Duke Journal of Gender Law & Policy: Barbara J. Flagg, An Essay For Keisha (And A Response To Professor Ford), 14 Duke J. of Gender L. & Pol'y 179 (2007)

Cited: 14 Duke J. Gender L. & Pol'y 179
[*pg 179]

AN ESSAY FOR KEISHA (AND A RESPONSE TO PROFESSOR FORD)

BARBARA J. FLAGG^*

I. PROFESSOR FORD'S CRITIQUE

II. KEISHA IN CONTEXT

III. RESPONDING TO PROFESSOR FORD
   A. Specific References
   B. Generic References
   C. Refashioning Keisha's Claim

FOOTNOTES

Some years ago, for a law review article, I created a character named Keisha Akbar.¹ The article had to do with Title VII; Keisha made her appearance in order to illustrate one pervasive form of employment discrimination. In the end, the article set forth two alternative doctrinal interpretations of Title VII, either of which would provide a legal remedy for the particular form of race discrimination that the fictional Keisha had suffered.

Keisha has reappeared in print a number of times in the past decade, twice in other works of mine,² and several times in the works of other scholars.³ In general, both Keisha and her claim have been treated well in these publications, even when their authors disagree with me on one or another point of Title VII law. Now, however, Keisha's claim has come under a more sustained attack, at the hands of Professor Richard Ford. Initially in several articles,⁴ and more recently in a book titled Racial Culture: A Critique,⁵ Ford has elaborated a thesis questioning the utility for racial justice of at least one version of multiculturalism.

It seems that for Ford the claim I fashioned for Keisha is an example of a "rights-to-difference" proposal, a particular variety of legal recommendation that is an instance of what he calls a larger "difference discourse." The latter, Ford asserts, "has the potential to stall egalitarian and humanist reforms and deprive us of much of the exciting potential of cosmopolitan society."⁶ Under Ford's analysis, Keisha could be harmed more than helped by my Title VII remedy.

[*pg 180]

I have two somewhat contradictory reactions to Ford's arguments. On the one hand, I regard his presentation of my views as infuriatingly misleading. I don't think there is much fit at all between his account of "rights-to-difference" proponents and my work. On the other hand, I do find many of his observations concerning multiculturalism illuminating. I especially take his point that norms of cultural pluralism may not be the best foundation on which to build remedies for racial subordination. It has prompted me to rethink the normative underpinnings of Keisha's claim.

Hence this essay has two objectives. The first is to respond to Professor Ford, insisting on some points that I think were set forth clearly enough in my Title VII article that they should not have been misread, and clarifying others that I now think might have been presented more intelligibly. The second objective is to purge Keisha's claim of its reliance on a norm of cultural pluralism and see what remains. It turns out, I think, that to whatever extent one sees my "models of Title VII liability" as doctrinal proposals, both emerge from the process of normative reconstruction pretty much unscathed.

The essay proceeds in three parts. In the first I describe the salient points of Professor Ford's critique,⁷ and in the second I review my Title VII article. In both of these sections I provide extensive quotations, for reasons that will become apparent. The final section responds to Ford, first by comparing his claims about my work with the work itself as I see it, and then by undertaking the project, for which he has provided the impetus, of normatively refashioning Keisha's claim.

I. PROFESSOR FORD'S CRITIQUE

My overview of Ford's project begins with his own description of his book:

Chapters 1 and 2 describe the set of ideas, policies, and literature I wish to attack and introduce the book's central thesis: Group cultural difference -- racial culture -- is a social discourse that produces rather than describes preexisting group cultural differences. I argue that the "recognition" of group difference is not an antidote to the forced assimilation of distinctive social groups. Instead the two -- recognition of difference and forced assimilation -- are both part of a single mechanism of oppression, a whipsaw dynamic that effectively produces and punishes group difference, making assimilation both mandatory and unavailable . . . .

Chapter 2 argues that racial and analogous social identities are never autonomously adopted or intrinsic; they are always, at least in part, the effect of this social regulation. . . . Chapter 2 suggests that, instead of presumptively protecting the traits associated with groups that have suffered from pervasive social and political oppression (which would reinforce the regulation that produced the identities in the first place), anti-discrimination law should only protect individuals from discrimination based on ascribed group membership or status. I conclude that "cultural difference" conflicts, which concern traits or conduct, are distinct from social struggles concerning racial and other ascribed statuses.

[*pg 181]
In chapter 3 I build on this conclusion and argue that political solidarity based on a common relationship to oppression and domination is the appropriate focus of (racial) identity politics and legal rights assertion; by contrast cultural claims are more contestable on both descriptive and normative terms and should be left to more fluid domains of conflict resolution such as social dialogue, the democratic process and the market economy . . . .

Chapter 4 addresses the limits of civil rights law with respect to cultural pluralism and group identity and suggests a direction for reform of anti-discrimination law consistent with my arguments above. The appropriate goals underlying anti-discrimination law are to counter the social and economic isolation of historically subordinated groups, which is produced by the ascription of stigmatized status. Rights-to-group cultural difference are of questionable legitimacy at best when evaluated in this pragmatic, consequentialist manner. The chapter ends by suggesting approaches to cultural difference and the expression of individual identity that would avoid the pitfalls of difference discourse.⁸

Thus, Ford presents both a positive and a negative program: On the negative side, Ford criticizes proponents of "difference discourse," defined as "a series of claims, proposals and practices" that set forth, or undergird, "a set of moral and legal arguments that promote what [Ford calls] 'rights-to-difference.' These arguments hold that a just society could and should prohibit discrimination on the basis of the [sic] cultural difference (thereby establishing a 'right-to-difference') for the same reasons it should prohibit discrimination based on statuses such as race."⁹ On the positive program side, Ford sets forth his preferred vision of multiculturalism and advances his own views on the appropriate contours of antidiscrimination law. In my opinion, the two programs are not equally successful; while Ford makes a number of solid and useful points on the positive side,¹⁰ his "critique" of the proponents of "difference discourse" and "rights-to-difference" is little more than a disappointing caricature.

Ford's discussion of "difference discourse" begins with a quotation from the district court opinion in Rogers v. American Airlines,¹¹ in which Rene Rogers, an employee of American Airlines, challenged her employer's grooming policy, which prohibited her from wearing her hair in cornrows. The significant portion of the quotation is as follows: "[T]he completely braided hair style, sometimes referred to as corn rows, has been and continues to be part of the cultural and historical essence of Black American women."¹² Ford objects that the plaintiff's reasoning -- which was reproduced in the district court opinion -- assumes (1) that there is a distinctive Black culture, or Black women's culture and (2) that this culture is unitary. Ford argues that legal enforcement of Rogers' complaint constitutes "an intervention in a long-standing debate among African-Americans [*pg 182] about empowerment strategies and norms of identity and identification."¹³ Moreover, he says, enforcement of such a right can increase pressure on members of the group in question to conform to the practice that gained legal recognition, provide a disincentive for nonmembers of that group to behave similarly, thus increasing racial disunity, and set precedent for other, even more troubling racial distinctiveness claims.¹⁴

Unproblematic as Ford's assessment of the Rogers argument may be, it unfortunately becomes a template for his presentation of all proponents of "rights-to-difference." Thus, under the heading "Rights-to-Difference Require an Official Account of Group Difference," Ford contends:

Rights-to-difference claims are presented as a simple matter of freeing individuals and groups from a mainstream cultural hegemony. The conceptual foundation of this claim is that there is a baseline of authentic behavior that members of minority groups would engage in out of unfettered free choice, in the absence of cultural hegemony.¹⁵

This passage seems to suggest that rights-to-difference proponents must believe in a relatively sharp distinction between mainstream and minority cultures (cultural distinctiveness), and definitely suggests that such proponents must believe in the unitary ("authentic") nature of any given minority culture (cultural essentialism).¹⁶

Neither suggestion receives further support from Ford. Surely one could adopt a vision of multiculturalism that understands cultural difference as fluid, overlapping, and contingent (as does Ford himself¹⁷). He fails to support either the descriptive claim that all proponents of cultural rights adopt the more rigid, categorical conception of cultural difference, or the normative claim that any rights-to-difference proposal would require such a conception. Along similar lines, it clearly is possible to conceive of culture as a set of social practices, norms, and ideologies that are internally varied, partially inconsistent, and sometimes contested.¹⁸ Thus, Ford's assertion that all rights-to-difference proponents subscribe to a rigid and categorical understanding of culture -- adopting some form of cultural essentialism -- either is descriptive of a contingent fact or is a normative claim about the nature of legal rights. Here again he fails to provide any support for either proposition.

Much of Ford's critique of "rights-to-difference" focuses not on proposals grounded in cultural pluralism, but on proposals grounded in conceptions of identity. Ford uniformly conflates notions of difference among cultures with notions of difference based on individual traits that vary along lines of social identity.¹⁹ More precisely, Ford treats as indistinguishable proposals that, in one [*pg 183] way or another, prohibit discrimination based on some sort of cultural difference and proposals that, in one way or another, prohibit discrimination based on traits linked in some manner to social identity.²⁰ Though each might well be brought under a general rubric of "difference discourse," clearly there is a conceptual distinction to be made between culture and individual traits, and it is a distinction that has some significance if one is concerned primarily about legal rights.

To illustrate the conceptual distinction: I can be a member of the same culture as persons with whom I have almost no shared individual characteristics, and I can share quite a few, even a great number of, individual characteristics with persons who inhabit and are members of wholly different cultures from mine.²¹ The payoff of the distinction for rights proponents is this: If one wished to frame a "cultural right" proposal of some kind, one would need some conception of "culture" (though it need not be a categorical notion; it might, for example, be a functional definition), and some conception of membership in a culture. For example, I can imagine trying to frame a right to fish by gill-netting (which is otherwise prohibited) for individual Native people. In order to do so, I would need some account of the relevant culture in order to support a claim that members of that culture should have unique fishing rights, and I would need some way to test cultural membership. In contrast, if one wanted to frame some sort of legal protection for one or more individual traits, one would need to specify the trait(s), but would not necessarily have to say anything at all about the culture or social identity of persons bearing such traits. Of course, many "trait discrimination" proposals are identity-trait proposals; that is, they link the (desired to be) protected traits with some (desired to be) protected social identity. Even in these instances, however, the proponent of such a right would not have to be concerned about the culture of members of the social category in question. Thus, if I wanted to frame some sort of individual right for gay persons, I would need some definition of "gay," which could be framed in terms of some set of traits (if I wanted to theorize "gay" in that way), but even so I wouldn't have to specify, or even know anything about, gay culture.

Ford systematically overlooks this distinction in his own presentation, moving back and forth without pause between considerations having to do with identity-linked traits and considerations having to do with culture. Consider this sentence: "Rights-to-difference are premised on a belief that identity has a [*pg 184] relatively fixed content that can be protected by rights assertion: Rogers's claim was that cornrows were and had always been the cultural essence of black womanhood . . . ."²² The second half of the sentence is about a cultural "essence"; it has no necessary connection to a claim that "black woman" has some fixed trait-based content. Perhaps the example itself contributes to Ford's confusion: Is wearing a cornrows a trait or a cultural practice?²³

Having thoroughly elided identity and culture, Ford applies the same process of overgeneralization to the former as he did to the latter. "These conceptions of identity share a crucial conceit: Social identities are things in the world and/or reflections of things that can be taken note of as a matter of fact."²⁴ In contrast to this "matter of fact" conception of identity, Ford sets forth his own vision of identity as performance. Here again, however, Ford makes no effort to support his claims that "rights-to-difference" proponents do not also ground their analyses on understandings of identity as performance; there is not even an attempt to name proponents or provide examples.

Finally, Ford tells us that "[r]ights-to-difference are informed by a conception of rights as a limit of social and political power. The proposals envision a legal right that will simply remove a source of repression, freeing the underlying identity to develop without restriction."²⁵ Ford contrasts this "negative rights" approach to his own, which understands rights as constitutive as well as protective: "As policy, legal entitlements in general and especially rights do not simply protect people from outside interference; they also channel energies and shape perceptions about what is important, necessary and good in life. Rights have a tutelary function; they send a message about what society values."²⁶ Once more, there is no support offered for the claim that "rights-to-difference" proponents subscribe to the "negative rights" conception, rather than to Ford's more sophisticated one.

Though Ford's critique of "rights-to-difference" proponents is a conceptual and substantive disaster, there nevertheless is more than a little good to be found in his positive program. Its core, and most valuable, contribution is Ford's insistence on the distinction between racial culture and ascribed race, where the latter is the bare fact of membership in a racial category, by social designation.²⁷ In Ford's view, addressing domination and subordination that attach to ascribed racial status is a proper aim of anti-discrimination law, but addressing "cultural discrimination" is not.

Shorn of its negative program, Ford's discussion of "difference discourse," broadly construed, also has a good deal to offer. Points such as the futility of drawing sharp or absolute distinctions between cultures, the perils of a tendency [*pg 185] to see some aspects of any given culture as "authentic," marking others as "inauthentic," and the value of understanding identity as social performance all are sound and worth making, even if not for the first time. In addition, Ford advances some interesting analyses of multiculturalism -- especially if we read him as a commentator on "multicultural" culture generally -- and the role of law vis-à-vis racial and cultural domination, considered separately.²⁸ On the former subject, I find especially engaging Ford's presentation of the history of cultural pluralism among civil rights advocates, his comments on institutional culture, and his portrait of "cosmopolitan" pluralism. On the topic of law, Ford's comments on the inevitable distributive effects of any legal regime, on understanding civil rights laws as at best a limited vehicle of social justice, and on the social discipline function of law, though points also made elsewhere by others, all add value to his critique of multiculturalism.

II. KEISHA IN CONTEXT

Keisha Akbar is a black woman; by profession, she is a scientist. In 1995 I described Keisha and her employment difficulties in this way:

Keisha majored in biology at Howard University, and after graduation went to work as the only black scientist at a small research firm dedicated to identifying and developing environmentally safe agricultural products for commercial uses. . . . Keisha excelled at the technical aspects of her work, but she brought to it a much less assimilationist personal style. At first, her cultural differences had no particular impact on her job performance. This changed, however, when the once-small firm began to grow rapidly and reorganization into research divisions became necessary. For the most part, the firm planned to elevate each of the original members of the research team to positions as department heads, but Keisha was not asked to head a department because the individuals responsible for making that decision felt that she lacked the personal qualities that a successful manager needs. They saw Keisha as just too different from the researchers she would supervise to be able to communicate effectively with them. The firm articulated this reasoning by asserting a need for a department head who shared the perspectives and values of the employees under her direction. When Keisha raised the possibility that her perceived differences might be race-dependent, the decisionmakers replied that they would apply the same conformity-related criteria to white candidates for the position of department head.²⁹

Footnotes added the following information: "Keisha often wears clothing that features African styles and materials, frequently braids her hair or wears it in a natural style, and at times speaks to other black employees in the dialect linguists designate 'Black English,' though she always uses 'Standard English' when speaking with whites";³⁰ "[a] principal reason for the decisionmakers' perception that Keisha's values were different from theirs was the fact that in [*pg 186] lunchroom conversation she analyzed current events as instances of 'racism' far more frequently than did her white coworkers."³¹

Though Keisha undeniably was the focus of the article in which she first appeared, she was not alone. Keisha has a sister, Yvonne, who in fact was introduced first, at the very beginning of the article:

Goodson, Badwin & Indiff is a major accounting firm employing more than five hundred persons nationwide. Among its twenty black accountants is Yvonne Taylor, who at the time this story begins was thirty-one years old and poised to become the first black regional supervisor in the firm's history. Yvonne attended Princeton University and received an M.B.A. from the Kellogg Graduate School of Management at Northwestern University. While employed at Goodson, she was very successful in attracting new clients, especially from the black business community. In all other respects her performance at the firm was regarded as exemplary as well.

Yvonne always was comfortable conforming to the norms of the corporate culture at Goodson, and in fact has been comfortable with "white" norms since childhood. Her manner of speech, dress, and hairstyle, as well as many of her attitudes and beliefs, fall well within the bounds of whites' cultural expectations. However, Yvonne may have adapted to the corporate culture too well. It is common practice at Goodson to be less than absolutely precise in keeping records of one's billable hours. Instead, accountants generally estimate time spent on clients' accounts at the end of each day, and tend to err on the side of over- rather than under[-]billing. On the rare occasions this practice is discussed, it is explained in terms of the firm's prestige in the business community; the subtext is that clients should consider themselves fortunate to be associated with Goodson at all. Like other young accountants, Yvonne at first attempted to keep meticulous records, but she soon realized that others were surpassing her in billable hours without spending more time actually at work. Consequently, and consistent with her general pattern of conforming to prevailing norms, she gradually adopted the less precise method.

Under Goodson's promotion procedure, the decision whether to promote an accountant to regional supervisor rests on senior partners' evaluations of the candidate's accounting knowledge and skills and, to a lesser extent, on assessments of her interpersonal skills solicited from clients and from peers in the office in which she works. The reports on Yvonne's accounting skills were uniformly excellent. Comments from some peers had overtones of distance and mild distrust suggesting that they were somewhat uncomfortable with Yvonne as a black woman, but these comments fell far below the level necessary to raise serious doubts about her interpersonal skills. However, several of Yvonne's clients took the occasion to register complaints about possible overbilling. The firm launched an extensive investigation and eventually reached the conclusion that Yvonne had been careless in her recordkeeping and that therefore she should not be promoted at that time. As a practical matter, this episode ended Yvonne's prospects for advancement at Goodson; the firm has an informal policy of not reconsidering an individual once she has been passed over for promotion.

[*pg 187]
Yvonne has a younger sister who, sometime during college, legally changed her name from Deborah Taylor to Keisha Akbar . . . .³²

Yvonne played three roles in this introduction to the article. First, she provided a means by which I could set forth and illustrate at the outset the distinction between disparate treatment and disparate impact discrimination as a matter of Title VII law. Yvonne's story exemplifies disparate treatment discrimination because, as the footnotes explained, hers is a case of the "differential application of legitimate employer expectations."³³ In contrast, Keisha's case, resting on the differential effects of seemingly neutral criteria, presents a disparate impact problem. Second, the juxtaposition of Yvonne with Keisha enabled me to make the point sub silentio that Keisha would not necessarily have been better off had she conformed to white norms and expectations; discrimination against people of color in a white-dominated world takes many and varied forms. Third, and perhaps most importantly, Yvonne allowed me to place Keisha's self-presentation in a larger context of black culture, one that did not assume that Keisha's way of being was the only, or best, way for a black person to be. Thus, though I said, "[a]s her decision to change her name suggests, Keisha places an emphasis on her African heritage that Yvonne does not . . . ,"³⁴ a footnote carried an important caveat: "This is not to suggest that Yvonne denies her African heritage, but only that she interprets it differently; she sees being black as congruent with many of the norms of the dominant culture."³⁵

The article went on to feature Keisha, addressing the question whether Title VII could be read to provide a remedy for the injury she suffered. I divided this project into three analytic parts.³⁶ First, because I hypothesized that the decisionmakers in Keisha's case would have genuinely believed that race had nothing to do with their decision (as would many white readers), I wanted to explain why I saw it as a case of "transparently white decisionmaking." Second, given that it was an instance of race-specific decisionmaking in the sense just identified, I needed an argument that fundamental Title VII norms would reach this form of race discrimination. Finally, because existing (judicially created) doctrinal models of Title VII liability would not be adequate to provide a legal remedy for Keisha, I set forth two new models, either of which would do so, and I explored the advantages and disadvantages of each.

With respect to the first of these steps, I'd like to provide readers of this essay the complete actual text and footnotes, as one portion of it figures prominently in Professor Ford's critique:

[I]n spite of the diametrically different cultural styles adopted by Yvonne and Keisha,[7] their stories have the same ending: Each encountered the glass ceiling [*pg 188] at a relatively early stage of what should have been a very successful career.[8] A case can be made that both were disadvantaged because of race.[9] Yvonne would argue that there is no nonracial element of her performance or her personal characteristics that could account for the way her recordkeeping practices were singled out for special scrutiny, and therefore that race is left as the most plausible explanation of the different treatment she received. Even if the basis for the special treatment was unconscious, this is a relatively easily understood form of discrimination: Yvonne's contention would be that she was treated differently from similarly situated others because of her race.

Keisha, on the other hand, arguably was given the same treatment that would have been afforded anyone who was perceived as unable or unwilling to fit smoothly into the corporate culture. Nevertheless, it can be argued that she too was disadvantaged because of her race, in that the personal characteristics that disqualified her from a management position intersect seamlessly with her self-definition as a black woman. I previously have characterized this form of discrimination as an outgrowth of the transparency phenomenon:

White people externalize race. For most whites, most of the time, to think or speak about race is to think or speak about people of color, or perhaps, at times, to reflect on oneself (or other whites) in relation to people of color. But we tend not to think of ourselves or our racial cohort as racially distinctive. Whites' "consciousness" of whiteness is predominantly unconsciousness of whiteness. We perceive and interact with other whites as individuals who have no significant racial characteristics. In the same vein, the white person is unlikely to see or describe himself in racial terms, perhaps in part because his white peers do not regard him as racially distinctive. Whiteness is a transparent quality when whites interact with whites in the absence of people of color. Whiteness attains opacity, becomes apparent to the white mind, only in relation to, and contrast with, the "color" of nonwhites.[10]

Just as whites tend to regard whiteness as racelessness, the transparency phenomenon also affects whites' decisionmaking; behaviors and characteristics associated with whites take on the same aura of race neutrality. Thus, white people frequently interpret norms adopted by a dominantly white culture as racially neutral, and so fail to recognize the ways in which those norms may be in fact covertly race-specific.[11] Keisha would argue that she was not promoted because her personal style was found wanting when measured against a norm that was in fact transparently "white."[12]

[7] For discussions of differing cultural styles within the black community, see LURE AND LOATHING: ESSAYS ON RACE, IDENTITY, AND THE AMBIVALENCE OF ASSIMILATION (Gerald Early ed., 1993) (containing essays by black intellectuals and writers on black people's struggle between nationalistic and assimilationist models of collective identity); Jerome McCristal Culp, Jr., The Michael Jackson Pill: Equality, Race, and Culture, 92 MICH L. REV. 2613 (1994) (imagining colloquies among professors, judges, and citizens of various political orientations on subject of fictional "Michael Jackson Pill," which would remove all "blackness" from black people).

[8] For a discussion of the many difficulties faced by blacks in the corporate world, see GEORGE DAVIS & GLEGG WATSON, BLACK LIFE IN CORPORATE AMERICA: SWIMMING IN THE MAINSTREAM (1982).

[*pg 189]
[9] Throughout this Article, I treat "race" as a relatively stable category, and as a construct with which the reader is familiar. I do so in large part because that approach seems consistent with the conceptual terrain of Title VII. Nevertheless, I find great explanatory power in the analysis of race as socially constructed. See generally Ian F. Haney Lopez, The Social Construction of Race: Some Observations on Illusion, Fabrication, and Choice, 29 HARV.. C.R.-C.L. L. REV. 1 (1994); D. Marvin Jones, Darkness Made Visible: Law, Metaphor, and the Racial Self, 82 GEO. L.J. 437 (1993). . . .

[10] Barbara J. Flagg, "Was Blind, But Now I See": White Race Consciousness and the Requirement of Discriminatory Intent, 91 MICH. L. REV. 953, 970 (1993).

[11] For a more complete exposition of the transparency thesis, see id. at 969-79.

[12] Moreover, the criterion employed in Keisha's case was subjective. Some objective criteria of decision, such as scored tests or, in some contexts, educational requirements, also may be characterized as transparently white, but these are not the subject of this Article. The glass ceiling is maintained much more frequently through the use of subjective bases of decision.

The form of discrimination Keisha experienced also may be labeled institutional racism, defined as the maintenance of institutions that systematically advantage whites. See JAMES M. JONES, PREJUDICE AND RACISM 129-31 (1972). Treating an Afrocentric personal identity as a negative factor in the decision whether to promote an individual to a supervisory position systematically advantages whites as a group over blacks, even though there may be significant numbers of black persons like Yvonne who would not be adversely affected by the use of that criterion. It should be noted, however, that institutional racism may take conscious as well as unconscious forms; thus, the concept overlaps only partially with the notion of transparently white criteria of decision because the latter by definition are employed unconsciously.

Viewed from another angle, Keisha's case is an instance of cultural domination. See Martha Chamallas, Structuralist and Cultural Domination Theories Meet Title VII: Some Contemporary Influences, 92 MICH. L. REV. 2370 (1994). My approach to Keisha's story emphasizes the ways in which structural factors combine with the impulse toward cultural domination to produce adverse employment outcomes for nonwhites.³⁷

This first step was not doctrinally necessary, as Title VII allows a plaintiff to state a claim whenever a criterion of decision has racially disparate effects in an employment context. Given that I ultimately turned to racial disparities as the central way to identify anything as white-specific,³⁸ one could argue that "transparently white" criteria really are nothing more than criteria that have a disparate racial effect favoring whites. In turn, this line of thinking goes, there's no need to identify the criterion of decision in Keisha's case as "white" (and on that basis characterize the decision as one taken "because of" on race, thus bringing it within the scope of Title VII); the existence of a racially disparate effect alone suffices to trigger a Title VII analysis. I concede the point as a doctrinal matter, though I never meant to suggest that the set of transparently white criteria of decision is exactly coextensive with the set of criteria of decision [*pg 190] that have racially disparate effects.³⁹ At the same time, as Keisha's claim likely would not survive under existing disparate impact doctrine, it seemed to me important to provide as many reasons as possible for constructing a new disparate impact framework.

Even more importantly, for me the core undertaking of the article wasn't entirely, or even primarily, doctrinal. My central objective was to illuminate for whites the invisibly raced nature of criteria, norms, and expectations that seem to us race neutral. My decision to frame this project in doctrinal terms was driven by a number of factors: That's how one gets a good law review placement, that's how one demonstrates to certain colleagues that one can (or is willing to) do doctrinal analysis, etc. But even then, I offered a "soft" account of the doctrinal program:

I undertake this project of doctrinal construction with two objectives in view. First, I hope to make the case that in Title VII and its 1991 amendments there are conceptual strands that can be woven together to form a coherent theory of liability for transparently white subjective decisionmaking. In addition, I offer this analysis as an exercise in reflection -- on the nature of the transparency phenomenon and the nature of doctrinal formation, reciprocally. Exploring transparency may tell us something about what race discrimination doctrine might become, and examining doctrinal possibilities may tell us something about who we want and choose to be.⁴⁰

Accordingly, my explanation that Keisha was "disadvantaged because of her race" had much more to do with an "exercise in reflection" than with any doctrinal recommendation.

My argument that Title VII could be read, in principle, to provide a legal remedy for transparently white decisionmaking, the second step in my overall project, turned on an examination of the notion of "equal opportunity." I framed the problem as a choice between pluralist and assimilationist interpretations of that norm. My argument for the former interpretation had two components: I contended that an assimilationist interpretation would fail to implement the Title VII objective of eliminating race as a factor in employment, and that the pluralist interpretation "is more fully aligned with the remedial goals of Title VII than is the assimilationist interpretation."⁴¹ For present purposes the former argument is the more significant, and again I reproduce the entire text and relevant footnotes:

Under the assimilationist conception of equal employment opportunity, Keisha would be required to conform to the prevailing cultural norms in her dominantly white workplace at least to the extent that those norms implicate characteristics within her control. For example, grooming is generally thought to be a matter of personal choice, and to the extent that it is, each employee has an equal opportunity to conform to an employer's grooming code, regardless of race.[121] The decisionmakers in Keisha's case could argue that she had been [*pg 191] afforded an equal opportunity to conform to analogous, but more subtle, cultural norms.

This assimilationist position rests on a false dichotomy between race and individual choice. For Keisha, the two are inextricably intertwined because the aspects of personal identity implicated in the decision not to promote her are race-dependent. Thus the "choice" with which she is faced is in effect a choice to retain her racial identity as she understands it, or to renounce it. She would describe herself as having to shed or disavow crucial facets of blackness, if she wants to get ahead in her place of employment.[124]

One might well argue in response, as could Keisha's supervisors, that even if Keisha experiences her personal qualities as linked with her race, in reality she has not been denied employment opportunity on the basis of race because she had the same chance as any white candidate to conform or be denied advancement. One cannot measure the subjective discomfort entailed by such a choice, the argument would go, and in any event subjective experience should not be relevant. Individuals must make all sorts of choices in life, including the choice whether to "fit in" to a particular environment. So long as the same demands are placed on all employees regardless of race, the argument continues, one should not say that race is a factor in a decision adversely affecting the individual who chooses not to conform.

The foregoing argument is problematic because it reiterates the transparency error. Because it underestimates the centrality of race to personal identity for people who are not white, it incorrectly assumes that the identity costs of conformity to the norms of a white cultural setting for a black person are commensurate with the identity costs incurred by a white person required to conform in the same setting.

Racial identity is not a central life experience for most white people, because it does not have to be. Like members of any socially dominant group, white people have the option to set aside consciousness of the characteristic that defines the dominant class -- in this case, race. Thus whiteness is experienced as racelessness, and personal identity is conceived in a race-neutral manner. However, race plays quite a different role in the lives of people of color in this society. It is, again as a consequence of existing social structures that define and give meaning to racial identity, a central facet of life. One black feminist, bell hooks, describes her experience of race:

I often begin courses which focus on African-American literature, and sometimes specifically black women writers, with a declaration by Paulo Freire which had a profound liberatory effect on my thinking: "We cannot enter the struggle as objects in order to later become subjects." This statement compels reflection on how the dominated, the oppressed, the exploited make ourselves subject. How do we create an oppositional worldview, a consciousness, an identity, a standpoint that exists not only as that struggle which also opposes dehumanization but as that movement which enables creative, expansive self-actualization? Opposition is not enough. In that vacant space after one has resisted there is still the necessity to become -- to make oneself anew. Resistance is that struggle we can most easily grasp. Even the most subjected person has moments of rage and resentment so intense that they respond, they act against. There is an inner uprising that leads to rebellion, however short-lived. It may be only momentary but it takes [*pg 192] place. That space within oneself where resistance is possible remains. It is different then to talk about becoming subjects. That process emerges as one comes to understand how structures of domination work in one's own life, as one develops critical thinking and critical consciousness, as one invents new, alternative habits of being, and resists from that marginal space of difference inwardly defined.[127]

Thus, Keisha's employer is simply wrong in thinking that its conformity requirement is race-neutral; the standard places quite a different burden on nonwhites than it does on white employees. Moreover, this difference is not subjective, but structural. The social significance of race -- the existence of a racial hierarchy -- guarantees that race will intrude on the self-consciousness of nonwhites to an extent that most whites never will experience.[128] Thus the hypothetical white candidate for promotion is unlikely to experience as race-dependent the personal attributes called into question by her employer's workplace conformity rule. Even if she does experience these attributes as associated with race, they are not likely to be for that reason central to her self-definition. For Keisha, on the other hand, conformity is excruciatingly difficult precisely because it calls her racial identity into question.

Once one sees that race is inevitably implicated in matters of "personal choice," it becomes apparent that the assimilationist interpretation does not truly reflect a conception of race-neutral employment opportunity. Under the assimilationist interpretation, the mandate of equality is satisfied in Keisha's case because she could, in theory, conform to the employer's expectations, even though doing so necessarily would levy costs on her that are inseparably linked to her race. The pluralist conception of equal opportunity embodies a more thoroughgoing notion of race neutrality. This interpretation of equality would not hold the requirements of equal opportunity to be satisfied unless the employer at least explored ways of accommodating diverse, race-dependent means of achieving legitimate business objectives. Thus only the pluralist interpretation of equal opportunity can capture fully the vision of a workplace in which race does not matter -- in Title VII's language, a workplace in which the individual is not disadvantaged "because of" race.

[121] See, e.g., Rogers v. American Airlines, Inc., 527 F. Supp. 229 (S.D.N.Y. 1981) (upholding employer's prohibition of braided hairstyles). . . .

. . . .

[124] I reiterate that Yvonne also understands herself as black, though she has a cultural style different from Keisha's.

. . . .

[127] BELL HOOKS, The Politics of Radical Black Subjectivity, in YEARNING: RACE, GENDER, AND CULTURAL POLITICS 15, 15 (1990). Keisha's Afrocentric personal style may be seen as a move toward becoming "subject" in bell hooks' sense.

[128] I emphasize that I'm generalizing, but I do not claim that all whites fit the description in the text, nor do I claim that all black people regard race as central to self-identity in the ways Keisha and Yvonne do. On the dangers, and [*pg 193] necessity, of essentialism, see ELIZABETH V. SPELMAN, INESSENTIAL WOMAN: PROBLEMS OF EXCLUSION IN FEMINIST THOUGHT (1988).⁴²

Finally, the article set forth two models of Title VII liability each of which might implement the pluralist vision of equal opportunity. The first of these was the "foreseeable effects" model. This model tracked existing disparate impact doctrine, but with important modifications. Under this approach, a plaintiff could show the existence of a disparate effect by demonstrating that "an unfavorable employment decision was based on lack of a characteristic more frequently possessed by whites than by nonwhites . . . ."⁴³ This modification of the disparate effects requirement was intended to obviate some of the difficulties associated with "small sample size, bottom-line impact[s], and workforce statistics;"⁴⁴ in that way it was designed to capture the ubiquity of transparently white norms. In addition, the foreseeable effects approach required that the business necessity defense be interpreted narrowly, in a non-assimilationist manner.⁴⁵

The second model of Title VII liability explored in the article was called "the alternatives model." This approach departed from the familiar disparate impact framework altogether. Instead,

[t]he plaintiff's first step would be to analyze the racial structure of her workplace.[169] A showing that the plaintiff's place of employment is predominantly white, or structured in such a way that whites predominate in positions of authority, would trigger a presumption that the adverse action rested on white-specific criteria of decision. However, that showing alone would not shift the burden of persuasion to the defendant. The employer would have to articulate the criterion employed in reaching the challenged decision, and the objectives served by it, but the plaintiff would bear the burden of proposing an alternative criterion that would serve the employer's objective equally well, and in a manner satisfactory to the plaintiff.[170] Finally, the defendant would have an opportunity to persuade the court that adopting the proposed alternative would require unreasonable measures.

[169] Like the foreseeable effects approach, this model would apply to a white plaintiff in any situation in which nonwhites predominate and have final authority over the management of the business, because by its terms Title VII applies to any discrimination because of race. However, I think it exceptionally rare for whites to find themselves in the position just described. But see Ray v. University of Ark., 868 F. Supp. 1104 (E.D. Ark. 1994) (involving claim by sole white officer on campus police force of University of Arkansas at Pine Bluff).

[170] Placing the obligation of formulating a pluralist alternative on the plaintiff is designed to ensure meaningful accommodation of cultural differences and to give the outsider authority to determine what constitutes a satisfactory accommodation.⁴⁶

[*pg 194]

The article concluded with a comparison of what I saw as the relative costs and benefits of each, measured against the goal of providing a remedy for transparently white decisionmaking in employment. I did not explicitly opt for one over the other, though I did hint at a preference for the alternatives model. The conclusion read as follows:

Keisha confronts a form of race discrimination that is as pervasive as it is painful: the expectation that she must conform to norms that challenge her racial sense of self if she is to succeed in her chosen career. I have argued in this Article that though existing judicial interpretations of Title VII are not adequate to respond effectively to Keisha's case, the statute itself, as amended in 1991 and read in light of its underlying objectives, is broad enough to encompass her claim. The challenge, then, is to formulate a doctrinal framework tailored to the task at hand but acceptable in light of other widely held normative expectations. I have set forth two possible models of Title VII liability that would accommodate Keisha's claim, in the hope that they will generate further exploration of possible doctrinal responses to institutional racism.⁴⁷

III. RESPONDING TO PROFESSOR FORD

In this section I engage with Professor Ford's critique on three levels. I look first at his specific references to me in the text of Racial Culture and argue that each of them misrepresents my position in quite significant ways. Second, I show that the general claims he makes about "rights-to-difference" proponents do not apply to my Title VII article, because for each of Ford's contentions there is something in my article that disavows or at least problematizes the view that he associates with "rights-to-difference." One could say that these two subsections are my own "negative program," in which I take strong issue with Ford's representations of my work. However, I do have a positive program as well. It is the case, as I will elaborate below, that my Title VII article relies in important ways on a core norm of cultural pluralism. Professor Ford's argument that this is not the best foundation on which to build race discrimination law is well taken; it is this proposition that prompts me to revisit Keisha's claim. Thus the final portion of this section will examine whether it is possible to refashion a claim for Keisha, one that draws on anti-subordinationist norms rather than pluralist ones.

A. Specific References

Professor Ford names me in the text of Racial Culture on six occasions, though he quotes on those occasions just two phrases seemingly attributed to me. The first of these is the phrase "racially correlated traits," which appears three times, two of them in quotation marks. In fact, I never used this phrase in the Title VII article (nor in any other law review article). To be sure of this, I used the Westlaw "Locate in Result" feature, which permits one to search for specific strings of words within a selected law review article. This procedure turned up no hits for "racially correlated traits," "racially correlated trait," "racially correlated," "traits," "trait," or "correlated." (Just to be sure it was [*pg 195] working properly, I tried "seamlessly," which I know to be there, and that term popped right up.)

Nevertheless, here's what Ford says about the phrase "racially correlated traits" in my work:

The simplest claim is that practices that "correlate" with group identity should receive rights protection. For instance law professor Barbara Flagg argues for the protection of "racially correlated traits," while law professor Kenji Yoshino argues for an approach to anti-discrimination law that "observe[s] correlations between behavior and identity that exist in the world."[citing to Yoshino only]⁴⁸

Rights-to-difference conceive of anti-discrimination law as an injunction to ignore those individual characteristics that are plausibly, if controversially, described as "elements" of a protected group identity. For instance Juan Perea argues that discrimination against "ethnic traits" should be prohibited as such; similarly Barbara Flagg's [sic] proposes to prohibit discrimination against "racially correlated traits," whether or not such discrimination results in a racially segregated or exclusive workplace.[citing to Flagg at 2025]⁴⁹

Many of the rights-to-difference proposals seem to envision a disparate treatment claim in which certain traits would be defined as elements of a protected identity: For instance Juan Perea argues that discrimination against "ethnic traits" should be prohibited as disparate treatment because of the difficulty of proving such discrimination in terms of its disparate impact on a racial or national origin group; similarly Barbara Flagg's proposal to prohibit discrimination against racially correlated traits is designed to avoid the necessity of demonstrating a statistically disparate impact.[citing again to Flagg, supra note 1, at 2025]⁵⁰

Ford's presentation makes it seem as if I set forth a doctrinal proposal under which some traits of Keisha's would receive legal protection because they "correlate" with her racial identity. However, the phrase I actually used was "characteristics associated with whites" (and some variants thereof, such as "a facially neutral criterion . . . associated more closely with whites than with other racial groups"⁵¹). This is not merely a semantic distinction. My "characteristics associated with whites" differ in three very significant ways from "racially correlated traits." First, the characteristics I discuss are characteristics of whites, not generic "racial" characteristics; second, they are characteristics of decisionmakers, not of the persons disadvantaged by their use; and finally, they include not only "[characteristics] that are biological in origin, but [also] characteristics that are associated with whites as a consequence of the existing social hierarchy of race, as well as differences that are more purely cultural in origin."⁵² Thus in my foreseeable effects model "characteristics associated with [*pg 196] whites" become a trigger for further analysis when they are used by white decisionmakers in the workplace in ways that disadvantage persons of color.⁵³ This model would apply even if Keisha exhibited no traits associated with blackness.⁵⁴

The second phrase Ford attributes to me is "personal characteristics that intersect seamlessly with [one's racial] self-definition," and variants thereof. At least this is a phrase I did actually use, though Ford again reads it in a manner that is at best highly misleading. Here are his attributions on this front:

In 1995 Barbara Flagg suggested in the Yale Law Journal that Title VII should prohibit discrimination on the basis of "personal characteristics [that] . . . intersect seamlessly with [one's racial] self definition."[citing to Flagg at 2012]⁵⁵

[Describing another author:] This latter presumption reflects a psychological conception of identity: here identity is a matter of a subjective and internal development of a "sense of self" to which certain practices and beliefs become central. Similarly, Barbara Flagg, [sic] argues for legal protection for traits that "intersect seamlessly with [the plaintiff's] self definition."[citing to Flagg at 2012]⁵⁶

Barbara Flagg writes of traits that "intersect seamlessly with . . . self definition": a subjective, psychological theory of identity.[no citation]⁵⁷

I see now that the phrase "intersect seamlessly with . . . self definition" might be confusing, though perhaps it would be less so if not taken out of context. The complete sentence reads: "Nevertheless, it can be argued that she too was disadvantaged because of her race, in that the personal characteristics that disqualified her from a management position intersect seamlessly with her self-definition as a black woman."⁵⁸ As explained in the previous section of this essay, the discussion at that point in the article was directed at whites' race consciousness; it was intended to make whites more aware of the raced nature of the decision, which otherwise might seem to rest on race-neutral norms and expectations. Even so, I concede that the passage is a bit puzzling. On the one hand, I made Keisha the arbiter of what it means to be a black woman (in order to avoid racial essentialism). At the same time, I said that the raced nature of the conformity standard is not subjective: "[T]he [conformity] standard places quite [*pg 197] a different burden on nonwhites than it does on white employees. . . . [T]his difference is not subjective, but structural. The social significance of race -- the existence of a racial hierarchy -- guarantees that race will intrude on the self-consciousness of nonwhites to an extent that most whites never will experience."⁵⁹ In fact, I think that identity, including racial identity, is formed through a dialogue between self and society, a point I'll explore more fully below.⁶⁰ This passage is at best an indistinct gesture in the direction of that sort of theory of the social construction of self.

In any event, the passage in question plays no role in either of my models of liability. The foreseeable effects model turns on the criterion of decision itself, with no reference to the self definition costs it does or does not impose on an adversely affected employee. In looking to structural considerations, the alternatives model moves even further away from concern about self definition. Neither model bears any resemblance that I can see to Ford's account.

B. Generic References

As described in greater detail above, the principal elements of Ford's caricature of "rights-to-difference" proponents are claims that we (1) imagine there is a very sharp difference between cultures (at one point labeled "provincial multiculturalism");⁶¹ (2) understand individual cultures to be monolithic in nature, either in fact or ideally (so that, for instance, there is some "authentic" black culture);⁶² (3) believe that individual identity preexists society; it is neither socially constructed nor something that is performed;⁶³ and (4) understand legal rights to be negative rights only, having no positive or constitutive function.⁶⁴ Now, one might find it a bit uncharitable of me to make the effort to distance myself from these generalizations, as I already have (I hope) made the case that I never was a "rights-to-difference proponent" in Ford's sense of the term, but as I appear regularly in that role in Ford's presentation of his opponents, I think it worth a couple of paragraphs to respond on this level as well.

The first two of these generic allegations can be dismissed relatively easily, by reference to the text and notes of my article and the immediate implications thereof. As mentioned earlier, the presence of Keisha's sister Yvonne was designed primarily to make it clear that I did not understand black culture to be monolithic, and that I wished not to suggest that Keisha's manner of presenting herself as black was better or more "authentic"⁶⁵ than alternative ways of being. It seems to me that it also follows from one facet of the description of Yvonne that I could not have envisioned a sharp divide between cultures: "[S]he sees being black as congruent with many of the norms of the dominant culture."⁶⁶

[*pg 198]

The question of "identity performance" is a bit more complicated. I, at least, did not have this vocabulary in 1995, but I think that my presentation of Keisha falls within that general frame of thought. First, there is the characterization of Yvonne and Keisha as "interpreting" blackness.⁶⁷ Second, there is the implicit claim that there is no such thing as a "raceless" individual choice; the meaning of any particular choice is given in part by the cultural context in which it is made. At the same time, I do think there is something that can be described as choice (witness the differing personal choices made by Keisha and Yvonne).⁶⁸ This adds up, I think, to a picture of identity-in-social-context; something that is quite congruent with Ford's account of identity as social performance.⁶⁹ In any case, I think there certainly is enough openness in the text⁷⁰ to make Ford's caricature inapposite.

Finally, there is the question of the role and impact of legal rights. I reiterate that I did not advance any doctrinal proposal in the usual sense; I set forth two possibilities "as an exercise in reflection -- on the nature of the transparency phenomenon and the nature of doctrinal formation, reciprocally. Exploring transparency may tell us something about what race discrimination doctrine might become, and examining doctrinal possibilities may tell us something about who we want and choose to be."⁷¹ It seems to me that this passage has more than a little in common with one of Ford's: "As policy, legal entitlements in general and especially rights do not simply protect people from outside interference; they also channel energies and shape perceptions about what is important, necessary and good in life. Rights have a tutelary function; they send a message about what society values."⁷² Too much congruence, I think, to justify saddling me with the "negative rights only" label.

C. Refashioning Keisha's Claim

If my "proposals" do not aim to protect a general "right to difference," they do envision some sort of "right," held by people of color, not to be white.⁷³ I suppose this is some variety of "right to difference," though it is not the variety Ford describes. It is not primarily a right to "cultural difference," though it would encompass cultural differences between whites and people of color when that difference was implicated by a white-specific criterion of decision. Beyond cultural difference, the "proposals" contemplate a form of protection for all types of racial difference from whiteness, whether having to do with physical characteristics, status differences, or any other form of "difference." The key is [*pg 199] the use by whites of criteria that seem to them "neutral," but which in fact operate to favor whites.

Even so, my Title VII article clearly implicates an underlying norm of cultural pluralism, at least some of the time. References to the concept are scattered throughout the piece.⁷⁴ The clearest illustration of the point can be found in my discussion of Title VII's underlying values, while the presentation of the models of liability tells a more complicated story, though one also in which culture makes an appearance.

I framed the question of interpreting Title VII's "equal opportunity" as a choice between assimilation and pluralism, the latter of which might be read more broadly than "cultural pluralism." However, the ensuing discussion focused almost entirely on cultural assimilation and diversity. Thus, to point to the most egregious example, the analysis featured Keisha's grooming choices, and began in this way: "Under the assimilationist conception of equal employment opportunity, Keisha would be required to conform to the prevailing cultural norms in her dominantly white workplace at least to the extent that those norms implicate characteristics within her control."⁷⁵ This discussion centered on the "identity costs of conformity to the norms of a white cultural setting."⁷⁶ I think there can be no question that this section rested on conceptions of cultural diversity and assimilation.⁷⁷

However, both models of liability moved away from any emphasis on notions of cultural difference, each referring back to more ambiguous concepts of "pluralism" and "assimilation." In the section that set forth the foreseeable impact model I discussed what it means for a criterion of decision to be "white." In so doing, I made it clear that the criteria under consideration would not be limited to the realm of cultural differences:

[The requirement] that a facially neutral criterion be associated more closely with whites than with other racial groups if it is to be considered white-specific, is a broad requirement that implicates a wide range of characteristics that might be distributed unevenly across races. It includes criteria that are biological in origin, but extends equally to characteristics that are associated with whites as a consequence of the existing social hierarchy of race, as well as to differences that are more purely cultural in origin.[151] The requirement is easily satisfied; one need only show that the criterion of decision in question is one that occurs more frequently among whites than among other racial groups.

[151] In an earlier article I described a black woman named C.W., who had been denied supervisory positions at a bank because she was less assertive than supervisors were expected to be. This characteristic might be the result of adaptation to racial hierarchy, or it might be simply a matter of cultural style. . . .⁷⁸

[*pg 200]

At the same time, with regard to the business necessity defense I said: "Because the focus here is the problem of assimilationism, this defense should be interpreted narrowly to exclude justifications that reproduce assimilation in another form."⁷⁹ If one reads this sentence in the context of the immediately preceding pages, one might see in it a broad notion of "assimilation," but in the context of the broader Title VII discussion, it takes on a more restricted connotation having to do with cultural assimilation.

Similarly, I presented the alternatives model as a reflection of institutional or structural discrimination:

The alternatives model mirrors the institutional nature of some forms of race discrimination. Keisha's claim, for example, is structural in the sense that it is the consequence of a particular workforce composition and the nature of white race consciousness.[175] Therefore, it seems only natural to construct a doctrinal framework that reflects the structural character of this theory of liability, and thus disavow the intent-like connotations of the existing disparate impact approach. The alternatives model permits the plaintiff who has been disadvantaged by institutionally race-specific features of the workplace to rely on general knowledge about this form of race specificity, and to proceed directly to the exploration of more inclusive employment practices.⁸⁰

However, footnote 175, in the middle of this passage, links structural discrimination with "cultural dominance":

[175] In a workforce that is predominantly white, the tendency of whites to be unaware of the whiteness of facially neutral criteria is exacerbated; in effect, certain structures inevitably produce cultural dominance. However, I agree with the many theorists who maintain that cultural dominance is a powerful force with or without structural reinforcement. See, e.g., DERRICK BELL, FACES AT THE BOTTOM OF THE WELL: THE PERMANENCE OF RACISM (1992); Culp, supra note [7].⁸¹

Though "cultural dominance" might as easily be tied to concepts of racial subordination as to notions of pluralism, the reference to "culture" here likely muddies the waters.

In sum, I think it entirely fair to say that my Title VII article moved rather indiscriminately between notions of cultural discrimination, status hierarchy, and structural and institutional discrimination. Against this, consider Professor Ford's admonition: "[O]pposition to discrimination based on status does not imply opposition to discrimination based on culture. There is no normative inconsistency in a position that staunchly rejects hierarchies of status while adopting a stance of diffidence, ambivalence or indifference as to culture."⁸²

I entirely agree with Ford that there is a very important distinction to be made between status hierarchy (and the actions, including but not limited to discrimination, that reinforce it) and cultural difference (and the protection, or not, of cultural difference). It is this reminder that has prompted me to revisit Keisha's claim. However, I find that I am not in complete agreement with Ford [*pg 201] on several points ancillary to this central one. I'll explore those areas of disagreement before turning to the refashioning of Keisha's claim.

Ford takes the position that:

[P]olitical solidarity based on a common relationship to oppression and domination is the appropriate focus of (racial) identity politics and legal rights assertion; by contrast cultural claims are more contestable on both descriptive and normative terms and should be left to more fluid domains of conflict resolution such as social dialogue, the democratic process and the market economy.⁸³

Along with the distinction between oppression-based claims and cultural ones, it seems that Ford draws a relatively bright line between cultural practices and ascriptive identity: "there is no necessary correspondence between the ascribed identity of race and one's culture or personal sense of self."⁸⁴ Moreover, he states explicitly that he sees ascriptive identity as immutable:

Race and other ascriptive identities aren't like religious affiliations because membership is not chosen by the member or conferred or withheld by the group -- it is imposed by society at large. The Congressional Black Caucus can't kick Clarence Thomas or Ward Connerly out of the black club and they can't quit.⁸⁵

I disagree with each of these propositions. I don't reject all legal protections for cultural difference for two reasons. First, I can imagine some acceptable forms of legal protection for the cultures of indigenous peoples. Second, with respect to non-indigenous groups, I can envision a degree of protection for cultural difference because (and when) I see it as intertwined with other aspects of identity, such as ascriptive status. Because I don't see ascribed, cultural, or other aspects of identity as distinct from one another in the way Ford does, addressing status hierarchy may well take the form of protection of cultural difference.⁸⁶

For example, if we consider Keisha, ascription ("black woman"), status (color stigmatization), resistance to it, and culture (her "Afrocentric interpretation of blackness") all flow into one another; I don't think they can be neatly separated. All exist -- are formed and performed -- in interaction with the larger culture in which she lives, and also in interaction with the variegated black culture Ford so aptly describes.⁸⁷ In short, I see identity as complex and dialogic, and I don't believe the ascriptive strand can be neatly separated from the others.

I also take issue with Ford's assertion that ascriptive status is something entirely given by society. Historians tell us that is not the case; the content of [*pg 202] ascriptive categories changes over time even within a given culture, but there also is reason to think that individuals sometimes migrate from one category to another, and not always by passing.⁸⁸ Moreover, this image of a fixed, entirely socially-given ascriptive identity seems at odds with Ford's own conception of identity as performance.⁸⁹

Now, all of this has to do with the ways one theorizes race; there is no "fact" here, and, in that sense, no right or wrong. Thus I'm not going to argue that my view is correct and Ford's isn't, or even that my way of thinking about race is better in these respects (I'd define "better" by reference to a theory's prospects of contributing to genuine racial justice). Rather, I just wanted to clarify my position on these questions as a context for the reconstruction of Keisha's claim.

I begin with what I really think happened in Keisha's case. She was denied a promotion because of a package of characteristics that, for the white people who were making the decision, added up to her being "too different" from the employees she would supervise to be a department head.⁹⁰ The package in question consisted of dress and grooming choices, use of nonstandard English at times, and unfamiliar (to the white decisionmakers) views on race. Even at the time, and most certainly now, I was and am inclined to read this decision as having more to do with preserving racial hierarchy than with "difference," conceived as something that pertains among equals. That is, Keisha presents an image of resistance to white supremacy (admittedly, she herself is a bit of a caricature). Were this a real life event (as it easily could be), I would strongly suspect that it was just too much of a challenge to white dominance -- dignitary, material, and cultural -- for the white decisionmakers to be willing to place any real power in Keisha's hands.

I don't imagine that the whites who made the decision in Keisha's case thought about the problem in terms of power and privilege; even less do I imagine that they saw themselves as keeping a black woman "in her place," or that they harbored any animus toward Keisha.⁹¹ Rather, for me the point of analyzing instances of transparently white decisionmaking is to expose the very subtle ways in which white privilege can and does reproduce itself. There is no animus here, nor any intent to discriminate. There is just the desire to keep things as they are, which becomes problematic only when we focus on the fact that "the way things are" centrally involves a well developed, though suppressed, racial hierarchy.

Thus Keisha's claim properly rests on anti-subordinationist norm, rather than a norm of cultural pluralism.⁹² And, just as her moral claim is a claim [*pg 203] against racial hierarchy, it would be possible to construct a legal remedy for Keisha that was designed to combat racial subordination. I close this essay by considering, briefly, the contours of such a remedy for transparently white decisionmaking in the workplace, reiterating that these proposals are "exercises in reflection on who we want to be."

As we are speaking of private employers, one needs some sort of statutory anchor for such a right. I think it's entirely possible to interpret Title VII in this way. One would argue that "equal opportunity" requires the dismantling of racial caste. (It's possible, for example, to read Griggs v. Duke Power Co.⁹³ in this way, though one has to admit that the time for that sort of judicial interpretation has passed.) But the key ingredient isn't a clever legal argument interpreting the statute; it's the will to interpret the statute in this way. Or more generally, the needed ingredient is the will of white people to actually dismantle racial privilege.

But assuming we -- or some authoritative legal institution (it has happened before) -- do make some anti-subordinationist commitment, in our culture we'd need a legal doctrine with which to implement it. I think that both of the models set forth in the 1995 article would work well enough, because they really are doctrinal structures designed to put the relevant norms into play. Thus substitution of an anti-subordinationist norm for a pluralist one, at appropriate points, is all that is needed. With respect to the "foreseeable effects" model, the 1995 test for the first prong, the existence of a foreseeable impact, clearly encompasses more than cultural difference.⁹⁴ In the discussion of the business necessity defense one could substitute "subordination" for "assimilation," such that the defense should be interpreted narrowly to exclude justifications that reproduce, or contribute to, racial subordination.⁹⁵ Along similar lines, the alternatives model fairly easily could be made to resonate with anti-subordinationism. The most difficult part would be to replace the suggested "reasonable accommodation" approach to balancing the respective interests.⁹⁶ Something along the lines of a balance that took into account the nature of the challenged action and the purpose for which it was taken, along with the nature and degree of structural inequality in the workplace, probably would be necessary to reflect the focus on anti-subordinationist norms. However, even lacking an established doctrinal concept analogous to "accommodation" (which I attribute to the fact that the law rarely goes very far down the anti-subordinationist path), because the "foreseeable effects" model sounds in difference, rendered problematic in new ways by Ford's commentary, I now have an even stronger preference than before for the alternatives model.

[*pg 204]

Let me spell out again how the alternatives model would work. In any place of employment dominated by whites,⁹⁷ any person of color affected by an adverse employment decision would have an opportunity to require the employer to articulate the criterion used and the reason for using it. The adversely affected employee then could propose an alternative, less discriminatory means of reaching the stated objective, the employer could argue that adopting the proposed alternative would not be reasonable, and the balance between the two would ultimately be resolved by a court, if necessary. But the process of "bargaining in the shadow of the law" might well mean that most cases would be resolved by mutual agreement between the employer and the employee. It seems to me that giving nonwhite employees the ability, backed up by law, to initiate conversations about nondiscriminatory ways of doing things is a step toward the redistribution of racial power. It looks to me something like the way things ought to be.

FOOTNOTES

*	Professor of Law, Washington University School of Law. I thank Pauline Kim, Laura Rosenbury, Terry Smith, and Peter Wiedenbeck for their helpful comments on earlier drafts of this essay.
1.	Barbara J. Flagg, Fashioning a Title VII Remedy for Transparently White Subjective Decisionmaking, 104 YALE L.J. 2009, 2010 (1995).
2.	Barbara J. Flagg, Subtle Opposition, 34 COLUM. HUM. RTS. L. REV. 605, 605-06 (2003); Barbara J. Flagg, On Selecting Black Women as Paradigms for Race Discrimination Analyses, 10 BERKELEY WOMEN'S L.J. 40, 42-43 (1995).
3.	See, e.g., Tristin K. Green, Work Culture and Discrimination, 93 CAL. L. REV. 623, 646 (2005); Terry Smith, Everyday Indignities: Race, Retaliation, and the Promise of Title VII, 34 COLUM. HUM. RTS. L. REV. 529, 538 (2003); Theodore A. Schroeder, Fables of the Deconstruction: The Practical Failures of Gay and Lesbian Theory in the Realm of Employment Discrimination, 6 AM. U. J. GENDER & L. 333, 357 n.151 (1998); Dorothy E. Roberts, The Priority Paradigm: Private Choices and the Limits of Equality, 57 U. PITT. L. REV. 363, 382-83 (1996).
4.	See Richard T. Ford, Race as Culture? Why Not?, 47 UCLA L. REV. 1803 (2003); see also Richard T. Ford, Beyond "Difference": A Reluctant Critique of Legal Identity Politics, in LEFT LEGALISM/LEFT CRITIQUE (Wendy Brown & Janet Halley eds., 2002).
5.	RICHARD THOMPSON FORD, RACIAL CULTURE: A CRITIQUE (2005).
6.	Id. at 211.
7.	As this is not a book review; not all of Ford's points are mentioned here.
8.	Ford, supra note 5, at 20-21.
9.	Id. at 4.
10.	There are a few problems here as well, as I discuss below.
11.	527 F. Supp. 229 (S.D.N.Y. 1981).
12.	Id. at 232 (quoting Memorandum of Plaintiff in Opposition to Motion to Dismiss at 4-5, Rogers v. Am. Airlines, 577 F. Supp. 229 (S.D.N.Y. 1981) (No. 81 Civ. 4474)).
13.	Ford, supra note 5, at 25 (emphasis in original).
14.	Id. at 25-27.
15.	Id. at 70.
16.	See also id. at 73, 79.
17.	Id. at 156-62.
18.	Id. at 23-27.
19.	See, e.g., id. at 101 ("But then aren't racially correlated traits-racial cultures-all potentially proxies for racial status?") Ford's response to this rhetorical question is that racially correlated traits should be seen as proxies if and only if they are so used intentionally. I quote the sentence, however, to illustrate his tendency to treat "racially correlated traits" as identical to "racial cultures."
20.	He may do this because he thinks that all individual traits linked to social identity are so linked by culture, but he's not clear enough about any of these terms for me to form a fixed opinion on what he says regarding this point. In any case, I think it evident that one could undertake a project of valorizing some discredited social identity without making any claims at all regarding a culture associated with that identity, and equally evident that one could describe and defend a discredited social identity without reference to any particular set of individual traits. That is, projects of "identity politics" do not necessarily require reference either to culture or to individual traits.
21.	Of course there will be some overlap in both cases; I am not one of those theorists who subscribe to a view of sharp cultural distinctiveness, and one most likely can find some shared characteristics between any two individuals.
22.	Ford, supra note 5, at 60.
23.	I think the answer is that it's both, and thus one who wants to create a framework for either a culture-based or an identity-trait-based legal claim should think very carefully about how to go about it. I don't think Keisha's claim is either culture-based or identity-trait-based. See infra text accompanying notes 43-47.
24.	Ford, supra note 5, at 61.
25.	Id. at 68.
26.	Id. at 69.
27.	Id. at 93-94.
28.	I don't necessarily agree with all of Ford's contentions, but I do find many of them interesting.
29.	Flagg, supra note 1, at 2011 (internal citations omitted).
30.	Id. at 2011 n.4.
31.	Id. at 2011 n.5.
32.	Id. at 2009-10 (internal citations omitted and alteration added).
33.	Id. at 2010 n.2 (explaining that "[a] white male was promoted to regional supervisor in Yvonne's place. This accountant's recordkeeping practices were not investigated in the same manner as Yvonne's; had that investigation taken place, it would have uncovered practices functionally identical to hers.") (alteration added).
34.	Id. at 2010.
35.	Id. at 2010 n.3.
36.	There's also a section that simply describes existing Title VII doctrine. See id. at 2015-30.
37.	Flagg, supra note 1, at 2011-13 & nn.7-12 (alterations added).
38.	Id. at 2040.
39.	For one thing, transparently white criteria by definition favor whites; clearly there are criteria with racially disparate effects that favor other racial groups.
40.	Flagg, supra note 1, at 2015.
41.	Id. at 2033.
42.	Id. at 2034-36 & nn.121, 124, 127-28 (footnotes omitted).
43.	Id. at 2041.
44.	Id. at 2041 n.155.
45.	Id. at 2041-42.
46.	Id. at 2044 & nn.169-70 (alterations added).
47.	Id. at 2051.
48.	Ford, supra note 5, at 60 (internal citations omitted and alterations added). Ford's use of quotation marks here is especially troubling, as the phrase attributed to Professor Yoshino does in fact appear in his article, though I think it is taken out of context. See Kenji Yoshino, Covering, 111 YALE L.J. 769, 934 (2002).
49.	Id. at 97-98 (internal citations omitted).
50.	Id. at 197 (internal citations omitted).
51.	Flagg, supra note 1, at 2040.
52.	Id. at 2040-41.
53.	Id.
54.	Though it's not the central focus here, I have to point out that Ford's phrases "whether or not such discrimination results in a racially segregated or exclusive workplace" and "is designed to avoid the necessity of demonstrating a statistically disparate impact" also are misleading, independently of the mistaken reading captured by "proposal to prohibit discrimination against racially correlated traits." At page 2025, I was describing existing doctrine, and I pointed out that "the existing disparate impact plaintiff must have statistically significant evidence of racial imbalance in the workforce." I suppose, as I was describing doctrines that would function as barriers to Keisha's claim, that one could infer that I disapproved of this requirement. But it surely does not follow that I was proposing some approach that would be entirely indifferent to whether a workplace was "racially segregated or exclusive." Indeed, if one looks at the alternatives model, one finds a proposal precisely to the contrary, as described above. See supra text accompanying note 46.
55.	Ford, supra note 5, at 11 (internal citation omitted).
56.	Id. at 60 (emphasis in original) (internal citation omitted).
57.	Id. at 98 (alterations added).
58.	Flagg, supra note 1, at 2012.
59.	Id. at 2035.
60.	See infra text accompanying note 87.
61.	Ford, supra note 5, at 167-68.
62.	Id. at 70.
63.	Id. at 61.
64.	Id. at 68.
65.	See supra text accompanying note 35.
66.	Flagg, supra note 1, at 2010 n.3.
67.	Id. at 2010 n.3, 2011 n.4.
68.	See supra text accompanying notes 29-32.
69.	See Ford, supra note 5, at 61-64.
70.	After all, the article is not about "identity."
71.	Flagg, supra note 1, at 2015.
72.	Ford, supra note 5, at 69.
73.	At this point it might be useful to distinguish between a "claim"-a term I use in this essay to encompass both moral claims and legal rights-and a "right"-by which I mean only a legal right. The distinction makes some difference in that I am quite sure that Keisha has a claim in the moral sense, but I'm not completely sure whether its best expression is in the form of a legal right. This essay does not grapple with that question, but seeks only to refashion the proposals set forth in my 1995 article.
74.	See generally Flagg, supra note 1.
75.	Id. at 2034.
76.	Id.
77.	It didn't have to. See infra text accompanying note 93.
78.	Flagg, supra note 1, at 2040 & n.151 (citing Barbara J. Flagg, Enduring Principle: On Race, Process, and Constitutional Law, 82 CAL. L. REV. 935, 971-72 (1994)) (alterations added).
79.	Id. at 2041-42.
80.	Id. at 2045-46.
81.	Id. at 2045 n.175.
82.	Ford, supra note 5, at 93 (alterations added).
83.	Id. at 20 (alteration added).
84.	Id. at 117.
85.	Id.
86.	I accept Ford's basic distinction between anti-subordination norms and cultural pluralism but reject his rigid distinction between ascribed and cultural identity because while the former is a matter of theory, the latter has to do with the real world, in which, in my experience, there rarely are tidy or absolute distinctions.
87.	See Ford, supra note 5, at 23-27.
88.	See, e.g., Daniel J. Sharfstein, Crossing the Color Line: Racial Migration and the One-Drop Rule, 1600-1860, 91 MINN. L. REV. (forthcoming 2007): Ian F. Haney-Lopez, The Social Construction of Race: Some Observations on Illusion, Fabrication, and Choice, 29 HARV. C.R.-C.L. L. REV. 1 (1994).
89.	See Ford, supra note 5, at 61-64.
90.	Flagg, supra note 1, at 2011.
91.	Ford seems to limit status discrimination to discrimination based on animus, though he does at one point acknowledge a limited form of "subconscious" discrimination. See Ford, supra note 5, at 191.
92.	I think I went in the direction I did in part because I tended to collapse differing norms supporting racial redistribution (of various kinds), and partly because, of the seemingly available alternatives, I wanted to select one that I saw as enjoying some degree of popular support. Unlike anti-subordination, cultural pluralism had (and apparently still has) some cachet.
93.	401 U.S. 424 (1971).
94.	Flagg, supra note 1, at 2040-41.
95.	Cf. id. at 2042 ("[T]his defense should be interpreted narrowly to exclude justifications that reproduce assimilation in another form.") (alteration added).
96.	Id. at 2047.
97.	A white-dominated workplace is one in which "nonwhites constitute less than roughly fifteen percent of the workforce"; "is racially stratified, with whites occupying all or nearly all of the upper-level positions"; or "a significant percentage of ostensibly managerial positions are occupied by nonwhites, but in which whites wield most or all of the ultimate policymaking authority." Id. at 2046. I think I would add to this list a racially segregated workplace, under circumstances in which whites were responsible for the segregation.