Rights that vest on the basis of an individual’s membership in a particular social or cultural group are an increasingly common aspect of modern liberal legal systems. Such rights have been granted to individual members of a broad array of social groups to remedy inequities associated with, for example, the members’ race, sexual orientation, gender, age, economic or disability status. Rights similarly have been afforded to individual members of cultural groups constituted according to nationality, ethnicity or religion, to acknowledge and accommodate particular beliefs or practices, or in recognition of collective claims to self-government or property. Yet since group-differentiated rights openly distinguish among classes of persons in the distribution of social benefits and burdens, this form of right has long been the source of significant controversy within liberal political theory.
This essay (and the longer paper from which this essay is derived) concerns a moral cost that I believe should be more strongly associated with efforts to evaluate group-differentiated rights. Liberalism is committed at its deepest levels to individual constitutive autonomy.1 And yet group-differentiated rights, through the processes of legal inclusion and exclusion, sort and construct individual human beings as members of particular social groups. As such, group-differentiated rights threaten to impede individual self-invention.
The qualification in the previous sentence is essential, however. We should ask merely whether rights explicitly conscious of differences among groups threaten to obstruct self-invention because such rights, and the groups of which they take note, are critically diverse. At times, the social differentiation inherent in legal rights will constrict constitutive autonomy. At other times, group-differentiated rights may enhance the individual’s capacity for self-construction. Moreover, even where constitutive autonomy is so constrained, the advantages engendered by group-differentiated rights may exceed the costs. Indeed, the primary descriptive and normative contentions in this paper center on the notion that the group-differentiated form of right in fact subsumes three importantly distinct models of legal rights, and that a fair evaluation of the form entails a contextual assessment.
The first part of the longer paper describes the debate over group-differentiated rights, with special attention to concerns raised from the perspective of liberal neutrality. Part II describes analytically the processes by which rights categorize persons and influence social identity. Part III constructs a liberal conception of membership, the central precept of which is individual constitutive autonomy. Part IV then depicts three models of the process by which group-differentiated rights construct social groups, and evaluates each contextually, according to the principles that inform the liberal conception of membership previously constructed. The analysis demonstrates that while the group-differentiated form of right does indeed imperil constitutive autonomy, the true extent of any loss is, in significant part, contingent upon the particular model invoked.
II. Rights and Membership
A. Rights and Categorization
Rawls tells us that the concept of justice concerns the proper distribution of benefits and duties across persons in society.2 The concept of justice provides us with a formula, the formula of formal justice, the purpose of which is to arrive at an appropriate balance among competing claims: like cases are to be treated alike, and different cases are to be treated differently. Notice, then, the effect that the application of this precept will have upon a society and its individual members. For as we distinguish relevant similarities and differences among individual rights-claimants we simultaneously categorize persons according to those characteristics deemed relevant. As a consequence, any legal right will engender at least two sets of persons: a set of rights-bearers and also a set (or sets) composed of persons excluded from the right.
The notion that legal rights necessarily engender sets of rights-bearers distinguished from sets of mere rights-claimants is true both of rights that arise in virtue of legislation and of rights that arise as a result of adjudication. First, with respect to legislation, the investitive conditions of rights in legislation – the conditions which describe when a given right becomes pertinent to an individual – are deemed to have been drafted justly only when drafted according to a universalizable format. In consequence, an individual can be a legislative rights-bearer only in relation to some more general set of rights-bearers. Each member of the resultant set would be distinguishable from persons excluded from the right in virtue of the characteristic(s) indicated by the right’s investitive conditions. So, for example, a legislative right granting, say, landowners the authority to exclude persons at will and by force from their land would engender a particular class of rights-bearers, here a class of landowners, described by the right's investitive condition, namely that of owning land.
Second, and quite unlike legislation, adjudicatory processes may engage merely private parties to seemingly isolated disputes. Nevertheless, any right announced as a result of such an adjudication takes on the character of a rule of law and so engenders a prospective set of rights-bearers otherwise indistinguishable from that generated by rights in legislation. Any later rights-claimant possessing the criteria deemed investitive in the previous adjudication must be accorded the status of rights-bearer and so membership in the class of rights-bearers engendered by the legal right. Indeed, any contrary result would abandon the precept of legal generality that similar cases be treated similarly. Formal justice thus induces the categorization of persons in virtue of the legal rights which individuals may claim. And this will be true both of rights in legislation and adjudication.
B. Rights and the Construction of Social and Cultural Groups
Now, the classes engendered by rights in legislation and adjudication may be nothing more than formal sets. That is, they may be mere aggregates of individuals, as opposed to fully constituted social or cultural groups. What then distinguishes mere sets from groups? “A social group,” as described by Iris Marion Young, “is defined not primarily by a set of shared attributes, but by a sense of identity.”3 Clearly, though, most rights are not accorded on the basis of characteristics deemed especially critical to the rights-claimant’s social identity. For example, if Valdez has extended credit to Smith, and so now has a valid legal claim for payment, that right will only barely affect our perceptions of who Valdez is. Yet the grant or denial of the rights of citizenship to either Smith or Valdez might powerfully impact social and personal perceptions of these individuals. Often, then, it will be the case that no meaningful social group is affected by the grant or denial of a right. But in any case where the investitive criteria invoked by a right cohere in a meaningful sense with an aspect of the rights-claimant’s social identity, a social group will indeed be constituted in virtue of the grant or denial of the right in question.4
On one hand, then, an individual’s status as a member of a social group described by a legal right may constitute a meaningful aspect of that individual member’s identity. And yet, on the other hand, an essential commitment of liberal justice is to individual human agency in the construction of one’s self. Hence, from a liberal perspective special attention is warranted because to the extent persons are categorized, and identities are constructed, by law, rather than in accordance with individual volition, we face a potential loss of constitutive autonomy.
III. Liberal Membership
Membership is often uncritically conceptualized as an exclusively communitarian ideal. Yet while a communitarian conception of membership may be well engrained in our literature, there is nothing necessarily (indeed, there is necessarily nothing) communitarian about the concept of membership.5 In the longer paper, I discuss two prominent such conceptions. Michael Walzer, I suggest, writes of membership as a distributive policy of communal inclusion and exclusion according to contingent internal norms. Michael Sandel, focusing on an intrinsic relationship he perceives between community and individual identity, theorizes membership as constitutive attachment. From these prominent conceptions, I then attempt to discern a more abstract concept of membership. Membership, I suggest, is a good, distributed exclusively to a particular class of persons, that describes an aspect of each member’s social identity.
In what then consists a liberal conception of membership? In liberal thought, since at least Kant and Mill, human beings are conceptualized foremost as fundamentally equal and autonomous agents. Liberalism holds that individuals must remain free, consistent with an equal freedom for all, to act in accordance with their own sense of what is good and valuable in life. An autonomous person is a maximally self-defined person, a person free in virtue of one’s own successive choices to determine one’s own life path. As such, on the liberal view, individuals must be free to the greatest possible extent to constitute and rationally to revise their own sense of the person they shall become.6
At the same time, membership within the broader polity is often a precondition for the exercise of rights, and thus for access to the sorts of opportunities and personal liberties which enable liberal self-invention. As a general matter, then, inclusion itself is often (at least) instrumentally conducive to liberal norms.7 What this means for present purposes, though, is that liberal membership must countenance a certain degree of involuntary construction of the self. For with inclusion, comes definition; persons will at times be included, and to that extent partially defined, specifically in order to secure to such persons genuine equality of treatment and the capacity for further self-invention.
Further, included among the many choices individuals must (to the extent possible) freely make are the social attachments they will maintain, sever and newly establish. This is why official recognition and accommodation of diverse cultural attachments is increasingly recognized as essential to liberal individualism; cultural associations provide the critical “contexts of choice” within which we as individuals may define and revise aspects of our selves.8
Clearly, then, in assessing group-differentiated policy, a liberal conception of membership must mediate among diverse and at times conflicting demands – demands for individual constitutive autonomy and demands for cultural toleration, demands for broad inclusion in, and demands for self-exclusion from, various social groups. Below I suggest three forms that these group-differentiated demands might take. These models should prove useful both in describing the nature of group-differentiated rights and in assessing the extent to which such rights are consistent with a liberal conception of membership.
IV. Three Models of Group-Differentiated Rights
Any right granted will of necessity be granted to a class of persons. Rights-claimants able to demonstrate sufficient congruity between their own particular circumstances and the criteria indicated by a right’s investitive conditions will be included in a class of rights-bearers. Rights-claimants who fail to meet such investitive criteria, and so are deemed in some important respect dissimilar from those entitled to exercise the right, are thereby excluded from the class. Where the dissimilarity between the classes of persons included and excluded from the right is founded upon an ascribed characteristic (e.g., a moral or intellectual trait associated with the right-claimant’s race or gender), both the inclusion and the exclusion will result in the construction of social groups. Membership in such an ascriptive social group will contribute both to social and self-perceptions of the individual members’ identities.
Consider, for example, the evolution of the class of persons granted the right to vote in the United States. At the founding, American suffrage remained conditioned on state imposed landed property qualifications, usually combined with specific legal exclusions for persons who were not free, white, male and often adherents of a particular religion. Each such categorical exclusion was justified on the basis of inegalitarian ascriptive assumptions regarding the excluded persons’ race, culture, gender, religion or economic status.9 Persons within these categories, it was thought, could not possibly possess the moral, civic and intellectual traits required of the electorate. As a result, each such person was excluded from membership in the American political class. At the same time, persons thus ascriptively excluded were simultaneously included in a social group (or groups) constructed according to the characteristic (race, gender, etc.) that served as the basis for the assumed incapacity. And membership in each such social group, together with its inegalitarian ascriptive subtext, would thus come to define an aspect of each individual member’s identity. Further, on the opposite side of this rights equation, the categorical inclusion of freeholders was justified on the basis of similarly inegalitarian, though now more sanguine, ascriptive assumptions regarding such persons’ intellectual and moral capacities and civic propensities. And, hence, membership in this more favored social group would come then to constitute an aspect of each individual “citizen’s” social identity.
Further examples of ascriptive differentiation in legal rights abound. In the Dred Scott case, individuals of African descent were ascriptively excluded from the class of constitutional persons deemed competent to bear rights, with obvious effects on social and self-conceptions of such “non-persons.”10 Disabled individuals too were long subject to ascriptive classification as non-persons for constitutional purposes.11In the private law context, married women were in most states deemed incompetent to contract or possess property, reflecting the strikingly ascriptive notion that a woman might cease to possess a separable identity upon marriage.
In each of these cases, individuals were sorted, and aspects of human identity defined, by law on the basis of inegalitarian ascriptive criteria. The loss in constitutive autonomy in such a context is at an extreme. Of course, such cases are now far less common in modern liberal democratic legal systems. Yet the constitutive influence of such laws is instructive in assessing two other, currently more prevalent, forms of differentiated citizenship. Both rights that seek specially to affirm the status of individual group members, and rights that would enable members of cultural groups to self-exclude, share with ascriptive rights a formal resemblance and an influence on their claimants’ social identities. And yet in each case the moral calculus is importantly different.
At times a right will be granted only to a particular subset of persons as part of an attempt to reverse the inegalitarian consequences of a previous ascriptive exclusion, or to remedy the exclusionary effects of social practices other than law itself. And yet even this effort, an effort ultimately to include otherwise subordinated persons, will of necessity result in the legal system excluding a category of persons. Although formally similar, this type of legal exclusion is of a substantively different nature than the ascriptive exclusionary process discussed above. The exclusion of a dominant class from a right granted to oppressed persons is justified not on the basis of negative inegalitarian characteristics ascribed to the excluded class, but rather on the ground that the particular remedy afforded the included class will be conducive to genuine equality of treatment for all. In such a case, the exclusion may function affirmatively to include a category of persons in need of special protection.
Mention of the affirmative form of group-differentiated right will bring immediately to mind disputes over the legitimacy of programs of affirmative action. And, indeed, a right to affirmative action on the basis of, say, past or continuing racial discrimination, clearly would fall within this category. Other examples of such rights of affirmation considered in the longer paper include the recently passed Vermont Civil Unions and Reciprocal Beneficiaries Law, which in an effort to protect the interests of same-sex couples constitutes a new form of legal relationship, the “civil union,”12 and the Americans with Disabilities Act, which grants rights against discrimination specially to disabled persons.13
As in the ascriptive exclusionary cases described above, group-differentiated rights of affirmation sacrifice not only formal equality of treatment but also, potentially, individual constitutive autonomy. This is so because the bearers of such rights are sorted and defined by law as members of particular social groups. At the same time, this does not mean that such rights are necessarily inconsistent with a liberal conception of membership, although presumptively they may be so. Whether the presumption may be overcome will depend on whether the moral cost in constitutive autonomy is sufficiently offset by the advantage inclusion brings. Thus, while the right to form a civil union, as opposed to a marriage, perpetuates legal and social differentiation, the law also importantly grants to same-sex couples all legal benefits to which married couples are entitled. While rights to affirmative action may further construct social perceptions of the members of subordinated social groups, such rights also afford their bearers precisely the sorts of opportunities necessary more fully to construct their identities in other spheres. Indeed, group-differentiated rights of affirmation are inherently constitutive, but they are also often among the most crucial to the realization of justice for oppressed persons.
In contrast both to rights that ascriptively constitute individuals as members of social groups, and rights that seek affirmatively to include members of subordinated groups, are rights that permit members of cultural groups, or the groups themselves in a collective capacity, the freedom to exclude themselves from some aspect of social life. Unlike ascriptive and affirmative rights, such culturally-differentiated rights often directly foster constitutive autonomy by enabling cultural group members to construct their own particular social identities. In the United States, religious conduct exemptions constitute perhaps the most prominent example of this form of group-differentiated right. Such exemptions permit cultural group members to differentiate themselves from broader social groups, and in the process to constitute further their particular identities.
A dilemma, though, arises when the collective constitutive autonomy secured by culturally-differentiated rights is used by elites within minority cultural groups to subordinate more vulnerable members. Indeed, at times the nomos or traditional precepts of certain cultures explicitly prescribe repression of individual members and internal sub-groupings, most particularly women.14 Where this is the case, the moral costs apparent in the ascriptive differentiated citizenship model are merely revived at a different level. Cultural self-exclusion often facilitates the capacity to self-define that is so central to a liberal conception of membership. But that constitutive capacity loses its liberal coherence when aspects of individual identities are constructed involuntarily, as they are when cultural groups use accommodationist policy to subjugate vulnerable members.
All legal rights categorize claimants. Group-differentiated rights do this and more. Such rights serve also further to construct aspects of human identities. Yet, the group-differentiated form of right in fact subsumes three distinct models of rights, and each model bears a different relationship to a liberal conception of membership. Sorting out the moral implications of this form of right will thus require a renewed focus both on the significance of individual autonomy in the construction of the self and on the particular social contexts in which such interests ordinarily arise.
1* Assistant Professor, Thomas Jefferson School of Law; Doctoral Candidate, Department of Politics, Princeton University. A.B., Cornell University; J.D., University of Michigan; M.A., Princeton University.
“[I]t is the privilege and proper condition of a human being . . . to use and interpret experience in his own way . . . He who lets the world, or his own portion of it, choose his plan of life for him has no need of any other faculty than the ape-like one of imitation.” John Stuart Mill, On Liberty 122-23 (Gertrude Himmelfarb ed., 1974, 1859).
2 John Rawls, Theory of Justice 5-10 (1971).
3 Iris Marion Young, Justice and the Politics of Difference 44 (1990).
4 The longer paper pauses here to explore constitutive methodology.
6 See, e.g., John Rawls, Reply to Alexander and Musgrave, in Collected Papers 232, 240 (1999, 1974) (“[F]ree persons conceive of themselves as beings who can revise and alter their final ends and who give first priority to preserving their liberty in these matters.”).
7 This disposition toward inclusion is reflected in classical liberal thought as well, as for instance in the idea that legitimate government is founded only upon the consent of the governed, see John Locke, SecondTreatise, ch.VIII, and in the Rousseauian notion that law’s claim to obedience rests upon one’s having participated in its making, see Jean Jacques Rousseau, On the Social Contract, Bk.II, ch.6.
8 See, e.g., Will Kymlicka, Liberalism, Community and Culture 166 (1989).
9 See Rogers M. Smith, Civic Ideals: Conflicting Visions of Citizenship in U.S. History (1997).
10 See Scott v. Sandford, 60 U.S. (19 How) 393 (1856).
11 See Martha Minow, Inclusion, Exclusion and American Law 101-120 (1990).
12 15 Verm. Stat. Ann. §1201.
13 42 U.S.C. §12182.
14 See Ayelet Shachar, Multicultural Jurisdictions: Cultural Differences and Women’s Rights (2001).