INTRODUCTION
When Bo Derek made headline news in 1979 for appearing in the movie 10 with beaded and braided hair, she could not have known that her choice to style her hair in that way would have legal significance in an antidiscrimination case two years later (Fig. 1). The case, Rogers v. American Airlines (1981), centered around an American Airlines policy, which prohibited its employees from wearing all-braided hairstyles. Renee Rogers, a Black female employee, challenged the policy because it discriminated against her based on race and gender.
Fig. 1. Bo Derek in Braids. Courtesy ©Getty Images.
The court disagreed. It reasoned that the grooming policy did not reflect sex discrimination because it applied to both women and men. The court further noted that because the policy restricted braided hair, irrespective of racial identity, the policy was race neutral. In reaching this conclusion, the court invoked Bo Derek's braided hairstyle in 10. The court credited American Airlines' argument that Renee Rogers “first appeared at work in the all-braided hairstyle on or about September 25, 1980, soon after the style had been popularized by a White actress in the film ‘10’” (Rogers v. American Airlines 1981, p. 232). The court rejected Rogers's claim that braided hair “‘has been, historically, a fashion and style adopted by Black American women, reflective of cultural, historical essence of Black women in American society’” (pp. 231–232). For the court, from an antidiscrimination perspective, braided hair had no significance. Consequently, it concluded that American Airlines' prohibition on braided hairstyles had “at most a negligible effect on employment opportunity” (p. 231).
Both Paulette Caldwell (Reference Caldwell1991) in a seminal article on the case and Angela Onwuachi-Willig (Reference Onwuachi-Willig2010) in a more recent essay have critiqued the court's approach for failing to consider the constitutive role hair plays in shaping Black women's identity. As Onwuachi-Willig puts it, “[a]lthough the Rogers court clearly understood that there were significant differences in the structures and textures of Black and White hair—Afros and non-Afros—its ultimate conclusion was rooted in an incomplete or flawed understanding of Black hair, especially as it relates to Black women” (p. 1093).
A more general way to critique the court's analysis is to say that the court separately analyzed Rogers' racial discrimination and sex discrimination claims. To put it the way Crenshaw might, the court failed to adopt an intersectional approach. Both Caldwell and Onwuachi-Willig advance this critique as well. According to Caldwell (Reference Caldwell, Moran and Carbado2008), “[i]n a case such as Rogers, an intersectional analysis would necessarily examine the issue at the core of the plaintiff's complaint: that race and gender discrimination operated together to affect her [Rogers] as a Black woman in a way that was not experienced by either White women or Black men” (p. 573).
Central to intersectionality is the idea that all of us have multiple identities—race, gender, class, and sexual orientation, etc.—and these multiple social identities intersect in ways that shape the form and extent of the discrimination we experience (Crenshaw Reference Crenshaw1989). Under this view, because of the difference of gender, Black women and Black men are not equally vulnerable to the same forms of racial discrimination; and, because of the difference of race, Black women and White women are not equally vulnerable to the same forms of gender discrimination. Thus, intersectionality suggests that when courts adjudicate discrimination claims, they should utilize the plaintiff's intersectional identity to determine whether the person was subject to discrimination in relation to their identity as a whole, as compared to utilizing one part of an identity (e.g., race), separate and apart from another (e.g., gender).
Part of the significance of paying attention to the plaintiff's intersectional identity is that it allows courts to consider evidence on the question of whether the plaintiff experienced discrimination based on a sub-group status (for example, being an Asian American woman) within a larger identity group (Asian Americans). Informing this sub-group approach to discrimination is the idea that it is plausible that an employer would not discriminate against all Asian Americans, but would discriminate against Asian American women. Under such a scenario, the employer would be making an intra-racial distinction—a distinction between people in the same racial group.
The standard discrimination claim involves people from different identity groups—for example, a company discriminating against Asian Americans in favor of Whites. This approach requires an Asian American plaintiff to demonstrate that she was treated differently from a similarly situated non-Asian American (usually a White) employee.Footnote 2 But our hypothetical plaintiff might not be the victim of this standard form of discrimination. As noted in the prior paragraph, it is possible that her firm prefers Asian American men to Asian American women, discriminating against the latter but not the former. Framing the discrimination question solely in terms of the plaintiff's Asian American identity ignores the fact that the plaintiff's discrimination could be based on her intersectional identity as an Asian American female.
This essay demonstrates how the identity performance theory we advance further develops one of intersectionality's insights that discrimination is based both on perceived or real inter-group differences (differences among people from different identity groups), in addition to intra-group differences (differences among people in the same identity group). Central to identity performance theory is the idea that to appreciate a person's vulnerability to an intra-group distinction, one must take into account how a person “works” or is perceived “to work” their identity (Carbado and Gulati, Reference Carbado and Gulati2000, Reference Carbado and Gulati2012).
It bears mentioning at the outset that the notion of a performative identity means different things in different disciplines. We invoke the idea here simply to refer to identity-associated ways of being, captured by expressions such as, “Susan doesn't act Black,” “Brian doesn't act like a gay man,” or “Why doesn't Leslie act more like a woman?” Underwriting each of the foregoing statements is the notion that there are existential (and falsifiable) ways to act out the identities we are already presumed to have. This social understanding of identity means that the discrimination question is not simply whether we look the racial part but whether we are perceived to act the racial part. Concretely, then, and thinking in performative terms, while a firm might prefer Asian American men to Asian American women, it is also true that a firm might utilize performative criteria to prefer one Asian American woman over another. Under this scenario, the firm's decision-maker could be asking himself something like the following question: Between two Asian American women, which one acts more like (or embodies the stereotypes we associate with) an Asian American female? The possibility for this kind of intra-group differentiation creates an incentive for people to “work their identities” (Carbado and Gulati, Reference Carbado and Gulati2000, Reference Carbado and Gulati2012) to signal that they do not embody the stereotypes associated with their stigmatized social category. We do not in this essay focus on those strategic performative or identity management strategies (Carbado and Gulati, Reference Carbado and Gulati2000, Reference Carbado and Gulati2012). Our aim instead is to describe the ease with which institutional decision-makers can employ identity-associated ways of being—performative criteria—to engage in intra-group differentiations, intra-group differentiations that currently reside beyond the regulatory reach of antidiscrimination law. To achieve this goal, we build on intersectionality.
Our starting point is an articulation of the theoretical scope of intersectionality—and its location in law. Here, we address, albeit in a rather limited way, some of the problematic ways in which scholars have re-read (and disciplined the reach of) intersectionality. We draw from a paper entitled Colorblind Intersectionality, which Devon Carbado (Reference Carbado2013a) wrote for a symposium issue on intersectionality. We perform this intellectual brush clearing to lay the ground for the working identity or performative conception of race we subsequently describe.
THE BIRTH OF INTERSECTIONALITY IN LAW: CLARIFYING THE BOUNDARIES OF THE THEORY
Crenshaw's article Demarginalizing the Intersection of Race and Sex (Reference Crenshaw1989) published over two decades ago is a classic in antidiscrimination theory.Footnote 3 In it, Crenshaw identifies an antidiscrimination problem that derives from the employment of “single axis frameworks” (p. 139) to adjudicate discrimination claims brought by Black women. These frameworks typically focus on just race or just sex, failing to consider that these two identities interact and intersect in ways that materially shape a person's vulnerability to and experiences of discrimination.
Because Crenshaw's intervention was directed at naming and eliminating these structural disadvantages, some scholars have assumed that intersectionality is only about Black women and/or that the theory is only about race and gender (Carbado Reference Carbado2013a). This assumption conflates the work a general theory of intersectionality might perform with the specific work Crenshaw mobilized her theory to do. Nowhere in Demarginalizing does Crenshaw state that intersectionality is concerned only with Black women or that the theory is just about race and gender (Carbado Reference Carbado2013a). Crenshaw focused on race and gender and Black women specifically because of the intellectual tradition upon which intersectionality is built and “because of the particular juridical and political sites in which Crenshaw sought to intervene. These sites targeted Black women for condemnation, erasure, and marginalization” (Carbado Reference Carbado2013a, p. 812). Crenshaw's engagement of Black women's experiences “should not lead one to conclude that there is an already-mapped terrain over which intersectionality must and only can travel” (p. 812).
We think it important, at the outset, to contest the reification of intersectionality as a theory that is only relevant to Black women and/or race and gender. What is ironic about this reification is that it reproduces an iteration of the representational problem Crenshaw interrogated in her article. Part of Crenshaw's claim was that Black women, because they are Black, cannot represent White women; nor, because they are women, can they represent Black people. The critique and reification of intersectionality as only about Black women reflects another representational assumption—namely, that Black women cannot stand in for, or “function as the backdrop for the genesis and articulation of a generalizable theory about power and marginalization” (Carbado Reference Carbado2013a, p. 813). Our race and gender—and Black woman-centered—engagement of intersectionality is not intended to instantiate the foregoing misreading of intersectionality.
Nor do we mean to reify the idea that intersectionality is only about marginalization. Carbado (Reference Carbado2013a) reminds us that “[t]he theory seeks to map the top of social hierarchies as well” (p. 814). Reading intersectionality this way helps to make clear that all of us have intersectional identities, including White, heterosexual men. This is why Carbado (Reference Carbado2013b) has argued that we should differentiate between “‘intersectionally marginalized groups’ (or IMGs) and ‘intersectionally privileged groups’ (or IPGs)” (p. 14). Doing so would help us to disrupt the extent to which some scholars perceive intersectionality as only about marginalized social categories.
Our final caveat is this: in mobilizing intersectionality we do not mean to acquiesce in crude notions of identity. Like Crenshaw, our project—broadly articulated—is to understand the relationship among power, social structures, and social categories, primarily with respect to law, civil rights advocacy, and academic theorizing. Significantly, when we speak of “social categories” in this context, we are not referring to “identity” per se (though that is a perfectly appropriate discursive shorthand and one that we will employ in this essay). Think of social categories as social locations. Under this framing, the fact that one might refer to Blackness as an identity, for example, should not obscure that it is a complex social terrain within which people are put and whose social boundaries people contest, embrace, navigate, and delimit. This is the understanding of identity that shaped Crenshaw's intervention. Some of the questions motivating this intervention were: How does law produce categories (e.g., race and gender)? What social positions do these categories constitute? What does the production of social categories tell us about power (e.g., racism and/or sexism)? What is the relationship between the conceptualization of social categories (e.g., race or gender) in law and social policy and the performance of civil rights (e.g., antiracism or feminism)? Do articulations of social categories (e.g., race or gender) embed identity defaults (e.g., Whiteness and maleness) that advantage some (e.g., White women and Black men) and disadvantage others (e.g., Black women)?
The “e.g.” parenthetical additions to each of the preceding examples is precisely to disaggregate the range of intersectionality problems one might engage from the particular interrogations of race and gender Crenshaw (provisionally) performed. To make this more concrete, one could replace “race” with “sexual orientation” everywhere that race appears as a parenthetical example (keeping in mind the social location formulation of social categories we set out above). One could do the same with gender—replace it with class. Indeed, effectuating these replacements would both underscore one of intersectionality's core insights—namely, that social categories do not exist prior to but are a product of social forces, including but not limited to law and politics—and advance one of intersectionality's core interventionary interests—namely, to ascertain whether and to what extent those social forces mutually or intersectionally constitute social categories along axes of both privilege and/or marginalization. In this respect, and to borrow from Judith Butler (Reference Butler1993), Crenshaw was interested in the “interarticulations” of power (p. 19).
We should be clear to point out that the preceding claims about identity substitution is not an invitation to scholars and activists to, for example, replace race with class (or vice versa) in their organizing or theorizing. Nor is the point that categories produced by and targeted for subordination or privilege are simplistically interchangeable, or that the particularities attaching to specific intersections generate predictably coterminous outcomes. In other words, we are not suggesting that identity categories are fungible. Our point is rather that intersectionality need not start with a commitment to, and is not exhausted by an engagement of, preexisting categories.
Still, inasmuch as intersectionality trades on the idea of an intersection, one can read the theory to mean that personhood (or identity) can be separated out into discrete social parts. For example, race can be separated from gender. This is because the notion that two things “intersect” brings readily to mind a Venn diagram within which each thing exists both inside and outside of the intersection. Indeed, this is the conception of intersectionality that students being introduced to Critical Race Theory most often articulate. Some even reproduce this understanding schematically along the lines of the diagram in Figure 2.
Fig. 2. The Intersectionality Venn Diagram.
The diagram invites us to imagine social circumstances in which race and gender exist apart from each other as “pure” identities (Fig. 2). Although the metaphor of intersectionality conveys this idea, one need not understand the substantive theory of intersectionality in that way. Our understanding of intersectionality is that race and gender are interconnected, and as a result, they do not exist as disaggregated identities.Footnote 4 In other words, there are no non-intersecting areas in the diagram. In this sense, there might be a tension between the conception of identity that the intersectional metaphor invites and the substantive theory that intersectionality articulates. Perhaps because of this tension, scholars have employed other terms—cosynthesis, multidimensionality, multiple consciousness, compoundedness, interconnectivity, and multiplicity—to discuss the “single axis” problem (or some variation of the problem) that Crenshaw identified (Carbado Reference Carbado2000; Hutchinson Reference Hutchinson1999; Kwan Reference Kwan1997; Valdes Reference Valdes1995; Wing Reference Wing1990). It is arguable whether these terms are better metaphors, since each seems to be predicated on the idea of separate identities categories coming together (as multi- or co-configurations) in some way. Moreover, none of the terms, several of which Crenshaw herself employs in her original articulation of the theory, alters the core insight that the overlapping nature of our social produced identities affects whether and how we experience discrimination.
ONE SIDE OF THE CLASSIC INTERSECTIONAL PROBLEM
To better understand how intersectionality implicates antidiscrimination law, consider the following hypothetical.
Tyisha, a Black woman, is interviewing with an elite corporate law firm. There are eighty attorneys at the firm, twenty of whom are partners. Only two of the partners are Black, and both are men. The firm has three female partners, and all three are White. There are no Asian American, Native American, or Latina/o partners. The firm is more diverse at the associate rank. There are fifteen female associates: three are Black, two are Asian American, and one is Latina. The remaining female associates are White. Of the forty-five male associates, two are Black, two are Latino, three are Asian American, and the rest are White.
Let us stipulate that five other recent law school graduates are interviewing for the job along with Tyisha: a Black man, an Asian American man, one White man, and one White woman. The firm does not hire Tyisha or the White male applicant. Tyisha brings a disparate treatment discrimination suit under Title VII (Civil Rights Act of 1964). She advances three separate theories: race discrimination, sex discrimination, and race and sex discrimination. She does not have any direct evidence of animus against her on the part of the employer. That is, Tyisha can point to no explicit statements such as “We don't like you because you are a woman,” or “We think that you are incompetent; all Blacks are.” The evidence is circumstantial: Tyisha was qualified, but was rejected for a position that was arguably open.
The court rejects all three of Tyisha's claims. With respect to the race discrimination claim, the court reasons that the claim is not supported by evidence of intentional racial discrimination or hostility. According to the court, there is no evidence that the firm dislikes (or has a taste for discrimination against) Blacks. In fact, argues the court, the evidence points in the other direction. The very year the firm denied employment to Tyisha, it offered an associate position to another African American. Moreover, the court points to the fact that the firm had, in the past, promoted African Americans to the rank of partnership. The court concludes that the simple act of not hiring one Black person, especially when other Blacks have been promoted, is insufficient to establish racial discrimination.
The court disposes of Tyisha's gender discrimination claim in a similar way. That is, it concludes that the fact that the firm hired a White woman the same year it did not hire Tyisha, and the fact that the firm has promoted White women to the rank of partnership, suggests that the firm did not engage in intentional sex-based discrimination against Tyisha.
The court concludes its dismissal of Tyisha's compound discrimination claim (the allegation of discrimination based on her race and sex) with an argument that such claims are beyond the reach of antidiscrimination law. More particularly, the court explains that while Tyisha may argue that the firm discriminated against her based on her race or based on sex, she may not argue that the firm discriminated against her based on her race and sex concurrently. According to the court, there is no indication in the legislative history of Title VII that the statute intended “to create a new classification of ‘Black women’ who would have greater standing than, for example, a Black male” (Degraffenreid v. Gen. Motors Assembly Div., St. Louis 1976, pp. 142, 145). Further, “[t]he prospect of the creation of new classes of protected minorities, governed only by mathematical principles of permutation and combination, clearly raises the prospect of opening the hackneyed Pandora's Box” (p. 145).
The foregoing articulates one aspect of the classic intersectionality problem wherein Black women fall through an antidiscrimination gap constituted by Black male and White female experiences.Footnote 5 The problem can also be framed in terms of essentialism. Consider first the court's response to Tyisha's race discrimination claim. In determining whether Tyisha experienced race discrimination, the court assumes that there is an essential Black experience that is unmodified by gender. This approach paints a totalizing picture of racism—that racism affects Black men and Black women in exactly the same way. In other words, racism is totally about race. Thus, this formulation implicitly suggests that it is unlikely that a person who holds negative racial views will distinguish between Black men and Black women (Harris Reference Harris1990).
Yet this is precisely what Tyisha is arguing. Her intra-racial distinction argument is that the firm distinguishes between Black women and Black men, that it prefers the latter, and that this preference is discriminatory. However, to the extent that a court essentializes race (by, for example, conceptualizing race without taking gender differences into account), it makes it likely that the court will not view Tyisha's identification of the firm's preference for Black men as racially discriminatory. Put another way, if, as in our hypothetical case, a court's antidiscrimination starting point is based upon an essential conception of race, that court may have difficulty understanding how a racist firm might promote some Black people (e.g., men) but not others (e.g., women).
Consider now the court's adjudication of Tyisha's sex discrimination claim. Here, too, the court's analysis reflects essentialism—namely, that women's experiences are unmodified by race. The court assumes that if a firm engages in sex discrimination, such discrimination will negatively affect all women in the same way. The court fails to consider that an institution might make an intra-gender distinction between Black women and White women. Yet this is the crux of Tyisha's gender discrimination claim—that intra-gender distinctions constitute actionable gender discrimination. But because the court essentializes gender, it does not view the employer's preference for White women as gender discrimination. Under an essential conception of gender, it is difficult to understand that a sexist firm might promote some women (e.g., Whites) and not others (e.g., Blacks).
Finally, consider the court's rejection of Tyisha's compound discrimination claim. Here, the court doctrinally erases Black women's intersectional identity as Black-women. In effect, the court is saying that, for purposes of Title VII, Black women exist only to the extent that their experiences comport with the experiences of Black men or White women. The contrary view recognizes that Blackness is not gender neutral (it is shaped by and experienced through gender) and gender is not race neutral (it is shaped by and experienced through race).
But now assume that a court is sympathetic to intersectionality or at least recognizes discrimination claims based on both race and sex (Caldwell Reference Caldwell, Moran and Carbado2008). This stipulation helps to introduce the identity performance issue. To appreciate how, assume again that Tyisha is an African American female seeking employment with a predominantly White elite corporate law firm. Four other Black women are also interviewing with the same firm. The firm hires the first four Black women, but it does not hire Tyisha, the fifth Black woman.
This hiring decision creates a buzz around the firm. The firm had never hired so many non-White attorneys. Moreover, the firm has never hired a class within which all the associates were non-White attorneys. Prior to 1980, the firm had never hired a single Black female associate. Further, most of those who were hired after that date left within two to three years of their arrival. Given the history of Black women at the firm—low hiring rate, high attrition rate, and low promotion rate—associates at the firm dubbed this the “year of the Black woman.”
Tyisha, however, is not happy with her rejection by the firm. She files a Title VII discrimination suit, alleging: 1) race and sex compound discrimination, i.e., discrimination against her on account of her being a Black woman; and 2) discrimination based on identity performance. The firm moves for summary judgment (urges the judge to throw out the case without a trial) on two theories. First, it argues that Tyisha may not ground her discrimination claim on her race and sex. According to the firm, Tyisha may separately assert a race discrimination claim and a sex discrimination claim. She may not, under Title VII, advance a discrimination claim combining race and sex. Second, the firm contends that whatever identity Tyisha invokes to ground her claim, there is no evidence of intentional discrimination.
With respect to the first issue, the court agrees with Tyisha that antidiscrimination law does recognize compound discrimination claims based on both race and sex. The court has read, understood, and agrees with the literature on intersectionality. Under the court's view, Black women should be permitted to ground their discrimination claims on their specific intersectional identity as Black women. According to the court, failing to do so would be to ignore the complex ways in which race and gender interact to create social disadvantage: a result inconsistent with the goals of Title VII (Lam v. University of Hawaii 1994).
Turning now to the second issue, the court agrees with the firm. The court reasons that recognizing Tyisha's intersectional identity alone does not prove that the firm discriminated against her because of that identity. The court specifically notes that the firm hired four associates with Tyisha's precise intersectional identity—that is, four Black women. Why, the court rhetorically asks, would a racist/sexist firm hire, not one or two of these women but four, if the firm held discriminatory views against Black women? The court reasons that when there is clear evidence of non-discrimination against members of the same identity category at issue, this evidence produces an inference that the plaintiff was not the victim of discrimination.
The court rejects Tyisha's arguments that Title VII itself and the Supreme Court's interpretation of Title VII focuses on protecting individuals, not groups, from discrimination.Footnote 6 According to Tyisha, a Black applicant who is not promoted may bring a discrimination claim even if another Black person is promoted instead, or if there are other Black employees represented in the position for which the plaintiff is applying and/or are represented in the workplace more generally. Central to Tyisha's argument is the idea that an employer cannot escape liability for having a group represented in the workplace; there is no “bottom line” defense to discrimination.Footnote 7
The court acknowledges that, as a “theoretical matter,” Tyisha is right. The firm's non-discrimination against the four Black women is not proof positive that it did not discriminate against the fifth. The court insists, however, that such evidence is persuasive. It explains that:
[p]roof that [the employer's] work force was racially balanced or that it contained a disproportionately high percentage of minority employees is not wholly irrelevant on the issue of intent when that issue is yet to be decided. We cannot say that such proof would have absolutely no probative value in determining whether the otherwise unexplained rejection of the minority applicants was discriminatorily motivated. Thus, although we agree that such proof neither was nor could have been sufficient to conclusively demonstrate that [the employer's] actions were not discriminatorily motivated, [it is proper] to consider the racial mix of the work force when trying to make the determination as to motivation
(Furnco Construction Corp. v. Waters 1978, p. 580).Footnote 8The court also rejects Tyisha's performance argument. It reasons that Title VII protects against discrimination based on immutable characteristics. Specifically, the court argues that, because Title VII provides no protection for an employee's choices relating to appearance, there is no need to engage the question of whether Tyisha's means of self-presentation (e.g., her hair style and manner of dress) caused discomfort to the lawyers who interviewed Tyisha for the job (Adamitis Reference Adamitis2000).
What's wrong with the court's approach? After all, the court explicitly invoked and took a particular understanding of intersectionality into account. What more should the court have done? To answer that question, we focus on the extent to which performative dynamics might be implicated in the hypothetical we have presented.
ADDING PERFORMANCE TO THE ANALYSIS
Our central claim is that the court should have considered whether Tyisha was the victim of an intra-intersectional distinction based not simply on her intersectional identity as a Black woman, but based on how she worked or was perceived to work her identity. Whereas in the previous hypothetical, the court was essentializing race by not taking gender differences into account and vice versa, the problem with the court's approach here is that it essentializes all “Black female” experiences without taking into account working identity differences among and between the five Black women. In this respect, the court assumes that Tyisha and the other four Black women are equally vulnerable to discrimination. However, this might not be the case. How Black women work or how others perceive them to work their identity affects whether and how they are discriminated against.
Consider the extent to which the following performance issues might help to explain why Tyisha was not hired, but the other Black women were (stipulating that the following information was visually apparent, disclosed on the resume, or revealed in the context of the interview).
Name. Each of the four Black women has a name that is perceived to be American: read, White (Mary, Susan, Helen, Tiffany, and Sarah). Tyisha's name has a Black racial signification.
Hair. Tyisha wears her hair in dreadlocks; the other Black women relax their hair.
Dress. Tyisha does not wear makeup and wore a trousers suit with a Kente cloth scarf to the interview. Each of the four Black women wear makeup and each wore a skirt suit with a white cotton blouse.
Political Identity. Tyisha's resume revealed that, as a law student, she was a student activist and served on, among other committees, “The Black Student Solidarity Committee” and “The Students for Faculty Diversity Committee.” Only one of four Black women participated on an identity-related committee—“Students for Interracial Cooperation.” One was a member of the Federalist Society.
Social Identity. All four of the Black women play tennis and two of them play golf; Tyisha appears disinterested in sports.
Marital Status. All four of the Black women are married. Two are married to White men and each of them is married to a professional. Tyisha is a single mother.
Residence. Each of the other Black women lives in predominantly White neighborhoods. Tyisha lives in the inner city, which is predominantly Black.
Religious Affiliation. Tyisha is a member of the Nation of Islam. The religious identities of the other four Black women are unknown.
Because the court conceptualizes Tyisha's discrimination case solely in terms of her intersectional identity as a Black female, it does not consider any of the foregoing performance dynamics. Yet any one of them could (and all of them together likely would) explain the firm's decision not to hire Tyisha. In other words, it is possible that the firm's hiring decisions reflect an identity preference based on perceptions about how each woman worked her identity.
Note that in this hypothetical, we purposefully exaggerated the working identity dimensions of Tyisha's identity. We do not mean to suggest that Tyisha can stand in for most or even a significant number of Black women, as we are not advancing an empirical claim. Our argument is theoretical in that it strives critically to examine the relationship between a person's vulnerability to discrimination in the workplace, on the one hand, and their working identity, on the other.
But let's assume now that, as an empirical matter, the foregoing performance dynamics are indeed at play in Tyisha's case, such that the firm disfavors Tyisha's working identity and prefers the working identity of the other four Black women. Does this preference constitute discrimination? The answer is not obviously yes. Perhaps the partners' preference for the four Black women is based on their sense that, unlike Tyisha, each of them is likely to fit comfortably within the law firm.
This is not far-fetched. After all, working in an organization is not only about doing work in the literal sense of completing one's assignments. It is also about getting along and establishing relationships with one's colleagues, essentially getting them to like, trust, and feel comfortable around you. This shadow work is particularly important in the context of law firms where there is a strong expectation that associates complete at least some of their work in teams. Perhaps not surprisingly, then, there is now an entire genre of books that stresses the importance of getting along and fitting in in the workplace. These books are very much concerned with some of the issues Dale Carnegie explored in his now classic, How to Win Friends & Influence People (Reference Carnegie1936).
With concerns about institutional fit and collegiality in mind, one could take the position that the firm did not discriminate against Tyisha when it refused to hire her. Instead, the firm was simply attempting to hire the associates most likely to get along with each other and work collegially and collaboratively in teams. By that yardstick, Tyisha looks decidedly less promising than the other four Black women—and some may argue that this has nothing to do with the salience of her working identity as a Black woman who is “too Black.”
One could also argue that none of the performative factors that constitute Tyisha's working identity—her hairstyle and dress, place of residence, and religious affiliation, among other factors—implicate race per se. At best, they are proxies for race—but not race itself. This argument brings us back to the Rogers case. Part of the reason the court rejected Renee Rogers's claim was that, from the court's perspective, race was an “immutable characteristic” (Rogers v. American Airlines 1981, p. 231) but a braided hairstyle was not. “An all-braided hairstyle is an ‘easily changed characteristic,’ and, even if socio-culturally associated with a particular race or nationality, it is not an impermissible basis for distinctions in the application of employment practices” (p. 232). The court went a step further by distinguishing braided hair from an “Afro/bush,” the latter being “a natural hairstyle” that “would implicate the policies underlying prohibition of discrimination on the basis of immutable characteristics” (p. 232). In concluding this much, the court left open the possibility that a grooming policy disallowing an Afro hairstyle “might offend” (p. 232) antidiscrimination law. In effect, the court's argument was that while braided hair was not race per se, an “Afro/bush” might be.
It is beyond the scope of this essay to pursue philosophical arguments about what is and is not race. At the same time, we reject the notion that race is immutable in the sense of being fixed and biologically determined—as do most scholars who write about race (thus the common refrain that race is a social construction). Our broader point is that whether or not the list of performative or working identity factors we outline above are race per se, an employer could draw upon any one of them, and certainly all of them together, to conclude that Tyisha is “more Black” or “too Black” as compared to the other Black women. Such a conclusion would make Tyisha more vulnerable than the other Black women to implicit or explicit negative racial stereotypes.
One can restate this point employing a concept from social psychology: priming. Tyisha's working identity is a stronger intersectional prime, i.e., a stronger catalyst, for the triggering of negative stereotypes about Black women, than the working identities of the other four Black women. We develop this idea more fully in the context of our book, Acting White? Rethinking Race in “Post Racial” America (Reference Carbado and Gulati2012). For now it is enough to understand that race could be (though is not necessarily) implicated in the employer's decision not to hire Tyisha, notwithstanding that the employer hired four other Black women.
CONCLUSION
In advancing the preceding claims, we do not mean to suggest that one could not frame the proceeding problem in intersectional terms. One could, for example, simply treat performance as another intersectional axis like, for example, class or religion. Alternatively, one could, as we do above, treat the performance problem we have identified as a manifestation of an intra-intersectional distinction. Our own view is that the working identity or performance theory we articulate here builds on intersectionality in a discursive frame that both names intersectionality as its intellectual progenitor and highlights some antidiscrimination dynamics that scholars have largely ignored. Our hope is that going forward, scholars will pay attention to the power dynamics at play vis-à-vis our hypothetical intersectional fifth Black woman and think about the implications of these dynamics for law and institutional reforms.
