California Law Review
88 Calif. L. Rev. 119
LENGTH: 4018 words
1998-99 Brennan Center Symposium Lecture: Response to Commentators
NAME: Robert Post +
+ Alexander F. and May T. Morrison Professor of Law, School of Law, University of California, Berkeley (Boalt Hall); Law Clerk, Honorable Justice William J. Brennan, United States Supreme Court (October 1978 term).
... As this Response demonstrates, I have learned much from them, particularly regarding the logical relationship between the dominant conception and the sociological account. ... She ultimately raises the key question of how we can "account for the transformation of the stereotype within the practice of gender if there were not something else in gender, as it were, that is not immediately co-opted or foreclosed by the stereotype." ... First, although the concept of stratification makes a great deal of sense within the core areas of race and gender, there are other areas of antidiscrimination law where the concept has less bite. ... Second, even within the core domains of race and gender, the concept of stratification may not exhaustively explain the operation of antidiscrimination law. ... This is because the dominant conception expresses a moral vision that is internal to the law, while the sociological account represents an external account of the operation of law. ... A convincing description of the purposes of modern antidiscrimination law, as well as of the complex circumstances that inform its normative postulations, would be, as Siegel's Response and her historical work surely demonstrate, a hugely daunting task, albeit a necessary one. ...
Academic life contains few pleasures more intense than writing for an audience as astute and as generous as the four commentators to my Brennan Lecture. As this Response demonstrates, I have learned much from them, particularly regarding the logical relationship between the dominant conception and the sociological account. What I chiefly wish to stress, however, is my deepest gratitude for their constructive and discerning insights.
Tom Grey, for example, makes an excellent and useful point about the potential tension between the rule of law and what I call in my Lecture the "sociological approach." 1 Grey is entirely correct to stress that the insights yielded by a sociological apprehension of antidiscrimination law could encourage overly ad hoc and contextualized judgments, which the internal morality of the law ought properly to resist.
This tension between contextualism and formal consistency, however, is not entirely new to the law, and we are not without mechanisms for its ameliorization. In tort law, for instance, where the common law seeks to reproduce the nuanced and individualized judgments characteristic of community norms, we typically place decisions in the hands of a jury. Jury decisions are figured as "merely" factual, and hence as outside of the rule of law. Juries are deliberately rendered non-accountable and their decisions are stripped of precedential force. Nevertheless, we do not entirely abandon values associated with the rule of law, because judges, in the context of motions to dismiss or motions for judgment NOV, can simultaneously facilitate highly contextualized jury decisions and yet confine such decisions within definite limits. Of course juries must also receive judicial instruction about the standards and criteria that ought to govern their decisions, but judicial instructions are usually phrased at a level of generality that adroitly reconciles contextualism with formal consistency. 2 We have developed a multitude of legal mechanisms, therefore, for negotiating the messy tensions usefully identified by Grey, and perhaps these [*120] mechanisms might also prove helpful in the context of antidiscrimination law.
I also entirely agree with Anthony Appiah's perception that judgments of equality and inequality must ultimately reflect a genuinely moral perspective capable of convincingly justifying the differential treatment of persons. I was particularly appreciative of Appiah's incisive parsing of the idea of a "stereotype." My own use of the word was simple, resting on the elementary notion of "generalization." But Appiah usefully distinguishes among "statistical" stereotypes, "false" stereotypes, and "normative" stereotypes. His discussion reveals how these distinctions can clarify our apprehension of different cases.
My only reservation derives from the sense that sometimes the distinction between "statistical" and "normative" stereotypes does not inhere in the fact of the matter, but in its apprehension. Often when it is said that women are "weak" or that they lack "aggression," for example, it is not clear whether the assertions signify empirical, statistical claims, or whether they signify instead normative claims about what it means to be a woman. Stereotypes not infrequently hover ambiguously between these meanings, sometimes assuming the status of descriptive claims, and sometimes modulating into what Appiah elegantly terms "scripts for identities." 3 It is thus an achievement, for which Title VII law is partly responsible, to be able authoritatively to interpret many of these stereotypes as merely "statistical."
Judith Butler's Response ably exposes and deepens many of the philosophical themes I had hoped to advance. 4 She ultimately raises the key question of how we can "account for the transformation of the stereotype within the practice of gender if there were not something else in gender, as it were, that is not immediately co-opted or foreclosed by the stereotype." 5 She worries that
in claiming that race and gender are stable and systematic features of social reality ... and in claiming that persons cannot legibly appear to us without these conditions of social appearance in place, are we perhaps fortifying these categories precisely in their stereotypicality and persistence? By what means, then, are they disrupted and revised? 6
[*121] These are exactly the right questions to ask, and they have no easy answers. But in thinking about them I recur to the analogy of language. The meaning of words depends entirely upon common conventions and shared understandings - "stereotypes," as it were. But these stereotypes do not prevent us from transforming language. We can forge new meanings and alter the conventions by which we script the world. We can speak poetry instead of prose. But even poetry is expressed within language; it successfully communicates only because it is always already situated within the conventions by which meaning as such is constituted.
It is in this sense that I believe that stereotypes "underwrite" the practice of gender. They do not exhaust the practice, because any given stereotype is susceptible to change and transformation, in exactly the way that the meanings of words are susceptible to change. But just as meaning remains necessarily conventional, so too altering stereotypes merely revises, rather than eliminates, the content of stereotypes.
Of course gender or race could cease to be relevant categories, the way that serfdom has ceased to be a relevant category. But the category of serfdom vanished because the forms of life that made the category pertinent disappeared, and I doubt that there is any imminent likelihood of a similar social transformation about to overtake either race or gender. To the extent that these categories remain relevant, their meaning will be underwritten by the shared conventions, the stereotypes, that make social meaning possible. And this will be true however successfully we seek to unsettle and revise the received meanings of these stereotypes.
It is possible, however, that Butler's worry may be directed not at the generic properties of social meaning, but rather at the specific account of law implicit in my Lecture. It is true that that I speak of law as having specific articulable purposes. It can fairly be said that I characterize the law as speaking prose, rather than poetry. But this is because, as Robert Lowell writes,
the law is a sledgehammer,
not a scalpel. 7
The law as an institution is a crude, blunt instrument of public policy. In part this is a consequence of the law's function, which is to coordinate the decisions of thousands of government officials and to direct the actions of millions of private persons. In part it is a consequence of the internal morality of the law; the rule-of-law values invoked by Tom Grey point powerfully toward stability, predictability, and internal consistency. And in part it is a consequence of the fact that important aspects of antidiscrimination law have a purposive structure; they seek to attain definite objectives.
[*122] These factors combine to push antidiscrimination law toward an internal organization that strives to accomplish discrete, relatively clear, articulated ends. As a practical matter, therefore, to speak of the efforts of antidiscrimination law to transform the practices of gender and race is to speak of its efforts to alter these practices from one form to another. The ends of antidiscrimination law can and should change over time, which in practice means that the law will unsettle and progressively revise these practices. But it is implausible to imagine the law in any given moment as pursuing anything other than specific objectives.
Of course antidiscrimination law could take as its specific goal the radical suspension of the practices of gender and race, so as to reveal persons simply as persons, and this may be what Butler has in mind when she speaks of "unsettling the social conditions by which persons become intelligible at all." 8 Surely Butler is right to stress this admirable aspiration. Although every person comes to us already enmeshed within the linguistic significations that define personhood, and to that extent always already caught within conventional understandings, we have all had the experience of breaking through stereotypes to acquire a more immediate, vivid, and seemingly "truer" perception of a person.
But strange things happen to the law when it aspires to this condition. Something like this aspiration may well lie behind the trope of blindness that presently controls antidiscrimination law. What I attempted to demonstrate in my Lecture, however, is that this aspiration does not produce the kind of "unsettling" that Butler seeks; instead it modifies conventions of race and gender in ways that are implicit and unconsidered. Because gender and race remain socially salient categories, and because they are likely to abide despite the best efforts of antidiscrimination law, the trope of blindness within law does not so much unsettle these practices, as it redefines them. In her extensive Response, Reva Siegel trenchantly argues that the trope of blindness does not transcend, but instead actively entangles antidiscrimination law in the actual practices of race and gender. 9 Particularly illuminating is her discussion of the elastic and permeable ways in which contemporary antidiscrimination law classifies actions and rules as "based upon" race or gender. These are significant doctrinal and ideological mechanisms by which antidiscrimination law regulates its own recognition of the manifold ways in which it intersects with the social practices of race and gender.
Reva Siegel has, of course, written extensively and profoundly about the history of race and gender. I have learned much from her pungent and subtle accounts of the actual operation of antidiscrimination law. In her [*123] Response, Siegel argues that race and gender must be understood as conditions of social stratification, and that antidiscrimination law has historically intervened to dismantle some, but not all, of these conditions. Through the trope of blindness, antidiscrimination law has paradoxically served to validate certain forms of stratification.
Siegel's argument seems to me largely correct, and I endorse it. But I would reserve three points of qualification. First, although the concept of stratification makes a great deal of sense within the core areas of race and gender, there are other areas of antidiscrimination law where the concept has less bite. The domain of religion, for example, is at least as venerable as those of race and gender, yet stratification does not seem a particularly useful way to understand contemporary American prohibitions against discrimination on the basis of religion. Appiah's invocation of the problem of "bigotry" 10 seems more pertinent. This would also seem to be true with respect to other areas of antidiscrimination law, such as sexual orientation or Vietnam Veteran status. There are still other areas of contemporary antidiscrimination law, however, such as age or marital status, where neither stratification nor bigotry seem especially relevant. While stratification may be a useful frame for understanding important aspects of antidiscrimination law, therefore, it is not comprehensive.
Second, even within the core domains of race and gender, the concept of stratification may not exhaustively explain the operation of antidiscrimination law. While I agree that the concept does illuminate exceedingly important dimensions of the relationship of antidiscrimination law to race and gender, there may remain other facets of this relationship that are not so centrally elucidated by the concept. Aspects of Title VII's upholding of conventional dress and grooming codes, for example, may be better explicable in terms of maintaining heterosexist norms than in terms of stratification between the sexes.
Third, and most importantly, Siegel's observations concerning stratification suggest an important ambiguity about what is meant by the "purpose" of antidiscrimination law. I should have stressed more clearly than I did in my original Lecture that the dominant conception and the sociological account are not symmetrical concepts. They cannot be substituted, one for the other. This is because the dominant conception expresses a moral vision that is internal to the law, while the sociological account represents an external account of the operation of law. 11 The dominant conception expresses a "purpose" of the law, but the sociological account does not.
[*124] As the work of John Rawls indicates, the trope of blindness has deep roots in contemporary theories of justice and of morality. If one were to ask a judge why the law ought to impose a regime of color blindness, she could reply that it was right and proper for the law to act in ways that render race invisible. The ideal of color blindness expresses a genuine moral commitment; it is neither a pretext nor a ruse. That is why the dominant conception functions, as Siegel accurately observes, affirmatively "to legitimate the distributive regimes that sustain group stratification." 12
My critique of the dominant conception is that it contains an image of how antidiscrimination law works that systematically masks and distorts the true operation of the law. This observation is relevant to the question of whether the dominant conception represents a desirable or adequate ideal for antidiscrimination law. I proposed the sociological account in order to provide an external perspective that would make visible how antidiscrimination law actually functions, and in this way to reveal the inadequacies of the dominant conception. My suggestion was that we ought seriously to consider whether to refashion antidiscrimination law to pursue a different internal ideal, one which takes account of and does not contradict the insights of the sociological account.
The sociological account cannot by itself provide such an ideal. Although the shift to an external perspective helps us better to understand the true effects of our internal purposes, it cannot simply be substituted as an alternative internal ideal. If a judge were asked to explain her decisions, she could not respond that she was engaged in a social practice that modified another social practice. A judge would instead have to articulate an internal justification for antidiscrimination law that was morally convincing, that acknowledged the actual interplay between such law and the social practices of race and gender, and that was defensible in terms of the particular requirements of legal doctrine. Brennan's rationale in Weber of breaking down "old patterns of ...segregation and hierarchy" might, for example, be a candidate for such a reformulated ideal. 13
The objective for antidiscrimination law that Brennan proposes in Weber belongs to the class of proposed internal purposes that we would now call "antisubordination" theories. Siegel eloquently summons the historical genesis and normative attraction of these theories, which challenge us to dismantle hierarchy, disadvantage, and subordination. 14 As I suggested in my Lecture, antisubordination theories do not seek to transcend history; they instead conceive law as an institution acting to modify other [*125] institutions. Unlike the dominant conception, therefore, they do not contradict the very premise of the sociological account.
But the sociological account is nevertheless relevant to the assessment of antisubordination theories. The power of such theories often flows from the implicit presumption that there exist independent metrics of hierarchy or disadvantage, like work or income, which it is the task of antidiscrimination law to equalize. Although there is surely some truth to this presumption, it is also clear that in its actual operation antidiscrimination law does not function in this purely redistributive fashion, even in the relatively progressive areas of disparate impact analysis and affirmative action. 15
Moreover antidiscrimination law seems in many contexts as much concerned with defining what should count as hierarchy or disadvantage as it is with applying any antecedent metric. Antidiscrimination law postulates normatively acceptable social meanings for race and gender, and these meanings, in turn, color what we perceive as subordination. The area of sexual harassment law is a particularly fascinating venue for the study of this dynamic. 16 In such contexts, notions of hierarchy and disadvantage do not so much measure the practices of race and gender against independent variables like work and income, as they express conclusions about what the proper nature of these practices ought to be.
This suggests that a sociological account of the actual operation of antidiscrimination law will prove important for our understanding of antisubordination theories. A convincing description of the purposes of modern antidiscrimination law, as well as of the complex circumstances that inform its normative postulations, would be, as Siegel's Response and her historical work surely demonstrate, a hugely daunting task, albeit a necessary one.
Of course whatever we learn from such a sociological account is not set in stone. Even if antidiscrimination law is not presently redistributive, for example, we can change our purposes and hence transfigure the actual practices of antidiscrimination law. We might in fact most accurately understand antisubordination theories as proposing just such a normative transformation. They remind us that we are not bound to repeat the limitations of our history. But in evaluating the promise of such theories, surely an essential (if modest) first step would be to apprehend the nature of our own current commitments, as they stand revealed in the actual operation of our law. It would seem that our capacity for firm and enduring change would depend upon the potential for such self-knowledge.
[*126] In this context, I would interpret Siegel's concept of stratification as an exceedingly useful amplification and elaboration of the sociological account. It offers a complex and systematic framework for apprehending the subtle ways in which antidiscrimination law modifies the actual practices of race of gender. It enables us to grasp in a deeper and more comprehensive way the impact of the dominant conception, and hence empowers us more accurately to evaluate that conception. But, like the sociological account, it is not itself an internal moral vision capable of guiding the law. As Siegel herself acknowledges, it is "messy" to apply the concept of stratification because "it is not always clear what equality looks like." 17
In my Lecture, I deliberately bracketed the question of the internal principles that ought to guide antidiscrimination law. I instead offered the modest observation that there were important advantages to formulating these principles in a manner that was cognizant of the insights that an external perspective, like the sociological account, could bring to bear. Particularly in light of Appiah's and Siegel's Responses, however, I should add that, whatever the substance of these principles, they ought to reflect genuine moral commitments. They ought to flow from a convincing social vision that identifies which aspects of the social practices of race or gender ought to be changed, and why. Although such principles would no doubt be strengthened if formulated in light of the insights provided by Siegel's concept of stratification, they can not be reduced to that concept. Stratification is a term of descriptive analysis; it is not a moral purpose.
I suspect that if courts were to take seriously the insights of the sociological account, they would recognize that antidiscrimination law serves different purposes in different contexts. If one compares the role of stratification in gender and in sexual orientation, for example, it is evident that antidiscrimination law must combat very different social phenomena in these two domains. It is plausible to expect these differences to affect the moral vision pursued by the law. Tom Grey perhaps puts the matter best when he notes that in articulating the objectives of the law we ought to "take account of the nature of the enemy." 18
For related research and practice materials, see the following legal topics:
Civil ProcedureJudicial OfficersGeneral OverviewLabor & Employment LawDiscriminationDisparate ImpactRemediesLabor & Employment LawDiscriminationGender & Sex DiscriminationCoverage & DefinitionsSexual Orientation
n1. Thomas C. Grey, Cover-Blindness, 88 Calif. L. Rev. 65, 68-69 (2000).
n2. See, e.g., Robert C. Post, The Social Foundations of Privacy: Community and Self in the Common Law Tort, 77 Calif. L. Rev. 957, 959-78 (1989).
n3. K. Anthony Appiah, Stereotypes and the Shaping of Identity, 88 Calif. L. Rev. 41, 51 (2000).
n4. The query that she puts to me - how to account for "those dimensions of personhood that do not, strictly speaking, appear" - is pertinent and important, and certainly beyond the confined scope of these comments. My guess is that I would want to think about such questions in roughly the way that Wittgenstein thinks about the issue of pain.
n5. Judith Butler, "Appearances Aside," 88 Calif. L. Rev. 55, 60 (2000).
n6. Id. at 62.
n7. Robert Lowell, Fetus, in Day by Day 34 (1977).
n8. Butler, supra note 5, at 63.
n9. See Reva B. Siegel, Discrimination in the Eyes of the Law, 88 Calif. L. Rev. 77, 83-84, 87-93 (2000).
n10. Appiah, supra note 3, at 50.
n11. For a discussion of the distinction between internal and external accounts of the law, see Neil MacCormick, Legal Reasoning and Legal Theory 275-92 (1978).
n12. Siegel, supra note 9, at 83.
n13. United Steelworkers of Am. v. Weber, 443 U.S. 193, 208 (1979).
n14. Siegel is, I think, correct to trace the normative structure of these theories to Fiss's "group-disadvantaging principle." Owen M. Fiss, Groups and the Equal Protection Clause, 5 Phil. & Pub. Aff. 107, 108 (1976); Siegel, supra note 9, at 109-10.
n15. See, e.g., Barbara J. Flagg, Fashioning a Title VII Remedy for Transparently White Subjective Decisionmaking, 104 Yale L.J. 2009, 2030-31 (1995).
n16. For a fine discussion, see Katherine M. Franke, What's Wrong with Sexual Harassment?, 49 Stan. L. Rev. 691 (1997).
n17. Siegel, supra note 9, at 115.
n18. Grey, supra note 1, at 70.