, “distributive” equality involves the equal distribution of a good such as welfare, resources, capabilities, or something else, whereas “social” or “relational” equality involves the absence of (certain kinds of) social hierarchy and the presence of (certain kinds of) equal social relations. Each of these concepts may be developed or specified in various ways, and so each concept allows for various conceptions. ² There are different conceptions of distributive equality that vary with regard to the good(s) that are to be distributed equally, as well as the principles that are taken to justify deviation from strict equality—for example, personal responsibility, as on luck egalitarian accounts, or benefit to the worst-off, as on Rawls's difference principle. Likewise, the relevant kinds of “relations of equality” may be specified in various ways—in terms of particular kinds of interpersonal dispositions and attitudes, say, or as a kind of institutional status shared as equals. ³

This distinction between the distributive and relational concepts is often regarded as significant, clarifying what is at stake in working out the relationship between equality and justice. It is tempting, in this regard, to draw a comparison with the case of freedom, for which the distinction between the so-called “positive” liberty of the ancients versus the “negative” liberty of the moderns has come to be widely regarded as useful and important. ⁴ Understanding the nature and significance of the contrast between distributive and relational equality, then, would appear to be important as well.

At present, however, the nature and significance of the distinction remains unclear, with little consensus emerging in the literature. On the one hand, some relational egalitarians suggest that distributive egalitarians have badly misunderstood the nature of egalitarian justice—“missing the point,” as the title of Elizabeth Anderson's well-known paper suggests. ⁵ Elsewhere, she writes that “the background conceptual framework of the ‘equality of what?’ debates in contemporary political philosophy is misguided…” having been “launched on the assumption that there exists a single good that egalitarians should want to see equally distributed.” ⁶ On the other hand, defenders of the distributive egalitarian approach resist these claims. Some pose a “reductionist challenge,” suggesting that relational egalitarians simply argue for the equal distribution of “social” or “relational” goods—for example, status, standing, nondomination, and so on—and so only propose another version of distributive equality. ⁷ Others suggest hybrid accounts: they hold that distributive and relational concerns are distinct, but can (and should) be incorporated into an overall account of egalitarian justice. ⁸ Without a clear understanding of what distinguishes relational from distributive equality, however, it is difficult to judge what is at stake in choosing between these various views. Moreover, there has been a tendency in the literature to contrast relational equality more narrowly with luck egalitarianism—one version of distributive equality among others. ⁹ While these discussions are certainly interesting, they do not always help to clarify whether and in what sense there might be a more general distinction between distributive and relational equality.

The aim of this article is to make progress on these issues. I believe it is indeed useful to distinguish distributive and relational equality, but the nature and significance of the distinction has not yet been given an adequate characterization. In what follows, I examine several proposals that have emerged in the literature. These proposals adopt various strategies, suggesting the distinction can be understood as a difference of justification, scope, or content (The justification strategy, the scope strategy, and the content strategy sections). In each case, I argue none of the proposals is entirely satisfactory. Then (in the civic status as a distinctive egalitarian concern section), I propose an alternative version of the claim that relational equality has a distinctive content, drawing on insights gained from the previous discussion. I suggest that relational equality is best understood as involving a concern with equality of civic status and that civic status has distinctive and normatively significant positional and temporal aspects. The final section concludes.

The justification strategy

Second versus third-person justification

One proposal, developed at length by Anderson, is that what is distinctive of relational egalitarianism is that it adopts the “second-person standpoint” of justification. Distributive egalitarians such as luck egalitarians, by contrast, view justification from a “third-person standpoint.” These notions are defined as follows:

In a third-person justification, someone presents a body of normative and factual premises as grounds for a policy conclusion. If the argument is valid and the premises are true, then the conclusion is justified. The identity of the person making the argument and the identity of her context are irrelevant to the justification. By contrast, most relational egalitarians follow a second-person or interpersonal conception of justification. This follows from their contractualism …. In a second-person justification, a claim of justice is essentially expressible as a demand that a person makes on an agent whom the speaker holds accountable. Justification is a matter of vindicating claims on others’ conduct. Vindication involves demonstrating that the claims are addressed to those properly held substantively responsible for the conduct in question, by persons entitled to the moral authority or standing to hold them to account. ¹⁰

It is doubtful, however, that adopting the second-person standpoint of justification is a necessary aspect of relational equality. As Anderson writes, it may be that most relational egalitarians adopt a second-person form of justification. But it does not seem that anything requires relational egalitarians to so justify their view. For example, one might claim that equal social relations are intrinsically good, and this fact alone gives us moral reasons to bring them about. This kind of justification, however, need not appeal to people's authority to demand of one another equal social relations in a second personal way. Instead, it might hold that equal social relations are valuable from a third-person perspective—from the “standpoint of the universe,” as it were. In this way, it is at least coherent for a relational egalitarian view to be justified from a third-person standpoint, however plausible such a justification is ultimately judged to be. So, it seems that the second-person standpoint is not strictly necessary for relational egalitarianism. ¹¹

It is also doubtful that endorsing the second-person standpoint of justification is sufficient for an account to qualify as a version of relational egalitarianism. Nothing rules out justifying a distributive notion of equality in these terms. In fact, Ronald Dworkin's theory of equality of resources appears to offer such an example. ¹² At the same time, however, Dworkin has been targeted by relational egalitarians as a case in point of the kind of distributive egalitarianism they reject. ¹³ Dworkin's view fits the distributive paradigm because he holds that people are owed an equal distribution of resources while allowing that inequality is justified when it results from voluntary choices. Dworkin justifies his view in a second-personal way, however. In order to determine what makes a distribution of resources count as equal, Dworkin asks us to imagine a group of people shipwrecked on a desert island. He supposes that these people “accept the principle that no one is antecedently entitled to any of these resources, but that they shall instead be divided equally among them.” ¹⁴ If people recognize each other's equal claims to the resources on the island, Dworkin argues that a justified distribution must pass what he calls the “envy test.” “No division of resources is an equal division if, once the division is complete, any immigrant would prefer someone else's bundle of resources to his own bundle.” ¹⁵ Dworkin suggests we can think of how this might be achieved through the notion of a hypothetical auction of the island's resources along with hypothetical insurance markets for disadvantages due to lack of talent or handicaps. Starting with an equal number of tokens, each person can bid on the resources they want, as well as pay into insurance schemes so they will be compensated for handicaps or low marketable talent. If through such mechanisms the people on the island reach a distribution such that no one prefers the resource bundle of another, the distribution is justified.

This theory seems to be very much a version of the distributive concept of equality, but in giving its justification Dworkin appeals to a kind of second-personal test. The requirement that the distribution of resources be envy-free is based on the idea the inhabitants recognize each other's authority to demand an equal distribution. Dworkin's idea is that each person has claims on the others that they do not appropriate a bundle of resources that would cause envy. The nature of their claims, in addition, depends on facts about their identities, such as facts about which resources they need or value as individuals. It is not the case, then, that Dworkin's conception of resource equality is justified from a third-personal “standpoint of the universe” where the identity of individuals does not figure into the justification. Instead, his justification of equality of resources directly appeals to the kind of claims individuals can make on one another in virtue of their needs and preferences, on the one hand, and to their position of equal authority, or equal entitlement to the island's resources, on the other.

It appears, then, that adopting the second-person standpoint of justification is neither necessary nor sufficient for relational egalitarianism. One might give a third-person justification for relational egalitarianism, or one might justify distributive egalitarianism second-personally. It might be suggested that, on the former possibility, there is a kind of relational egalitarian commitment missing; namely, the relevant kind of justification, and so the view is not relational egalitarian “all the way down.” In addition, it might be suggested that on the latter possibility, there is a kind of relational egalitarian commitment present; namely, its justification. This is a fair enough suggestion, but even if we accept it, there would still be something lacking in characterizing the distinctiveness of relational equality. Relational equality—at least in its most interesting form—is intended to illuminate an important concern with social hierarchy, distinct from equality of “distributive” goods like resources or wellbeing. However, pointing to a distinctive kind of “relational” justification alone fails to vindicate this claim.

Grounded versus ungrounded

Another suggestion along similar lines, put forward by Samuel Scheffler and Kristi Olson, is that distributive egalitarianism is ungrounded, whereas in relational egalitarianism, concerns with distribution are grounded by the ideal of people relating as equals. According to this proposal, distributive egalitarians simply assert that some form of distributive equality matters without appealing to any grounding commitment. By contrast, Scheffler argues that “any form of distributive egalitarianism, if it is to be persuasive, must be rooted in a more general conception of equality as a moral value or normative ideal.” ¹⁶ In Olson's view, this is the “central charge” of relational egalitarianism. ¹⁷

There are, however, several problems with this suggestion. First, relational egalitarians need not view every form of distributive equality required by justice as grounded in relational equality. ¹⁸ Second, it is not the case that distributive egalitarianism is necessarily ungrounded. To be sure, some such as G.A Cohen seem to hold an ungrounded view when he remarks “I take for granted that there is something which justice requires people to have equal amounts of, not no matter what, but to whatever extent is allowed by values which compete with distributive equality.” ¹⁹ But not everyone shares this view. By contrast, some luck egalitarians are explicit that their view is grounded in a more fundamental concern with autonomy. ²⁰

Third, and most importantly, this suggestion faces the same problem as we saw previously: it fails to vindicate the claim that the relational view involves a distinctive and important concern with social hierarchy that is not captured by distributive equality. Again, it would be disappointing if relational egalitarians only offered a distinctive kind of grounding for familiar forms of distributive equality, without identifying any distinctive concern with social hierarchy. So, if there is indeed something distinctive about relational egalitarian commitments in the domain of distributive justice, it seems we should consider other proposals.

The scope strategy

We’ve already noted, roughly, that relational egalitarians are concerned with equal social relations, whereas distributive egalitarians are concerned with equality in the distribution of goods. But what exactly does it mean for relational egalitarians to have a distinctive “social” or “relational” kind of concern? In what follows, we will consider several ways this thought might be specified. As will become clear, it is difficult to identify a kind of concern that might not also be shared by a version of distributive egalitarianism.

As a first pass, consider the following suggestion: distributive egalitarians care about an equal distribution of some good independently of whether it involves people who stand in social relations, whereas relational egalitarians are concerned with equality between people who stand in social relations. This way of understanding the distinction is suggested by Elizabeth Anderson when she says that

equality in the distributive conception consists in the mere coincidence of what one person has with what others in the comparison class independently have and need not entail that the persons being compared stand in any social relations with one another. They might even live on different planets and have no interactions with each other. ²¹

On this suggestion, relational equality would contrast most clearly with what Parfit describes as “telic egalitarianism,” which holds that inequality is in itself bad regardless of the place or even time between the individuals subject to comparison. ²² It is bad, according to telic egalitarianism, if an Incan peasant is worse off than someone today. By contrast, relational egalitarians only care about equality within (particular sorts of) social relations.

This suggestion identifies a very plausible necessary condition for a theory to count as a kind of relational egalitarianism; but it does not provide a satisfactory characterization of the difference between distributive and relational equality. If, on the one hand, one is concerned, like a telic egalitarian, with the equal distribution of something independently of social relations, then it's hard to see why we would count them as endorsing a relational conception of equality. On the other hand, however, distributive egalitarians are not necessarily committed to telic egalitarianism. Nothing rules out valuing equality in the distributive sense, but also holding that it matters only between those who stand in social relations. Kok-Chor Tan's view, for example, restricts the scope of the luck egalitarian principle to apply only to institutions. ²³ Furthermore, “statists” like Michael Blake and Thomas Nagel argue that the obligations of distributive justice only arise between people who share a coercive system. ²⁴ It is entirely coherent, then, to hold that distributive equality—for example, of resources or welfare—only matters between those who interact or stand in particular social relations. This restriction of scope to social relations, however, does not capture a sense in which relational egalitarians have a distinctive kind of “relational” concern within social relations.

The content strategy

Concern with “relational goods”

A more promising proposal, then, is that relational egalitarians are concerned with the quality of social relations, or with goods that in some sense inherently involve social relations. This is suggested by Anderson when she says that “we should be fundamentally concerned with the relationships within which goods are distributed, not only the distribution of goods themselves.” ²⁵ But how exactly should we understand what it means to be “concerned with relationships”? The thought seems to be as follows. On the one hand, relational egalitarians care about whether people have equal status and avoid relations of domination, stigmatization, marginalization, and oppression. On the other hand, distributive egalitarians are concerned with goods like well-being, resources, capabilities, and so on. The former kinds of concern, it appears, involve social relations in a sense not shared by the latter.

While this suggestion has intuitive appeal, it is difficult to specify what it means for some concerns to be “relational” in a way that is not shared by some “distributive” goods, such as resources, well-being, or capabilities. I will consider two interpretations of this suggestion, finding that neither can adequately distinguish the relational and distributive approaches. ²⁶

On one interpretation, what it means for a good to be “relational” is for it to be constituted by social relations. For goods such as well-being or resources, we can coherently imagine what it means for two people who have no interaction with each other (or anyone else), to have different amounts of that good, and thus to be more or less equal in this regard. Social goods like status, on the other hand, are inherently social: it is impossible to have a status of any sort without social relations. Likewise, social hierarchy that involves domination, stigmatization, oppression, and marginalization are all “internal to” social relations. ²⁷ One might think, then, that concern with these kinds of “relational goods” is distinctive of relational equality; and, by contrast, distributive equality is concerned with “atomistic” goods that can be enjoyed by lone individuals.

The problem with this suggestion, however, is that on closer inspection we can notice that many important goods central to the distributive approach are also constituted by social relations. On some accounts, for instance, social relations involving love or friendship might be taken to be constitutive of well-being. Likewise, some basic capabilities are inherently social, such as those which Martha Nussbaum identifies as involving “affiliation.” ²⁸ Consider, also, the case of resources. It is true that we can imagine a person alone on a planet having more or fewer resources. However, I take it that when egalitarians are concerned with resources as they matter for justice, they are concerned with resource ownership, and this constitutively involves social relations. In fact, Ronald Dworkin is explicit that he understands resources in this sense. He points out that “private ownership … is not a single, unique relationship between a person and a material resource, but an open-textured relationship many aspects of which must be fixed politically,” thus, “the question of` what division of resources is an equal division must to some degree include the question of what powers someone who is assigned a resource thereby gains.” ²⁹ And the relevant “powers” are inherently social: they involve, for instance, the ability to exclude or transfer to others. It is not obvious, then, that we can distinguish relational and distributive equality on the basis of the former being concerned with something constituted by social relations. Doing so ignores the sense in which so-called “distributive” goods such as well-being, capabilities, or resources may be constituted by social relations as well.

Another version of the suggestion that relational egalitarianism involves a distinctively “relational” concern involves not the existence, but instead the value of the goods in question. This interpretation is also suggested by Anderson when she writes that “within the distributive conception of equality, the good to be distributed equally … is such that the amount that one person has is typically logically independent of the amount of the good that the others in the comparison class have and also often logically independent of that person’s social relations to the others in the comparison class.” ³⁰

The issue with this claim, however, is that it underestimates the extent to which the value of many “distributive” goods, too, depends on social relations. This is because many distributive goods have positional aspects. ³¹ A good is positional if its value for its possessor depends on how the possessor's level of the good compares with others. A common example of this kind of good is the market value of education. ³² Having a master's degree will likely be much less valuable, in terms of the job opportunities it provides, if the majority of your competitors have greater qualifications than you do. On the other hand, a master's degree might be of greater value if your competitors have only a bachelor's degree or less. In each case, you have the same level of education, but its value to you in terms of the opportunities it provides is sensitive to how you compare with others. Resources have positional aspects as well, since they are worth more when others have less of them. There is, to be sure, a sense in which the value of resources is not positional. We can imagine people living on their own islands or planets and having equal amounts of fresh water, food, and so on. But this does not seem to capture the sense in which resources are relevant to theories of distributive justice. Many resources such as oil and gold have an entirely different value in a social world like ours, and as a medium of exchange, the value of any resource will depend in part on how scarce it is and how much others have.

Thus, while the distinction between “relational” versus “distributive” concerns may be intuitively appealing, it does not obviously divide the two approaches in the way some intend. Whether spelled out in terms of the relevant concern being constituted by social relations, or its value being sensitive to social relations, either way of understanding the distinction would count as “relational” a range of views commonly understood as “distributive.” I should note. however, that I do think the general strategy here warrants further consideration; and in what follows, I will propose that the distinctiveness of relational equality should be understood as involving concern with a particular kind of “relational” good. But before giving that account, it will be useful to consider a final suggestion.

Concern with institutional expression

Christian Schemmel offers another way to understand what it means to be concerned with social relations rather than (merely) distribution. In Schemmel's account, what is distinctive of the relational egalitarian approach is that it involves a concern with the expressive dimension of institutional action. This expressive dimension involves the kind of messages or attitudes that are communicated through the law and other institutions. ³³ By contrast, the distributive approach, Schemmel suggests, involves no such concern: it is “purely recipient oriented.” ³⁴

To illustrate the contrast, Schemmel asks us to consider a range of examples in which there is the same distribution of a good, but institutions express a different attitude in each case. Suppose that everyone in a society has access to excellent clean drinking water, except for a group that lives in a particular region (call it Dirty Water Region) who have barely potable water. ³⁵ In one case, the difference is due to the fact that there are natural pollutants in Dirty Water Region, and for whatever reason, it would be incredibly costly to remedy, such that there is no feasible policy that could do so. In the second case, the difference is again due to natural pollutants, but it would only be somewhat costly to remedy. The state, however, neglects to do so. In the third case, the difference is due to the fact that the residents of the Dirty Water Region are a particular religious minority, and for that reason, the region was chosen as a dumping ground for various contaminants.

This third case is the most objectionable because it involves an expression of contempt for a particular group, marking them out as inferior. The second case also seems objectionable, but less so. The state may express neglect or unequal concern in this case, but not the kind of stigmatizing contempt as in the third case. The first case, by contrast, may appear regrettable, but perhaps not unjust, expressing only a limitation of the state's resources. By expressing different attitudes across these cases, then, the state realizes different kinds of social relations; and intuitively, these matter for justice. It is more unjust, other things being equal, for the state to express contempt toward some of its members than (merely) neglect. ³⁶

I believe Schemmel is correct to point us to the importance of this expressive dimension, as I discuss below. Nonetheless, I do not think pointing to this dimension provides a fully satisfactory account of the distinctiveness of relational equality. This is because there is nothing in principle that prevents proponents of distributive equality from appealing to the importance of institutional expression as well. To see this, imagine a society in which there are two primary cultural groups. Now imagine two accounts that hold that equal resources are owed to everyone in the society. The difference between them, however, is this: on one account, equality of resources is owed on the basis of everyone being moral equals. On the second account, however, the state is explicit that one group is inferior to the other, but virtue dictates—as a matter of “noblesse oblige”—that the inferior group is given equal resources. Clearly, the second account is defective because it does not have the right kind of egalitarian institutional expression, and the first account is preferable in this regard.

But is the incorporation of the right kind of egalitarian institutional expression enough to regard the first view as involving a distinctive kind of relational equality? We might think that this view has incorporated an important relational egalitarian element, and so adopts a kind of pluralism. ³⁷ However, if all we could say is that relational equality involves incorporating this expressive dimension, it would seem that relational equality is more of an amendment to distributive equality—not a significantly distinctive concept that illuminates an important concern with social hierarchy. After all, the main concern is still with an equal distribution of resources, and not the quality of social relations in any substantial sense. This suggests, then, that we should still want to know whether there is more to be said about whether relational equality has a distinctive content, which I turn to next.

Civic status as a distinctive egalitarian concern

We’ve seen, then, that none of the accounts considered—whether spelled out in terms of justification, scope, or content—provide a fully satisfactory account of the distinction between distributive and relational equality. I now want to propose an alternative version of the content strategy. My proposal identifies a distinctive kind of concern with social hierarchy that is not captured by the accounts previously considered. On my account, the distinctiveness of relational equality is best understood in connection with the ideal of equality of status, where status is taken to be a kind of social ranking conferred by social, political, and legal institutions; and as we will see shortly, status understood in this way has distinctive and normatively significant positional and temporal aspects.

The central status relevant to social justice is the basic kind a person has as a citizen, or what I will call “civic status.” ³⁸ Civic status is distinct from moral status, which people may have equally however society is arranged. And it is distinct from what might be meant by “social status.” This qualification is important because it indicates that people need not be social equals in every dimension, since equality of civic status is compatible with inequalities of esteem or prestige, say, in arts or athletics. ³⁹ Civic status, in my view, has three aspects: one objective; another, expressive; and a third, intersubjective. ⁴⁰ Objectively, having a particular status (say, as a superior, inferior, or equal) depends on a person's possession of certain goods that are “status-conferring” in that they function to assign people a place or “rank” in here social hierarchy, significantly impacting how they see themselves and are seen by others. A person's enjoyment of the basic liberties, their level of income and wealth, as well as their access to goods like education and healthcare—all are plausibly of this kind. A chief task of relational egalitarianism is to identify which goods are “status-conferring,” and which distribution of them functions to rank people as equal citizens. An “equal distribution” of status-conferring goods may not require strict equality of all goods that confer status (e.g. income), but there will likely be significant limitations on the kinds of inequalities that are permitted.

Next, as we saw previously, relational equality, and so civic status, has an expressive dimension. An institution might distribute status-conferring goods on a variety of bases and so communicate a variety of messages. We could imagine, for instance, cases in which status-conferring goods are distributed equally on the basis that people are, luckily, regarded as equally talented, say, or equally intelligent. Alternatively, as suggested previously, we could imagine an equal distribution that expresses one group's fulfillment of “noblesse oblige.” Intuitively, these cases would fail to fully realize relational equality. What is needed, by contrast, is that status-conferring goods are given to people on the basis that they are antecedently equals. The equal distribution of status-conferring goods publicly communicates respect for persons as equals, manifesting an equal rank among them. ⁴¹

Finally, there is an intersubjective aspect of equality of status. It must be a matter of common awareness that people objectively enjoy an equal distribution of status-conferring goods, and that these goods are distributed on the basis that people are equals. By contrast, we might imagine a society in which the objective and expressive dimensions are realized, but there is little, if any, awareness among people that this is so. In this case, the objective and expressive dimensions of civic status will fail to structure people's social relations in any significant way. Thus, part of what it means to have a status as a civic equal is for there to be uptake or awareness—each knows they have an equal status with others, they know that others know they have an equal status with them, and so on. ⁴²

On this account, relational equality is not best understood in the first instance as a kind of interpersonal relationship, characterized by particular dispositions and attitudes. ⁴³ Instead, my account suggests we ask how people are given an equal status through the distribution of “status-conferring” goods along with the institutional expression that this distribution confers an equal rank, placing none above or below others in a fundamental way. This account is intended to be fairly general and open to development along several dimensions, as follows.

First, and of central importance, is the specification of the content of the status to be shared as equals. We could imagine two cases, one in which all social decisions are made by an algorithm without any input from society's members; and a second in which social decisions are made through well-functioning democratic processes. ⁴⁴ The second case, I take it, would be preferable. It might seem, however, that both cases involve people sharing a kind of civic status as equals (assuming that the algorithm is not under anyone's control). Likewise, we can imagine different specifications of equality of civic status that count different liberties as “basic” or are more or less permissive with regard to material distribution. ⁴⁵ Anderson's “democratic equality,” Rawls's “justice as fairness,” Pettit's “equality of nondominated status,” Schemmel's “liberal relational equality”—all might be regarded as different ways of specifying the nature of the kind of equal civic status required by justice. ⁴⁶ Choosing between these accounts—or, indeed, developing an alternative—will depend on an appeal to values and ideals outside of relational equality itself. ⁴⁷ For example, the relevant notion of status may be specified by appeal to a conception of the person—say, as a bearer of two “moral powers,” as in Rawls's view, which may be taken to delineate which liberties are “basic.” ⁴⁸

Second, a related issue concerns the relevant aspects of civic status. While the specification of civic status will in all likelihood involve concern with “distributive” goods such as the basic liberties, resources, and capabilities, it also directs our attention to less well-studied dimensions, such as norms of esteem and “microaggressions.” ⁴⁹ Relatedly, a further aspect of equality of civic status concerns its site. The realization of equal civic status certainly depends on a suitable basic structure. But one might argue it depends on suitable attitudes and dispositions beyond the basic structure as well. ⁵⁰

Third, the importance of equal civic status needs to be specified. ⁵¹ In my view, equality of civic status follows from the basic requirement that the government show respect for its members as equals. ⁵² If a government assigns some status to second-class or inferior citizens, whether explicitly or implicitly, it clearly fails to meet this requirement. Meeting this requirement, it is worth emphasizing, is not simply a matter of status-conferring goods being distributed equally. Recall previously the possibility that status-conferring goods could be distributed equally on the basis of a kind of “noblesse oblige,” where this distribution involves the message that one group deserves pity, and this is the received justification in the community. Intuitively, the members of this group would not be shown respect as equals. This suggests, then, that in order for government to show respect to its members as equals, the expressive and intersubjective dimensions of civic status must also be realized.

In addition, one might appeal to the personal value of having an equal civic status, and the personal disvalue of an unequal one. As some have suggested, equal social relations provide a basis of self-respect and encourage valuable relations such as civic friendship; by contrast, unequal social relations may undermine people's self-respect and encourage bad character traits such as deference or arrogance. ⁵³ On this suggestion, too, it is important that people have an equal civic status—in virtue of all three of the objective, expressive, and intersubjective dimensions being satisfied—and not just an equal distribution of status-conferring goods. The relevant threat to self-respect, for instance, is not due simply to an unequal distribution of goods but also involves the common awareness that one occupies an inferior place in society and is regarded this way by others.

With this general account of civic status in place, let us turn to the question of distinctiveness. How does civic status involve a distinctive kind of “relational” concern? In what follows, I show that civic status has certain positional and temporal aspects that are distinct from “distributive” goods such as resources or well-being. I will discuss each of these in turn and explain their significance.

Positional goods, as discussed briefly above, are such that their value to an agent depends on relative position. My claim is that civic status is a particular kind of normatively significant positional good. ⁵⁴ To explain this claim, it is worth pointing out that status-conferring goods may in fact have different kinds of positional as well as nonpositional aspects.

Consider education. Part of education's value is nonpositional: learning and gaining knowledge may be valuable to a person, regardless of how well-educated others are. But apart from this, we can distinguish two positional aspects. First is its labor market value, which may vary depending on how one's education compares to others: one stands a better chance of getting a job when one has higher qualifications than others, other things being equal. Second, the status-conferring value of education involves a different positional aspect. One's education (or opportunities for education) relative to others may serve to mark out their place in social hierarchy, impacting how they see themselves and are seen by others—in the way that attending, or being able to attend, an “elite” university, may function to rank a person as a social superior (at least in some contexts).

Similar observations hold for other status-conferring goods as well. Consider resources. Many resources, such as food or shelter, have nonpositional value. But as a medium of exchange, resources have a positional aspect: their value is impacted by what others have. Or consider healthcare: it certainly has nonpositional value, since being healthy is valuable independently of what others have. But in addition, healthcare also has comparative labor market value, similar to education: other things equal, being healthier gives one a leg up on the job market. ⁵⁵ Both resources and healthcare, too, have a status-conferring positional aspect, since the distribution of these goods may function to rank people in the social hierarchy.

Not only do status-conferring goods have various positional and nonpositional aspects; also, civic status is itself a kind of positional good. In particular, civic status is what I call a “position-defined” good. This means that a person's status is partly defined by a particular relative position: for example, as an inferior, superior, or equal. (It is only partly so defined because it is also defined, in absolute terms, by the particular bundle of status-conferring goods it involves). Other goods with positional aspects are typically not defined by relative position, but instead, their value is sensitive to relative position. Education, for instance, is not defined by a relative position; rather, the value of one's education is impacted by relative position. By contrast, a central part of what it is to have a particular civic status is to occupy a particular relative position. The kind of status that relational egalitarians care about is not simply a matter of each person possessing a certain level of status-conferring goods. Instead, the concern is that each person has publicly recognized status-conferring goods equally with anyone in society, giving all a basis to relate to any other as a fundamental equal.

Since civic status is position-defined in this sense, it has some significant features that set it apart from “distributive” goods such as well-being, resources, and capabilities. As a matter of definition, some holding a high civic status entails that others hold a low status. ⁵⁶ For this reason, eliminating positions of civic superiority does not count as “leveling down.” After all, some are made better off since they are no longer inferior. It may be, also, that no one is worse off in any morally relevant way. ⁵⁷ And for anyone to enjoy an equal status, it must be possessed by all. As a result, it is implausible to try to justify deviations from equality of civic status by appeal to personal responsibility, in the manner of luck egalitarian principles. ⁵⁸ It is also implausible to justify inequality of civic status insofar as it improves the situation of the worst off compared to what they would enjoy under equality of civic status, in the manner of Rawls's difference principle. The distribution of civic status best for the worst off is necessarily an equal one.

Consider, also, the notion of “relational sufficiency,” which requires that people are able to “relate as sufficients” or have “enough” of whatever relational goods are taken to matter. Kasper Lippert-Rasmussen suggests the idea of relational sufficiency can be illustrated by the following cases: “unequal relations between a worshipping religious follower and a sufficiently respectful guru … or a boss with suitably circumscribed powers and an employee with alternative employment opportunities. ⁵⁹ Similarly, Andreas Bengtson and Lasse Nielsen suggest examples of “the cultural elite [that] thinks of others as uncultivated, the highly educated assemble in intellectual communities in which less educated people are not allowed to participate, and the beautiful [that] look down with pity on the less attractive.” ⁶⁰ While these suggestions seem plausible, they do not motivate the notion of relational sufficiency with regard to civic status. The examples just mentioned are not cases of inequalities in civic status, and if they are judged to be innocuous, I contend, this depends on the existing against a background of equality of civic status. ⁶¹ In this way, then, civic status involves a distinctive kind of egalitarian concern in virtue of it being position-defined. Equality is the only plausible principle.

A further distinctive feature of equality of civic status is that it involves a different kind of temporal concern as opposed to equality of “distributive” goods such as welfare or resources. The orthodox view (for “distributive” goods) is that equality applies across people's complete lives. ⁶² This means that in judging whether two people enjoy distributive equality, we should not look at this or that particular moment, but at each of their lives as a whole. If people enjoy the same amount of whatever good summed across their life—aggregated intrapersonally—then they are equal in the distributive sense. This approach is problematic when applied to the position-defined good of equal civic status, however. To see this, consider how we might understand the notion of “complete lives relational egalitarianism.” Suppose, for example, that two people take turns serving one another, trading every so many years, such that each person spends the same amount of time being both a master and servant. ⁶³ We might think that since each person spends an equal amount of time being a master, and as a servant, then we can say that they enjoy the same amount of the relational good “status” over the course of their lives. Thus, they enjoy relational equality from the perspective of their complete lives.

This suggestion, however, also mistakes the nature of relational equality. To be sure, we could say that in this case there are some goods, such as power, that each person enjoys equally across their life. However, civic status is not the kind of good that can be aggregated intrapersonally in this way. The relevant concern is that people have publicly recognized status-conferring goods equally with others, giving all a basis to relate to others as fundamental equals. So, in the example, since no one at any point enjoys an equal civic status—after all, there is a master and servant at each point—then we cannot say people enjoy relational equality from the perspective of their complete lives. ⁶⁴ The point, then, as Julia Bidadanure has argued, is that relational egalitarianism involves a distinctive kind of synchronic concern, requiring that people enjoy equal social relations at each point in time. ⁶⁵

This is not to say, however, that there is no room for a diachronic perspective on the relational egalitarian approach. ⁶⁶ Consider the ways in which unequal relations of certain kinds are ubiquitous, as in those relations identified by Scheffler: “doctors to patients, teachers to students, parents to children, attorneys to clients, employers to employees, and so on.” ⁶⁷ If these relationships were permanent, we might indeed think them incompatible with relational equality. ⁶⁸ In most cases, however, these relationships are temporary, and part of why we may judge them to be compatible with relational equality depends on a background of enduring equality of civic status. In the case of (very young) children, it is not entirely clear that they should have an equal civic status—for example if they cannot—for good reason—vote; but it still matters that their status is not too unequal and that they are treated in ways appropriate to persons who will come to have an equal civic status. And even if young children are not given a strictly equal civic status, this may not be objectionable from a relational egalitarian standpoint, if each person enjoys equal civic status across the relevant diachronic period; that is, across their adult lives. ⁶⁹ Another way in which a diachronic perspective may matter for relational equality concerns our duties to the elderly suffering from or developing conditions such as Alzheimer's. ⁷⁰ A purely synchronic or “time-relative” view of relational equality may imply that we have overly demanding duties to ensure that such persons receive extensive and costly care to continue functioning as equal citizens. By adopting a diachronic perspective, relational egalitarians can suggest that these persons having functioned as equal citizens across their lives may reduce (but not eliminate) what is owed to them in these cases. In these ways, then, some aspects of relational equality may involve a diachronic perspective. Unlike diachronic distributive equality, however, it is not the case that concern is with a kind of “arithmetic” kind of equality across time, with each unequal social relationship being exactly compensated for at a different point of time. ⁷¹

Both of the distinctive aspects identified here—the positional and the temporal—make equality of civic status a particularly “strongly” egalitarian ideal. As Carina Fourie suggests, following T.M. Scanlon, a “strongly” egalitarian ideal is concerned with people's relative advantages and positions, and not how they fare in absolute terms. ⁷² But from the discussion above we can identify a further way in which an ideal might be strongly egalitarian: if it requires that people are equal a significant dimension at all times. Due to the position-defined nature of civic status, it is implausible that deviations from equality in this dimension can be justified; and it is likewise implausible to think that a loss of status in one stage of life can be compensated by a greater share of “relational goods” in another.

Conclusion

I have argued that none of the accounts surveyed on the distinction between relational and distributive equality is entirely satisfactory. In turn, I advanced a particular version of the claim that relational equality has a distinctive content. On the account put forward, relational equality involves a concern with equality of civic status. Relational equality, understood this way, has distinctive positional and temporal features which make it a particularly strongly egalitarian ideal.

One might still maintain that this account simply requires a “relational” good to be distributed equally, and so the “reductionist challenge” introduced at the outset still has bite. In reply, I am happy to grant that there is some force to this challenge and that the relational approach, understood in the way I propose, can be thought of as offering another answer to Sen's “equality of what?” question. As such, the difference between distributive and relational equality may be somewhat less deep than some relational egalitarians suggest. However, as I hope to have shown, civic status nonetheless possesses significant features that distinguish it as a distinctive target of egalitarian concern.