Printer Friendly

Intersectionality as "catch 22": why identity performance demands are neither harmless nor reasonable.

A. How Intersectionals Confound the Prior Assumptions of
B. What Intersectionals Can Teach the Anti-Discriminator


Private requirements to perform one's identity in a particular manner, even when imposed by employers, are not frequently recognized by either law or society as a form of actionable or even particularly worrisome subordination. (1) For example, neither the law nor most members of society usually choose to sanction employers when they prohibit employees from wearing their hair in cornrows, require them to wear suits, prefer job applicants who can speak comfortably about sports, wine, or any other topic, or fire employees who use profane words. Recently however, there has been an explosion of legal scholarship arguing that we should view at least some of these demands and requirements as a form of negative class subordination, with authors arguing for varying levels of legal and social sanctions on the demands. (2)

For example, in Covering, Kenji Yoshino describes varying forms and effects of pressures to "cover[]," (3) or assimilate, as well as pressures to "reverse cover," (4) or perform one's identity in stereotypical fashion. He argues that these demands are a form of class subordination, leaving aside the question of how many, if any, of these demands should be legally actionable as discrimination. (5) One could describe his work as treating the saliency of conformist demands along a suspect axis, like race or gender, at times as occasion to interrogate the demand, and sometimes as occasion to socially condemn it, but rarely as occasion to legally prohibit it. Devon Carbado and Mitu Gulati describe the actions minority employees must take to fulfill these demands as "identity work," in order to argue that when minority employees must engage in more identity work than other employees they are being discriminated against. (6)

These authors have blurred the line between identity performance demands and traditional forms of discrimination by arguing that our identities are socially constructed or performative, and that our behavior, dress, and other "performances" are at least to some degree constitutive of our identity. In contrast, I employ the case of "intersectionals" to show the potential harm of these role-playing demands. Unlike other authors who have interrogated these demands on the basis of anti-subordination theory, critical theory, or queer theory, (7) I focus attention on the case of intersectionals. I do so by arguing that negotiating multiple identity performance demands simultaneously often places intersectionals in a uniquely restricted situation, one that has been referred to in other contexts as a "catch 22" or "double bind." (8) Thus, the harm I point to in this Article is the harm of having one's choices restricted rather than the harm of covering identity. (9)

Inspired by the idea of intersectionality analysis first articulated by Kimberle Crenshaw, (10) I use the term intersectionals to mean persons who are members of more than one "low-status" category, such as women of color, queer persons of color, or indigent women. However, I engage in a different sort of "intersectionality" analysis than has been previously performed. Thus, as I assist in the project of interrogating role-playing demands, this Article also represents a new technique for describing problems faced by intersectionals and applies it to the field of employment discrimination.

Despite the recent scholarship likening identity performance demands to identity based discrimination, there will be, and is, resistance to condemnation of those demands, largely stemming from two forms of apprehension. These forms of apprehension may be characterized as slippery slope fears about what will happen to all conformist demands and anti-essentialist fears about what legal condemnation of identity performance demands might symbolize. I give credence to these fears. Some types of conformist demands are an inevitable part of life and some are, normatively, quite positive. (11) These demands are a large part of how a society expresses its values. Furthermore, equating bias against a typical form of identity performance for a group with bias against the group itself may naturalize the identity performance in question, thereby naturalizing and essentializing the "differences" between the group and others, promoting prejudice and pigeonholing. Thus, even for those who recognize the problematic nature of many of these demands, it does not necessarily follow that we should legally or socially condemn any identity performance demands rather than merely question them. By highlighting the manner in which intersectionals subject to identity performance demands are likely to experience a greater restriction on their personal freedoms, especially in the form of catch 22's, I hope to convince more people that at least some identity performance demands require legal condemnation.

In making this argument, I use the idea of intersectionality to make a very different sort of point and to inspire a very different sort of analysis than is traditionally performed by those who describe themselves as engaging in intersectionality analysis. Indeed, in some respects, the analysis is at odds with the lessons and advice that scholars writing about intersectionality analysis have given over the years. However, it is deeply indebted to those who first explored the idea of intersectionality analysis and shares their progressive goals.

Intersectionals experience a qualitatively different kind of subordination, and many have stated this point, especially in scholarship of the type that seeks to formulate generally applicable advice and analytical frameworks for approaching the problem of discrimination. (12) Intersectionality analysis generally takes to heart admonitions that gender is raced, race is gendered, and the like, (13) and creates something like a fine grid of identities in which broad categories of identity, such as race or gender, are replaced by smaller sub-categories that carry their own significant weight, such as "Asian American women" or "gay African American men." (14) In application, such analyses seek to understand, for example, discrimination against Asian American women qua Asian American women, or the cultural identities of and specific history of discrimination against African American women qua African American women. (15) For example, such analyses interrogate behaviors that confirm stereotypes specific to Black women, such as the Jezebel stereotype. (16) These analyses are excellent, and no understanding of discrimination would be complete or correct without them.

However, rather than articulate the harms suffered by intersectionals in terms of their specificity, I describe them in terms of the logically incoherent "intersection" of demands or stereotypes that intersectionals are subject to. In other words, while others have used the term "intersection" to mean something closer to either the compounding of subordinations or the specificity of subordination, I use the term to mean actual "conflict." This conflict or logical incoherence represents an untapped potential of intersectionality to expose discrimination, which I demonstrate by placing intersectionals directly at the center of my argument.

I argue that when seemingly innocent identity performance demands conflict for some groups, that conflict may evidence discrimination lurking beneath a facially "reasonable" demand, and furthermore may restrict and mold the choices and identity performances of protected classes of persons more tangibly than other classes, amounting to unequal conditions of employment. The model for this conflict is the "catch 22" experienced by the prevailing plaintiff in Price Waterhouse v. Hopkins, (17) who was denied partnership at an accounting firm after failing to act feminine. (18) The Supreme Court characterized this demand to be feminine as coming into conflict with the demand to be aggressive and masculine in order to succeed in the corporate environment, creating a catch 22. (19) I argue that intersectionals are especially likely to experience these conflicts because they experience multiple identity performance demands simultaneously, thereby increasing the chance for such demands, if not truly evenhanded, to conflict with each other and restrict their choices in a more visible and tangible manner.

In Part II, I describe the sorts of identity performance and assimilationist demands that other authors have recently described as improperly distinguished from traditional forms of discrimination, such as exclusion. These authors have argued that because identity is at least partially performative, some identity performance demands must be condemned as similar to exclusion, or "discrimination by proxy." (20) Part II lays out arguments against condemning these demands, in order to demonstrate why this Article's argument regarding intersectionals is crucial. First, a slippery slope argument, and second, an argument that characterizing identity performance demands as discrimination or subordination amounts to essentialism.

In Part III, I examine the catch 22 described in Price Waterhouse v. Hopkins, so that it can serve as a model for the catch 22 argument I will make. I suggest that intersectionals are especially likely to be subject to the sorts of catch 22's identified in Price Waterhouse.

Part IV presents evidence of intersectionals being subject to catch 22's by pointing to writings by intersectionals describing their experiences. These writings are authored by a diverse group of persons, including African American women, African American gay men, and South Asian women.

In Part V, I use the case of intersectionals to present hypothetical situations in which intersectionals might be subject to catch 22's in order to demonstrate how these catch 22's can clarify the discriminatory nature and impact on choice of certain identity performance demands. Part VI discusses and responds to some objections that could be raised to the paper. Part VII concludes.


A large number of prominent authors have recently written about demands, both formal and informal, placed on the "identity performance[s]" (21) of workers. These authors have argued that we should view at least some of this regulation of identity performance as a form of negative class subordination, with authors arguing for varying levels of legal and social sanctions on the demands. (22)

For instance, in a very representative article, Covering, Yoshino articulates three kinds of demands placed on members of low-status groups. (23) The first demand is "conversion," or the requirement to alter one's identity. (24) We frequently condemn conversion requirements, such as when an employer only hires white people. (25) Such a practice places a conversion requirement on people of color. If conversion is impossible, it amounts to outright exclusion, but even if people of color can convert their race successfully, we still would condemn the requirement that they do so. For instance, religion can be changed, but we legally prohibit and condemn employers who refuse to hire persons of a certain religion, (26) even if conversion would satisfy the employer and secure the victim of discrimination a job. Similarly, requirements to convert from gay, lesbian, or bisexual to straight are condemnable discrimination, legally so in many jurisdictions. (27)

The second type of demand is "passing," or the requirement to hide one's identity completely and pretend to be someone else. (28) For many identities, we condemn, legally and socially, the requirement to pass. (29) For instance, we would condemn an employer that was willing to hire an African American on the condition that the employee pretend to be white on the job. We also would condemn an employer who refused to hire non-whites, but who had unwittingly hired a staff entirely comprised of passing African Americans. The employer would still be discriminating even though he had hired many African Americans.

The third type of subordination Yoshino identifies is a "covering" demand, or the requirement that one not "flaunt" one's identity, "mak[ing] it easy for others to disattend" one's identity. (30) Many employers hire out gay people, but either explicitly or implicitly ask that they not "flaunt" their identity. An employer may do this by hiring men who state that they are gay and answer an interview question about hobbies by discussing football, while failing to hire men who state that they are gay and answer an interview question about hobbies by discussing their antique lamp collection. In other words, gay men are fine, and out gay men are fine, but all gay men must "cover" their gayness even though they are not required to actually pass as straight. Yoshino also identifies "reverse covering" demands, or demands that one somehow signal one's identity--a demand that he believes is frequently placed on women. (31) For example, women are sometimes penalized in the workplace for not acting "female" enough, such as by not dressing femininely. Others have described these "covering" and "reverse-covering" demands as regulation of "identity performance," "identity work," or "assimilationist bias." (32) One could even simply call them role-playing demands. We do not legally condemn such requirements generally, (33) and many people do not even socially condemn such requirements. Even Yoshino refrains from arguing for legal prohibition or condemnation of many such requirements. (34) In fact, he states that legal prohibition may often be inappropriate. (35)

I note an additional category of demands that are widely understood as a form of condemnable discrimination: unequal conditions. An example of unequal conditions would be requiring African American candidates to score ninety percent on a test for employment, while requiring all other candidates to score only eighty percent. Another example would be requiring women with children to work part time, while allowing men with children to work full time. (36) Still a third example would be requiring female employees to have sex with an employer to get a raise, while not requiring this of male employees. (37) These unequal conditions are widely understood to be condemnable discrimination, with cases in which the inequality might be justified, such as affirmative action, exempted from this rule for some, and covered by the rule for others. While different rules for the differently situated are understood as acceptable, even positive, by some (as in the cases of affirmative action or accommodation of people with disabilities, illness, or children) it is unusual for people to think that treating like persons differently, due to stereotypes or animus, is acceptable. If anything, we expect private actors to refrain from such behavior, and it is only more recently that anti-subordination thinkers have begun to call for the accommodation of certain differently situated persons. (38)

Resistance to using the law to combat the regulation of identity performance is understandable, as is resistance to construing these pressures to cover as negative forms of subordination, deserving of condemnation. After all, as human beings, all sorts of pressures are placed on the ways we choose to act, by the government, by employers, and by our communities. These demands are how a society expresses its values and norms. Of course, we can imagine these conformist pressures taken to a terrible extreme--for example, a rule that wearing one's hair in cornrows (39) is prohibited on penalty of imprisonment. Even as we would condemn such an extremely punitive and race-salient restriction, we might recognize that a very real slippery slope exists in the private context between something like a restriction on wearing one's hair in cornrows and a restriction on using "obscene" words. The use of certain vocabulary, even "obscene" vocabulary, just like the decision to wear a certain hairstyle, can be a cultural practice that helps constitute one's identity. (40) And yet, many people--even among those who would disapprove of an employer rule against cornrows--would be uncomfortable not only with legally requiring employers to allow the use of "obscene" words, but also with condemning employers for prohibiting those words.

Requirements to "do one's job" might also be characterized as requirements to conform. For instance, the requirement to speak in standard English can also be understood as the requirement to cover one's national origin or race. (41) While some use this as an opportunity to interrogate the normative value of the requirement to speak in standard English, and are happy to condemn it, others may worry that interrogation is one thing, but actual condemnation is another. Many demands will have salience along some suspect axis, like race, and while this may inform our judgment of the value of the demand, some will feel that this salience should not give rise to condemning or prohibiting the demand, either legally or socially. All demands might then be condemnable, which is quite a different thing than all demands being questionable. Thus, many people will understandably be resistant to legal or social condemnation of role-playing demands, since social pressure, especially conformist pressure, seems, to some extent, an inevitable fact of existence. Indeed, some conformist pressures, like the pressure to not discriminate or the pressure to be respectful of others' emotional privacy, are more than just inevitable, they are probably normatively good.

Additionally, some fear what kind of essentialist statement the law is implicitly, or perhaps explicitly, making when it treats identity performance demands as constituting unfair treatment. Roberto Gonzalez has critiqued the work in this area on this anti-essentialist basis. (42) If we assume that prohibiting typical African dress unfairly harms African American employees, are we implicitly naturalizing or even biologizing a preference for African dress among African Americans, rather than treating it as a sociological or cultural fact? It certainly sounds so if we describe the act of not wearing African dress as "covering' one's racial identity. (43) Even if we argue that African Americans must engage in extra "identity work" when subject to such a rule, are we implying that biologically or for some other "essential" reason, they are more inclined to wear, appreciate, or like African clothes than white persons are, and therefore have to work harder to figure out how to dress in a mainstream way? Does this legitimate a white person's personal dislike of African clothes, even as it protects the African American person's right to wear them? If we describe a requirement that employees act aggressively as a requirement that women "cover" their identity, are we implying that women are naturally timid? At the very least, are we privileging stereotypical acts, such as women acting timid, in a manner that encourages and validates them? Should we focus instead on encouraging people to break out of stereotypes in order to integrate various parts of the workforce?

These fears are serious and cannot be discounted. I share them. However, I believe they are an insufficient reason to draw the line of condemnation between covering-type demands and more traditional forms of discrimination, such as conversion and passing requirements. While some identity performance regulation is inevitable, I believe that the restriction on freedom and personal expression can be significant enough and discriminatory enough that we must use law to limit those restrictions. (44) I use the case of intersectionals to make that point.

An important baseline assumption this Article makes does not necessarily comport with federal anti-discrimination law, particularly Title VII. This is the assumption that intersectionals qua intersectionals should not be discriminated against in employment, promotion, and other tangible employer-provided benefits. Under the current state of Title VII jurisprudence, although an employer cannot exclude all Muslims, and cannot exclude all African American persons, it might be permitted to exclude all African American Muslims. (45) I believe most people would agree with me that if this is true, it represents a failure of either the drafting of Title VII or the constrained reading given it by courts. I also believe most would agree that the statute should either be interpreted to prohibit discrimination against African American Muslims or amended to prohibit such discrimination. Indeed, Title VII does doctrinally cover such intersectional cases when one of the multiple categories is gender, through the "sex-plus" doctrine, (46) and more courts appear to be recognizing claims of discrimination based on multiple factors. (47) This Article is concerned with what the law should be, and how to get there through legal and social change, not with working around flaws in the legal landscape that have already been addressed in academic literature.

This Article also assumes that we condemn traditional forms of discrimination on the basis of sexual orientation. Such discrimination is also not covered under current federal doctrine, although prohibitions on discrimination on the basis of gender could be interpreted to cover sexual orientation discrimination, and we have yet to see how Lawrence (48) plays out as precedent in the federal courts. Discrimination on the basis of sexual orientation is, however, prohibited by many state and municipal laws. (49) In any case, the contributions of this Article are twofold: first, demonstrating the harm to choice of identity performance demands, and second, demonstrating potentially positive uses for the incoherent conflicts intersectionals are subject to at the juncture of various forms of discrimination. The particular axes along which one believes discrimination is wrong are therefore less important than the ability to accept the axes this paper chooses for purpose of example.


The term "intersectional analysis," as used by Kimberle Crenshaw and the scholars who follow in her footsteps, (50) stands for an analysis that takes into account the fact that many people occupy multiple identities and that they may suffer subordination along these multiple axes. (51) In the classic example, the call for intersectional analysis is a call for feminists to remind themselves that some women are also people of color, as well as a call for antiracist activists to remind themselves that some people of color are also women. (52) Crenshaw has noted that both the compoundedness and uniqueness of the discrimination faced by women of color must be accounted for, lest, by studying "single-axis" discrimination along lines of race and sex, we allow ourselves to only study the injuries suffered by privileged subgroups within oppressed classes: white women or Black men. (53) In other words, discrimination against Black women has its own specific history and employs specific stereotypes about Black women, such as the Jezebel or Mammy stereotypes, rather than being equivalent to the "sum" of discrimination against Black men and discrimination against white women. (54)

After Crenshaw's call for intersectional analysis, many agreed with and joined in her call for this accounting of intersectionals. (55) It is now frequently stated that the existence of intersectionals must be acknowledged and accounted for. (56) Many have explained intersectionality analysis as concerned with the specificity of discrimination against intersectional groups. (57) Some excellent and frequently cited applications of this concept exist. (58)

This Article takes the very different, and nearly opposite, approach of appropriating the contradictions inherent in the way many intersectionals are perceived as a tool, rather than attempting to articulate specific discrimination against intersectionals. (59) Racial, gender, class, and sexual prejudice are confused, almost schizophrenic, modes of thought, leading to contradictions, especially when it comes to intersectionals. However, the contradictions that make intersectionals hard to fathom are the very same contradictions that, framed correctly, expose the unreasonableness of discriminators' demands. Thus, this Article describes the multiple forms of discrimination against intersectionals as sometimes conflicting with each other, resulting in a conflict experienced by the intersectional. For instance, women of color sometimes experience a conflict between racialized demands on their behavior on the one hand, and gender-based demands on their behavior on the other hand. This "intersection" of contradictory demands can be described as placing the intersectional in a catch 22.

For the sake of clarity, I emphasize that I am not engaging in a debate with intersectionality scholars about the most accurate or "true" way to describe the experiences of intersectionals and the nature of discrimination against them. Nor am I disputing the value of the now traditional approach to describing those experiences. Rather, I am proposing a different and additional way of describing the experiences of intersectionals and the discrimination against them--one that, like any rhetorical technique, will only have value some of the time. I propose this approach because it has particularly persuasive potential for those in the legal profession, especially in the realm of employment discrimination law.

In the next Part, I use the "catch 22" described by the Supreme Court in Price Waterhouse v. Hopkins (60) as a model for describing the harm experienced by intersectionals when subjected to multiple identity performance demands.


In Price Waterhouse v. Hopkins, Ann Hopkins was denied partnership at Price Waterhouse and successfully sued the firm under Title VII. (61) The denial of partnership was construed by the Supreme Court as illegal discrimination on the basis of sex, even though no formal rule excluded Hopkins from partnership because she was a woman. (62) The Court found that a component of the firm's denial of partnership to Hopkins, an incredibly successful employee, was grounded in impermissible "sex stereotyping." (63) The accounting firm demanded she conform to stereotypes of femininity, and when she failed because she was aggressive and used vulgar language, it denied her partnership. (64) Formally, many may feel that the demands placed on Hopkins are easily characterized as discrimination of the "unequal conditions" variety--she was required to act and speak femininely, while male candidates for partnership were not. (65)

However easy it may be for some to understand these demands as formally discriminatory, it may be just as difficult for others to see them this way. Some simply understand the demand to gender conform as an equal demand placed on both men and women. (66) Alternatively, many would argue that matters like dress, makeup, and mannerism are trivial, and that it does not matter if men and women are treated differently with respect to these issues. This is the basis on which federal appellate courts prior to Price Waterhouse created a doctrinal exception for dress and grooming standards to the statutory language of Title VII, holding that gender differentiated appearance standards do not generally violate Title VII. (67) These courts held that, although conditions of employment which apply to only one sex, or "sex-plus" discrimination, can violate Title VII, only "sex-plus" rules that affect an immutable trait or fundamental right, such as a rule excluding women, but not men, with children, run afoul of Title VII. (68) In fact, even post-Price Waterhouse, plaintiffs do not always successfully claim that requirements to gender conform constitute sex discrimination. (69)

After Price Waterhouse, the Ninth Circuit reached a case in which a man was subject to rather ugly harassment because of his failure to conform to gender stereotypes. (70) In holding that he had a claim under Price Waterhouse, the court felt it necessary to note, without explanation, that different standards of dress for men and women might not be discriminatory. (71) Later, a panel in the same Circuit held that an explicit policy requiring women to wear makeup while prohibiting men from wearing makeup was not discriminatory. (72) And yet, the "smoking gun" (73) evidence in Price Waterhouse was that she was told she should '"walk more femininely, talk more femininely, dress more femininely, wear make-up, have her hair styled, and wear jewelry.'" (74) Why did she win?

An examination of post-Price Waterhouse cases in which courts recognize claims that sex-stereotyping constitutes sex discrimination reveals that such cases generally involve facts which evoke additional sympathy from the federal courts beyond the sympathy that one might expect being forced to conform to a sex stereotype would evoke. These facts may be allowing courts to view such cases as exceptional, even where Title VII doctrine would seem to imply that they are not.

First, such cases are often sexual harassment cases involving extremely ugly and abusive homophobia. (75) Indeed, in Jespersen, the Ninth Circuit held that a plaintiff who was fired for refusal to wear makeup in accordance with a female-only makeup requirement was not fired because of sex. (76) The majority explicitly used the fact that the case did not involve sexual harassment to distinguish its prior, post-Price Waterhouse cases, holding that when an employer acts because of an employee's failure to conform to a sexual stereotype, the employer has acted because of sex. (77) The majority used the presence of harassment as a distinguishing factor even though the Supreme Court has held that whether an action, such as harassment, constitutes an adverse employment action is a separate question from whether that action is "because of sex." (78)

Second, in cases in which there was no harassment, the plaintiff who successfully states a sex-stereotyping claim has generally been a pre-operative transsexual. (79) In these cases, a transsexual plaintiffs failure to conform to gender stereotypes is medicalized through the diagnosis of gender identity disorder. Thus, nonconformity in these cases may appear as both more "exceptional" and less of a "choice" to federal judges.

Ironically, one could easily view Jespersen as the case raising a sex-stereotyping claim most similar to the case Ann Hopkins raised. Darlene Jespersen was, like Ann Hopkins, a woman working in a traditionally male industry--bartending (80)--who refused to wear makeup. And yet, Jespersen lost where Hopkins prevailed.

One of the reasons identified for why Hopkins as plaintiff evoked sympathy from the Court is that she was subject to what the Court called a "catch 22." (81) This so-called "double bind," (82) as others have termed it, was that she was subject to two conflicting requirements: (1) to be aggressive and act masculine, in order to perform well and become a partner, and (2) to be timid and act feminine, in order to become partner. (83) The Court said that it would not tolerate subjecting women to this double bind. (84) This double bind is what Kenji Yoshino describes as demands both to "cover," by assimilating to a standard of excellence as defined by men, and "reverse cover," by not assimilating and performing one's identity in gender typical ways. (85)

Yoshino criticizes the Court's understanding of the catch 22 because it focuses on the impossibility of a woman subject to a catch 22 making partner, rather than on the illegitimacy of the sex-stereotyping demand itself, and reads the Court's opinion as one of many instances in which antidiscrimination law privileges immutable over mutable harms. He notes how especially apparent this is when we realize how frequently women do in fact successfully negotiate contradictory demands to cover and reverse-cover. (86) It is not true that to leave the demands in place would have necessarily excluded women altogether from partnership at Price Waterhouse. Many successful professional women negotiate the demands to be both feminine and masculine at work. Some do it with ease; others find it very difficult. But many do it. (87) In other words, for Yoshino, the demands to cover and reverse-cover are illegitimate in and of themselves, not because they are impossible to meet simultaneously.

Even for Ann Hopkins herself, it is not clear that the demands were impossible to negotiate, despite the fact that they were contradictory, and despite the Court's use of the term "catch 22." For instance, it is unclear from the Court's opinion whether her performance with clients would have been significantly damaged had she behaved more femininely in precisely some of the ways she was asked to, such as by wearing makeup and not using vulgar language. No evidence, even outside the record, is cited in the opinion that would support such a claim. This paper maintains that the best way to characterize this so-called "catch 22" is not as a typical nod to immutability, but as the Court's recognition that even if it were theoretically possible to negotiate these demands, it was too much to ask women to do so. The inconsistency in demands was tangibly harmful to women by restricting their freedom on the path to partnership more so than for men, and also because it helped expose some form of irrational prejudice lurking beneath sex-stereotyping. We will see such a junction of incoherent demands again later, in the case of intersectionals.

We should take a closer look at exactly how the conflicting demands on Hopkins ought to be characterized. The Court noted that the demands conflicted, but in order to resolve the conflict, they could have ruled the demand to "be aggressive" as the impermissible demand, rather than the demand to gender conform. Why did they choose one over the other? Yoshino has characterized the "Be aggressive" type of demand as a covering demand--a demand to assimilate to masculine forms of behavior. He has characterized the "Don't be aggressive" or "Be feminine" demands as a reverse-covering demand--a demand to overperform the stereotypical or perhaps truly common characteristics of one's status group. He has noted that women especially are often subject to both covering and reverse-covering demands, like Hopkins. (88)

I would also characterize the "Be aggressive" demand as a covering, or assimilationist, demand. However, it represents an excellent example of why many may be hesitant to condemn covering demands. Is it really so condemnable for an accounting firm to demand that its partners be aggressive? Perhaps aggressiveness is necessary for the job--not just to impress clients, but also to do the job right somehow. (And, perhaps not; it is very difficult to know. But even if not at Price Waterhouse, there are certainly some workplaces in which aggressiveness is generally a good attribute for the job.) In contrast, the reverse-covering demand of "Don't be aggressive" or "Be feminine" can be characterized as a formally discriminatory demand of the unequal conditions variety. It is a demand that was placed on women, but not on men. It is therefore not too surprising that this is the demand the Court chose to condemn. This is the demand that is easier to condemn as discriminatory in the more traditional sense, while the covering demand was not evidenced to be discriminatory in the more traditional sense.

Notably, though, prior to Price Waterhouse, and even after Price Waterhouse, as described above, many would have characterized even the demand to gender perform as not discriminatory. I suggest that it is the "double bind" or "catch 22" that helped make courts and others who care about ending discrimination cognizant of the fact that the demand to perform gender-appropriately constitutes a kind of unequal or unfair condition. Two demands that might have been acceptable on their own--(1) the demand to be aggressive, and (2) the demand to act gender appropriate--suddenly appear unacceptable and unreasonable when imposed at the same time, on the same woman. For this to occur, the logical inconsistency between the two demands did not have to amount to a complete exclusion of women, or even exclusion of the woman suing her employer. Rather, because the conflict created a restrictive situation for women that men were free from, the Court found the demands exclusionary enough to call them a "catch 22" and condemn one of them as discrimination. And, this Article suggests that if this is the only way to resolve incoherent demands, the Court was correct--such inconsistency ought to be condemned.

Unlike coherent conformist demands, inconsistent demands are less likely to be job requirements or value-laden expression. They serve no socially or economically beneficial function to the actor who imposes them, while frustrating the choices of those, like Hopkins, who cannot or do not want to negotiate them. One might object by stating that an employer ought to be able to ask for a balance. What would be wrong with Price Waterhouse asking its employees to "Be a little aggressive, but don't be overbearing"? In fact, most reasonable conformist requirements would be of this form. Or, one might object that even certain essential job requirements are of this form, for example: Type at least sixty words per minute but do not make more than two mistakes per page. (89)

I do not argue that all catch 22's experienced by employees constitute discrimination, or that the presence or absence of a catch 22 is the test by which we should distinguish legitimate demands from illegitimate demands. Rather, I use the catch 22 to highlight the impact and discriminatory nature of certain identity performance demands, which I hope is persuasive to those who believe that identity performance demands are generally too trivial or too reasonable to be condemned, and that identity performance regulation should be left to employers without legal intervention. (90) The catch 22 is useful because it highlights the manner in which employee choices and freedoms are not only regulated and influenced by identity performance demands, but also can be regulated and influenced in an unreasonable manner--unreasonable because of its disparate impact on the choices of certain employees and unreasonable because of the self-contradictory nature of the regulation.

For example, in the case of Price Waterhouse, Price Waterhouse was not regulating gender performance in a trivial or fair way. This is because the demand to conform to a sex stereotype was neither a rational job requirement nor an even-handed demand. I do not argue that the catch 22 Hopkins experienced was, or should be considered, either a necessary or sufficient condition for finding that the sex stereotyping was discriminatory. Rather, I argue that the catch 22 helps to reveal that the sex stereotyping was problematic. First, women were subject to a catch 22, while men were not. Hence, the path to partnership for women was a messier one than it was for men. Second, the requirement to gender conform was not the arbitrary cultural norm it might have seemed, as highlighted by its conflict with the norms of an aggressive corporate culture. If employees are being treated fairly, the resulting conditions of employment should not be more restrictive or require more complicated negotiation on the part of some employees than on the part of others.

Thus, the catch 22 articulation of Hopkins' experience was useful as a litigation strategy and should also, I suggest, serve as evidence to the rest of us that the requirement to gender conform is neither an arbitrary cultural norm nor harmless. Of course, one could alternately use the insights of critical theory and the theory that identity is performative to reach the conclusion that sex stereotyping should be condemned. However, both the slippery slope argument and the anti-essentialist critique undermine that conclusion. Consequently, I propose the catch 22 as a means of persuading both courts and ourselves that the harms of some identity performance demands are significant and unfair enough to warrant attack.

In other words, Price Waterhouse's demand to act gender-appropriately was always wrong, and some people always saw it that way. But for those who didn't understand just how wrong it was, Hopkins' catch 22 opened their eyes. This Article suggests that this is entirely appropriate. Whatever one thinks of the pervasiveness and unavoidability of some irrationality in the world, (91) a deep suspicion of what appear to be contradictory sets of demands and behaviors is especially appropriate when the accusation of prejudice and group subordination is being made. (92) When identity performance demands conflict, this increases the likelihood that the demands are neither serving a positive function for the employer, nor expressing a cultural preference with trivial impacts on employees. Second, demands that result in greater restrictions on the freedom of some employees than on the freedom of others constitute unequal conditions.

Part V of this Article demonstrates why intersectionals are especially suited to take advantage of this form of argument. By existing at the juncture of various forms of subordination, intersectionals are often erased, or fragmented, by social actors with monolithic understandings. Such stereotypes of various status groups make it difficult to conceive simultaneously of the multiple aspects of intersectionals' identities. (93) And yet, this need not be a curse. It can also expose contradiction and irrationality on the part of those who subscribe to monolithic generalizations.

Indeed, some have proposed that the facts of Price Waterhouse are extremely similar to cases of employers discriminating against perceived lesbians. (94) Lesbians themselves may be understood by some as intersectionals since they have a low-status gender and a low-status sexual orientation. The conflicting demands could then be understood as: (1) the demand to be aggressive and masculine, or to assimilate to the cultural and stereotypical practices of higher status gender; (2) the demand to act feminine, or to assimilate to the cultural and stereotypical practices of the higher status sexual orientation. These two apparently neutral conformist demands could be understood as conflicting in the case of Hopkins' intersectional status as a perceived lesbian.

Untapped in all the talk about intersectional analysis is the fact that intersectionals are people whose very existence exposes the kinds of irrational catch 22 conflicts described by the Price Waterhouse Court. The existence of a woman nearing the partnership level in the male-dominated firm of Price Waterhouse disrupted prior stereotypes and social norms enough to give rise to an incoherent situation--the firm's requirement that a partnership candidate be simultaneously aggressive and not aggressive. Similarly, the existence of intersectionals is disruptive to those who understand subordinated social groups as monolithic. Both discriminators and anti-discriminators fall victim to this mistake. Thus, intersectionals can give rise to incoherence that similarly exposes discrimination.


Marlon Riggs states:</p> <pre> I am a Negro faggot, if I believe what movies, TV, and rap music say of me. My life is game for play. Because of my sexuality, I cannot be black.... I cannot be a black gay man because by the tenets of black macho, black gay man is a

triple negation. (95) </pre> <p>Although here Riggs is criticizing the tenets of "black macho" as the tenets of black America itself, he conceives of the black macho tenets as an internalized stereotype rooted in and responsive to the "racist myth of the super-macho black stud." (96) Isaac Julien and Kobena Mercer state that "[t]he stereotype of the threatening black 'mugger' is paradoxically perpetuated by the way black male youth have had to develop macho behaviors to resist harassment, criminalization, and the coercive intrusions of white male police forces into their communities." (97) Thus, the racist myth of Black men as super-macho comes into tension with the myth of gay men as sissy or effeminate. These tensions between the stereotypes surrounding those with deviant sexuality or gender performance and the stereotypes of minority ethnic communities and their members, as subscribing to hyper-patriarchal traditions, such as super-macho behavior or oppression of women, find resonance with members of other ethnic communities.

For example, Kaushalya Bannerji, a South Asian lesbian, describes having to "point[] out to [her] classes at the University that not all Third World women are passive or heterosexual," (98) in addition to being forced, with regard to living at home with her parents, to "explain[] to white lesbians that [her] living situation was not 'Eastern patriarchal oppression.'" (99) Bannerji states that "Living at home as a South Asian lesbian daughter has been a complex and sometimes contradictory process for both myself and my parents." (100) But it would seem that even being able to claim this identity, "South Asian lesbian daughter," must be instructive and transformative for her white lesbian friends as well, who initially understood this state of South Asian daughter as somehow inherently at odds with lesbianism.

Another form of contradiction pointed out in the experience of intersectionals is the incoherence of racist treatment of African Americans and patriarchal sexist treatment of women, in the case of African American women. (101) Sojourner Truth pointed to this contradiction:</p> <pre> That man over there says that women need to be helped into carriages, and lifted over ditches, and to have the best place everywhere. Nobody ever helps me into carriages, or over mud-puddles or gives me any best place, and ain't I a woman? Look at me! Look at my arm. I have ploughed, and planted, and gathered into barns, and no man could head me! And ain't I a woman? I could work as much and eat as much as a man--when I could get it--and bear the lash as well! And ain't I a woman? I have borne thirteen children, and seen them most all sold off to slavery, and when I cried out with my mother's grief, none but Jesus heard me! And ain't I a woman? (102) </pre> <p>By repeating, "[A]in't I a woman?," Sojourner Truth used her existence at the intersection of the categories slave and woman to reveal the irreconcilability of the norms of Southern patriarchy with the racist norm of brutal treatment of African American slaves. In later years, African American women have used this continuing contradiction in the simultaneous statuses of African American as worker and laborer, and woman as housewife and mother, to challenge the naturalization and biological justification for the patriarchal division of labor:</p>

<pre> Throughout most of the history of the United States normative notions of what a woman ought to be have reflected what upper-class women were able to be.... These women, who "become symbols of domesticity and of public service," influence our views of what feminine behavior

should be. The ideology of the division of labor, then, is often

far removed from the reality of the experience of minority and other working women, yet it remains the cultural ideal, often buttressed by the canons of religion and rationalized as "natural".... (103) </pre> <p>Mullings describes the historical "dilemma created by the distance between the patriarchal model and the real experience of African American women." (104) What she calls a dilemma, the Supreme Court in Price Waterhouse termed a catch 22. Such dilemmas are often created by discriminatory treatment or perception. In the case of the Southern dilemma, it derives from understanding some women as categorically different than others on the basis of their race. As Mullings has stated, "The dilemma ... is in part addressed through representations of African American women as defeminized, inappropriate, or bad." (105)

This has translated into a modern double bind for African American women. In order to cover against racist stereotypes of African Americans as lazy, African American women are subject to simultaneous demands to fulfill patriarchal ideals of good mothering and racialized demands to be workers. (106) This conflict results in African American mothers who work being vilified as "matriarch[s]," whose failure as mothers underlie the problems of the African American community. (107) At the same time, African American mothers who do not work are vilified as lazy "welfare mother[s]." (108) As Mullings notes, "[t]hese women, who must work to ensure the survival of their families, are then attacked as 'matriarchal'," (109) and are accused of being "bad mother[s] responsible for low educational attainment, crime, and delinquency." (110) If they do not work, they are subject to the "demonization of 'welfare mothers'" (111) as "promiscuous and lazy." (112)

In Arpana Sircar's extensive interviews with Indian immigrant women who work, she finds that most of these women feel that the structural constraints of sexism and racism in the United States workplace are far less significant than the "structural inhibitors existing in the Asian Indian subculture itself." (113) They describe conflicts between pressures to culturally assimilate at work and domestic pressures to conform to gender-stereotypical behaviors at home, with some finding coping mechanisms for dealing with these conflicts to be "a great asset." (114) Others, however, find them "too stressful and hypocritical," (115) stating that '"[i]t is like becoming someone else!'" or describing the situation as "role overload." (116) And yet, these conflicts between the demanded performances at work and home occasionally become dilemmas imposed by the same institution, rather than different institutions each making their own selfish demands. For instance, one respondent, a molecular geneticist, "was told by coworkers that she spoke too much for an Asian Indian woman who is generally soft-spoken and does not know much about the United States." (117)

These examples of the tension between different irrational stereotypes and demands placed on people of low-status identity indicate more than frustrating situations. Rather, they demonstrate the potential of intersectionals to highlight the harms of the performance demands underlying the conflicts. For the sake of clarity, I note that many intersectionals have described conflicts between what is expected of them by one identity group, such as a gay community or an employer, and what is expected of them by another identity group, such as a Black community or a family. Without demeaning the hardships imposed by these other conflicts, what I have limited my examples to, and what I mean when I refer to catch 22's or conflicts, are situations where conflicting demands or presumptions arise in a single environment and are made by the same agent, be it an employer, a court, the government, or a group of social acquaintances. In other words, descriptions of role overload by the Indian immigrant women in Sircar's book do not serve as examples of those subject to a catch 22. (118) However, when those women describe employers who make self-contradictory demands, (119) their experiences do serve as examples of catch 22's.

In the next Part, I use hypothetical examples of intersectionals subject to these catch 22's in order to highlight the harm done by identity performance demands. However, before continuing, a word of acknowledgement is necessary. Seizing on the contradictions presented in this Part, and describing them as a "conflict" between individual axes of discrimination, is at odds with both the tenets of many critical race theorists, with whom the concept of intersectionality has long been identified, as well as the typical lessons drawn from Crenshaw's original work on intersectionality.

First, Critical Race Theory encompasses a rich tradition of pointing out the value and pervasiveness of dual or multiple consciousness among people of color, (120) as well as resistance to the overuse of the concept of formal equality. (121) Critical Legal Studies scholars have rightfully challenged the notion that legal answers can always be determined from logic and reason. However, the fact is that judges, law professors, and lawyers alike all continue to strive for some measure of reasoned argument and consistency in their work. Many of these legal actors do so even as they accept that some norms sustaining existing power structures will inevitably be incorporated in their work, probably cloaked with a false air of neutrality. (122) Indeed, for all the success of Critical Legal Studies, very few in the legal profession have been willing to throw out the ideals of formal equality and the rule of law entirely, (123) and even those who come close tend to backpedal enormously. (124) Because they personally strive to be fair and reasonable, when confronted with self-contradictory belief and action, legal actors get suspicious. If progressive scholars should embrace this tendency anywhere, it is in the context of antidiscrimination law, where we know that racism, sexism, and other forms of prejudice are often rooted in unreasonable and unsustainable approaches to and understandings of people and the world.

Second, much intersectionality scholarship warns against "single-axis" study of discrimination, and especially against treating discrimination along multiple axes as equivalent to "additive discrimination": Discrimination against Black women Discrimination against Black (men) + Discrimination against (white) women. (125) However, while this Article's approach is at odds with some of the tenets and methodologies of those who have self-identified as performing intersectional analysis, our progressive goals are shared, and it would be remiss to neglect the potential of intersectionals to expose discrimination when their experiences are framed as the "conflicts" between single-axis discrimination that I describe in this Article. In other words, were we to reject all rhetoric of equality, reason, consistency, and group-based prejudice and harm, we would take criticism of the inadequacy of these approaches too far. (126)


In Part VI. A, I describe how intersectionals, by existing at the convergence of various forms of discrimination, can highlight injury and expose the unreasonableness of those who discriminate. I then describe, in Part VI.B, how those who do not discriminate could learn from all of this.

A. How Intersectionals Confound the Prior Assumptions of Discriminators

Imagine someone who harbors certain stereotypes. Let's call him "The Discriminator." The Discriminator, like many people, hyper-sexualizes gay men--he thinks that they all have uncontrollable, even predatory, sex drives. The Discriminator, again like many others, also desexualizes Asian men--he thinks they are all asexual and passive, maybe even impotent. One day, The Discriminator meets a gay Asian man. Suddenly, he doesn't know what to think. Is this man hypersexual, or asexual? He can't be both. Whatever he is, at least one of The Discriminator's assumptions--either about gay men or about Asian men--cannot be sustained with respect to this man. Perhaps even both assumptions will be challenged. Legal scholars have continually pointed out the fact that intersectionals suffer from multiple and unique kinds of discrimination, but they have not clearly pointed out what this hypothetical demonstrates--intersectionals have the potential to confound the Discriminator, because they suffer from logically inconsistent discrimination and stereotyping, exposing the irrational and often self-contradictory or conflicted nature of prejudice. This is an untapped potential of intersectional analysis.

More examples may help, as well as demonstrate that this kind of intersectional conflict may occur in many situations. Consider the gay African American man. The Discriminator thinks that gay men are effeminate, but he hyper-masculinizes African American men. What will he think when he realizes that some gay men are African American? How about the Asian lesbian? The Discriminator hyper-sexualizes and hyper-feminizes Asian women, as well as perceives them as passive and docile. He also thinks that lesbians are asexual, mean, butch, and aggressive. The Asian lesbian thus cannot make sense. What about the Latina woman? The Discriminator thinks that Latinos are universally strong, fiery people, and this is why he likes hiring Latinos. The Discriminator also thinks that women are quiet and weak, which he does not like, and this is why he never hires women. What happens when a Latina woman applies for a job? What about the homeless gay man? The Discriminator thinks that gay men are more well-educated, intelligent, and hardworking than the average American, and therefore should not get "special rights," but thinks that homeless persons are unintelligent and lazy. What does The Discriminator think when he finds out about the homeless gay man? His very existence is something of a challenge to The Discriminator's world view.

I do not maintain that this form of conflict always occurs in the case of intersectionals. Indeed, The Discriminator's stereotypes frequently complement each other, such that an even more exaggerated expectation can result. For instance, for The Discriminator who perceives both gay and black sexualities as depraved and deviant, the existence of a gay Black man will not confound his or her world view. Rather, The Discriminator would approach and perceive the gay Black man with a heightened and even more extreme prejudice. The two stereotypes would feed and sustain each other. (127)

I also do not maintain that it is only certain categories of intersectionality, such as "gay Asian man" that result in unsustainable conflict. It is not the type of intersectional that determines whether The Discriminator will find himself confounded; it is the type of stereotypes that The Discriminator maintains. This is made clear by the fact that many low status racial and gender groups find themselves subject to animus-based stereotypes or irrational expectations that are themselves contradictory. Thus, whether or not an intersectional will confound a Discriminator depends on which of many contradictory racist, sexist, homophobic, and other stereotypes a Discriminator gives primacy to in a particular context. For instance, white supremacists in the South schizophrenically perceived Black women sometimes as "Jezebels" and sometimes as "Mammies." (128) Currently, some homophobes think of gay men as predatory, hypersexual aggressors, while others perceive them as weak and effeminate. Thus, while one Discriminator's assumptions about race and sexuality may be complemented and heightened when confronted with a gay Black man, as I have described above, another's might be confounded. The Discriminator who thinks of gay men as extremely weak, passive, and effeminate, and who also thinks of Black men as hyper-aggressive, macho, and masculine, will be confounded when confronted with a gay Black man. The fact that the existence of conflict depends on The Discriminator makes sense. After all, the conflict is internal to The Discriminator, who is irrational in holding a world view that does not comport with reality--it is not a conflict created by the intersectional; it is a conflict created by discrimination and animus, but revealed and experienced by the intersectional.

I do not propose that all or even most Discriminators experience a grand enlightenment and see the error of their ways when confronted with such intersectional conflicts. Of course, prejudice is tenacious, and the most likely result is that The Discriminator develops exactly what Crenshaw and others have described--a new and more nuanced stereotype or belief specific to the intersectional. For example, The Discriminator is likely to revise his or her opinion and decide that where he previously thought African American men are threatening, he now thinks only straight African American men are threatening; gay African American men are not. Of course, there are times when confronting an intersectional can be so confusing and unexpected that The Discriminator experiences a more fundamental paradigm shift.

Finally, let us take the case of The Assimilating Firm. At The Assimilating Firm, everyone goes out for drinks on Friday evenings. You do not have to participate, but if you want to advance within The Firm, everyone knows that you are supposed to go because the happy hour is considered an important event where collegiality and trust are built. (129) At the happy hour, many of the younger employees sit at a table and talk about their sexual escapades from last weekend. For now, leave aside the issue of whether this constitutes sexual harassment, and assume that nobody feels sexually harassed because this is all in good fun. There's nothing "hostile" about this environment, and The Firm is well-integrated, with both men and women occupying all levels of the hierarchy equally. (130) One day, The Firm hires an out, lesbian South Asian woman. She goes out for drinks on Friday with everyone, but when it is her turn to talk about sex, she says "Oh, I don't really want to." Everyone boos in disappointment, and she finds out later that her boss thinks she is too reserved and docile, and probably can not handle much responsibility yet. (131) The new hire is very irritated because she thinks these are stereotypes about Asian women, and she's pretty sure that if she were not an Asian woman, the boss probably would have excused her decision to not talk about sex as the understandable hesitation of a new person. She decides perhaps she has to overcompensate for being Asian.

The next Friday, at the happy hour, she talks about how she had sex with a woman she met last weekend at a bar in the East Village. On Monday, the new hire overhears a manager at the firm talking about how inappropriate he thought the story was. He says "I am not against gay people. I just really think there's no need for her to flaunt it like that. Plus, I don't think anyone, gay or straight, should be telling the group about anonymous sex in a public place. Yuck." The new hire feels like she does not know what she is supposed to do. Note that the only way to resolve the logically inconsistent demands is for the woman to pass as straight. This is not to say that passing is the only mechanism available to negotiate the demands. Our hypothetical employee could probably discuss her sex life, but excise mention of behaviors that would make her coworkers uncomfortable. But unless she passes, the demands placed on her will remain logically inconsistent, regardless of the technique she develops to negotiate them. Requiring members of certain low-status groups to pass is a demand that lies firmly within traditional notions of discriminatory behavior.

These stories are meant to demonstrate how discriminatory stereotypes and conformist demands can be revealed as such in the case of the intersectional. The rationality of the Discriminator's presumptions and the rationality of the Assimilator's demands are challenged by the intersectional. They are revealed as logically inconsistent.

The ability of intersectionals, via the concept of a catch 22, to highlight this irrationality and prejudice may be useful as a matter of social transformation. This would mean that accounting for intersectionals is important, not just for the reasons identified by Crenshaw--that intersectionals suffer additional and different discrimination that must be dealt with--but also because they are useful in the fight against even single-axis discrimination and stereotyping. (132) It would mean that public service announcements about safe sex that include queers of color, anti-poverty campaigns that include queers, Asian American political groups that make queers public, rather than hidden, all could reap the gains of the intersectional conflict.

The leaders of the campaign to end the exclusion of homosexuals in the military have been criticized for largely ignoring Perry Watkins, an African American gay man who "repeatedly acknowledged his sexual orientation to commanding officers" and even put on "military drag shows" during his tenure. (133) He was dismissed after fellow soldiers tried to gang rape him, and won a court battle because the Army had allowed him to reenlist repeatedly despite their knowledge that he was gay: (134)</p> <pre> According to Tom Stoddard, a white gay lawyer who directed the Campaign for Military Service, "[T]here was a public relations problem with Perry." Ostensibly, the problem was not simply that Watkins was black, but also that he wore a

nose-ring.... Watkins nevertheless indicated that race was the determinant. "[W]e'll go with a [white] woman who lied for twenty years before we go with a black man who had to live the struggle nearly every day of his life." (135) </pre> <p>But instead of theorists and activists imagining that the exposure of and focus on intersectionals will somehow complicate or dilute their political and social progress, the leaders of low-status groups (persons who are often non-intersectional) could realize the untapped capital of the intersectional in countering stereotype and presumption. This is not to say that activists and theorists should not be accounting for intersectionals anyway--even if it were not in their strategic interests. Fairness and the democratic principles of inclusiveness are sufficient for that result. But it is to say that perhaps feminists, antiracist activists, and queer activists should begin thinking of intersectionals as named plaintiffs, poster children, (136) and leaders of the common cause of antisubordination.

But even if pointing to irrational prejudice and exposing conflicting stereotypes will do little in the social and political realm (after all, people do think and behave irrationally, and discrimination is notoriously tenacious) (137) at least in legal discourse, and especially in the realm of antidiscrimination law, the exposure of this irrationality ought to count for something.

B. What Intersectionals Can Teach the Anti-Discriminator

How about the Anti-Discriminator? What does the case of the intersectional teach her? By the Anti-Discriminator, I mean the person who wants to legally and socially condemn and prohibit traditional forms of discrimination, but does not want to legally and socially condemn demands to dress, act, speak, or otherwise perform one's identity in a particular way.

The stories in Part VI.A should make clear that some instances of identity performance or role playing demands can impose catch 22's on those subject to them. I suggest that these catch 22's should highlight the fact that we simply cannot categorically dismiss the regulation of identity performance as only influencing individual choices in a trivial manner. In the case of the Assimilating Firm above, the lesbian Asian woman was confronted with logically incoherent demands--the demand to be uninhibited and social about sex, and the demand to be inhibited and silent about sex. These demands were not evenhanded because they asked of the employee that she overcompensate. In other words, she was required to meet a higher bar than others. The first demand was a demand to overcompensate for being Asian, while the second demand was a demand to overcompensate for being lesbian.

Unfortunately, these demands in isolation are easily disguised as innocent, and even when they are explicitly differentiated on the basis of factors like sex, they are often understood as trivial without extra evidence otherwise, as they were in Jespersen, where the plaintiff was told that her sex-stereotyping claim was unworthy in part because she was not harassed, implying that she was not seriously harmed. (138) Simply asking of employees that they be uninhibited about sex does not seem to deserve condemnation from the person who only condemns more traditional forms of race discrimination. Also, simply asking of employees that they be inhibited and silent about sex seems equally fair. (I leave aside questions of whether either of these might constitute some form of sexual harassment. Without more, I believe that they do not.) (139)

However, once we take a look at the case of intersectionals, the incoherence of the demands placed on them--the catch 22 or double bind--makes clear first that when identity performance demands come in the form of higher or different burdens placed on some employees than others, the impact can be much more than trivial. Second, the catch 22's experienced by intersectionals can expose the unreasonableness of identity performance demands that come into conflict with other identity performance demands. I do not argue that the presence of a catch 22 is either necessary or sufficient to prove unreasonability, and therefore do not argue that something like a rule requiring fast typing with few mistakes is unreasonable. However, I do propose that catch 22's created by identity performance demands should make us suspicious of the supposed reasonableness of an employer's choice to make those demands, just as they made the Supreme Court suspicious in Price Waterhouse. (140)

In other words, the compounding of the different forms of discrimination in the case of intersectionals results in more visibly restrictive terms of employment, highlighting the injury, while the conflict of the different forms of discrimination results in an embarrassing incoherence on the part of the employer, highlighting irrationality and prejudice.

Another example may be of assistance here. Suppose an employer requires its employees to dress somewhat formally, but the standard is vague. There is a "business casual" rule. Suppose the employer also requires its employees to not flaunt their sexual orientation if they are lesbian, gay, or bisexual. Finally, the employer perceives traditionally feminine clothes, like dangling earrings, flared skirts, and open-toed sandals, as less formal than traditionally masculine clothes, like button down shirts and tailored pants. These two rules--(1) do not flaunt your sexuality, and (2) masculine clothes are more formal than feminine clothes--might appear as reasonably innocent conformist demands each on their own. To reward women who assimilate to an androgynous style of dress seems somewhat innocent, and perhaps even progressive. It certainly does not seem to necessarily have a cultural meaning that denigrates women. (141) Let us suppose though, that without even realizing it, the employer actually has a double standard for people of color and white employees. People of color who dress on the less formal side, such as women who wear flared skirts and dangling earrings, or men who wear polo shirts instead of long-sleeve shirts, are more frequently perceived as not being serious, and are subconsciously associated with the many support staff of color who are not required to dress as formally as professional staff. Subsequently, they are given less responsibility and less opportunity to demonstrate their skills. But white professional staff who dress less formally are not subconsciously associated with support staff, and therefore experience no loss of responsibility. Such a subtle double standard may be difficult for the employer, a court, an observer who condemns discrimination, or even an employee of color to perceive. (142) Professional employees of color may feel the effects slightly, but may be unsure just how severe or substantial the double standard is.

What happens, however, when a lesbian or bisexual woman of color comes to work in this environment? This intersectional is subject to three assimilationist demands: (1) do not flaunt that you are a lesbian or bisexual, or you will be fired; (2) dress professionally; the more masculine you dress, the better you are meeting the standard; (3) (Implicitly) dress very professionally, because you are a person of color. These three demands translate into: (1) dress femininely; (2) do not dress too femininely; (3) dress very masculinely. Even demands 1 and 2--the explicit demands-are somewhat contradictory, but easily dismissed by some as trivially so. After all, nobody is going to tangibly punish a queer employee for dressing somewhat femininely, since there is considerable freedom to do so at this workplace. All she has to do is dress somewhat femininely in order to avoid incurring any risk that she will be fired or suffer any tangible loss. However, once demand 3 enters the picture, this employee does experience a tangible punishment if she dresses somewhat femininely. She is given less responsibility.

There are two clear ways for her to resolve these incoherent demands: (1) Convincingly pretend to be hetero or convincingly pretend to be white (this would allow her to either dress more masculinely, or dress more femininely, respectively, without punishment); or (2) Convincingly pretend to be a man (this would allow her to dress masculine without punishment). However, asking her to do either of these would be discrimination, because these are requirements to pass.

Again, I do not claim these are the only two mechanisms by which she could negotiate the incoherent demands. Rather, I claim these are the mechanisms by which the conflict in demands could be removed. There is probably some way for this employee to satisfy the three demands. Perhaps she can wear extremely feminine shoes while wearing masculine clothes. Maybe she can wear makeup with masculine clothes, as well as a very feminine hairstyle. She might also just wear feminine clothes but be extremely industrious in order to prove her skills beyond a doubt. That however, is somewhat beside the point. The point is that, like Hopkins, this woman is subject to incongruent demands that make it obvious that at least one of these demands cannot be fairly placed on all employees. In this case, there are actually two demands that can be characterized as unfair disparate treatment. First, do not flaunt your sexuality, though straight employees are free to flaunt. Second, dress hyper-professionally, though white employees are free from this constraint.

These multiple demands place the intersectional employee in a highly constrained situation, and therefore highlight the injury. They also subject the intersectional to irrational, conflicting directions, proving that they simply do not stem from some kind of arbitrary yet rational set of cultural norms that the employer desires to promote. While it takes an understanding of social and historical context to determine whether the set of cultural norms being promoted is both racially and sexually based, the catch 22 should make us care that it is so by identifying the harm of these demands--an unreasonable constraint on human choice. (143)


In this Part, I identify and address a few objections to my argument. Many may object that, of course, the particular conformist demands I have identified are discriminatory because they constitute disparate treatment of people based on irrelevant factors such as race and gender. They may object that therefore, these conformist demands are already captured by traditional antidiscrimination rules, given the proper evidence of disparate treatment. In fact, it may be wrong to even call these demands "covering," "conformist," "assimilationist," or "identity work"--they are just regular old discriminatory treatment. Critics would say that it is disingenuous to group the demand that Asian men overcompensate and meet a higher standard than white men along with the demand that all employees be aggressive. These demands are very different and should be thought of as such.

I do not object to the point that these demands are very different. In fact, my point is precisely that these demands are very different, although they might appear similarly arbitrarily conformist on an initial read or observation. Even demands that constitute explicit disparate treatment, such as gender differentiated appearance requirements, have been permitted by certain courts as somehow not harmful, or trivially so, even after Price Waterhouse. (144) Assessing when a demand is "facially neutral" with respect to categories such as race, gender, and sexual orientation is virtually impossible, as demonstrated by the facts of Price Waterhouse. What one person calls disparate treatment, e.g., "women only must act feminine," another calls evenhanded conformist requirements, e.g., "everyone act gender appropriate," or even "everyone dress 'professionally,' 'normally,' or 'according to accepted standards,'" where terms like "normal" and "standards" incorporate notions such as femininity.

Moreover, even instances of identity performance demands that are applied evenhandedly are significant in that they can disparately impact intersectionals' space for negotiation and choice.

Traditionally, disparate impact refers to employment policies which exclude members of one group disproportionately. Such policies are forbidden under Title VII unless they are a "business necessity." (145) However, the disparate impact doctrine is less often applied to policies which disproportionately change the behavior of members of one group, or disproportionately regulate one group's preferred mode of dress. Under traditional understandings of discrimination, the group members have no claims to behave the way they would like on the job, and therefore, no claim of disparate impact where the aspect of their lives impacted is mutable. (146)

For those who agree with the current state of affairs, the case of intersectionals and the catch 22's they experience may help serve as one reason why we should care about impact on mutable behaviors. The harm of even a universally imposed identity performance demand is very significant when it comes into conflict with demands that are not evenhanded.

Almost all of us have felt the need to negotiate identity performance demands that are in tension with each other, whether at work, at school, in job interviews, or with family. This is unsurprising given that our various commitments are to different people and institutions, all of which demand different behaviors from us. But sometimes the same entity, be it an employer, a spouse, a friend, or a person with whom we make financial transactions, makes self-contradictory demands of us, and this is particularly frustrating. This truly irrational kind of obligation has been imposed on many of us, a fact which is unsurprising given the fact that everyone is an intersectional on some level. (147) Rather than assume this kind of madness is simply part of life, we should be interrogating it and treating it as evidence of discrimination.

Just as Ann Hopkins' very existence at Price Waterhouse was something of an incongruity, intersectionals' very existence in many walks of life are something of an incongruity. Their very presence can teach private actors, courts, and the public something about themselves and something about discrimination. In particular, their presence helps to highlight how identity performance regulation can be problematic even under the core of our socially shared principles concerning discrimination and fairness.

Another objection may come from those who promote intersectional analysis. This criticism would point out that my approach may strip intersectional analysis of one of its most important assets, contextual humility and acknowledgement. My analysis might be criticized as turning a historically and socially contextual and contingent analysis into an oversimplified one that describes discrimination as occurring along single axes in order to characterize the resulting stereotypes as "conflicting." This criticism would point out that stereotypes like the welfare queen or the Black matriarch might be best characterized as stereotypes and demonizations specific to African American women, rather than conflicts between stereotypes of African Americans and stereotypes of women. (148) This critique would argue that because racism and sexism are gendered and racial, respectively, attempts to isolate the components of intersectional stereotypes along single axes of discrimination are too great an oversimplification.

The response is simple; this Article's characterization of irrational stereotypes and demands is by no means the only appropriate one. However, in the context of identity performance demands and regulation there is a clear benefit to articulating harms in terms of this model of "conflict" between discrimination along single axes. Rather than seek to treat race, gender, and the like as monolithic categories, this Article seeks to use single-axis models to actually render clear the fact that discrimination is incredibly complex, gendered, raced, and, most important, thoroughly irrational. Mine is not the best or only mode of intersectional analysis. Rather, I think it is a neglected form of intersectional analysis that deserves attention.


I have hoped, in this Article, to use the case of Price Waterhouse to remind us that what some think of as formally discriminatory, like the requirement to gender conform, others think of as formally nondiscriminatory. I also hope the case reminds us of the importance of legal suspicion of irrationality, especially in antidiscrimination law. The case of intersectionals demonstrates to both discriminators and anti-discriminators that what may be perceived as fair treatment can in fact entail significant harm.

Another aim of this Article is to revitalize the field called intersectionality analysis. The conflicts experienced by intersectionals, conflicts that monolithic and stereotyped understandings of identity give rise to, lead to erasure and ignorance of their circumstances. But these conflicts are also of the incoherent sort that, especially in legal discourse, can expose our own presumptions and unfairness to the light of day.


We can depict how the intersectional brings the harm to light using the following figures.


In the figures, the shaded areas represent the type of dress or behavior that is "off-limits" for a particular group of persons. For instance, the areas shaded by vertical lines represent the behavior that is off limits to all men--dressing or acting too femininely. The area shaded by horizontal zig-zag lines represent the additional behavior that is off limits to queer men, who must overcompensate and act particularly masculine. Some shaded areas represent behavioral restrictions or demands that derive from demands to reverse-cover, or perform one's identity stereotypically, while others represent assimilative restrictions and demands. The field of allowable behavior, the leftover white crescent, gets very small in the case of some intersectionals, especially queer women of color, and this makes the discriminatory treatment both more tangibly harmful and more noticeable.

Of course, these pictures seem to quantify what may be unquantifiable, and the sizes of the various shaded areas are quite arbitrary. These figures are not meant to represent any empirical findings. Rather, they are meant to illustrate how intersectionals can highlight the injury inflicted by multiple identity performance demands and how intersectionals can demonstrate the incoherence of many identity performance demands. The diminishing white space in the pictures for intersectionals depicts their diminishing space for negotiation. The opposing directions of the shaded crescents depict catch 22's: the conflicting directions in which certain intersectionals--in this example, straight and queer women of color--are pulled.

(1) See Devon W. Carbado & Mitu Gulati, Working Identity, 85 CORNELL L. REV. 1259, 1262, 1293-98 (2000); Camille Gear Rich, Performing Racial and Ethnic Identity: Discrimination by Proxy and the Future of Title VII, 79 N.Y.U. L. REV. 1134, 1200-02 (2004); Kenji Yoshino, Covering, 111 YALE L.J. 769, 781 (2002).

(2) See supra note 1.

(3) Yoshino, supra note 1, at 772.

(4) Id. at 906.

(5) Id. at 930-33.

(6) Carbado & Gulati, supra note 1, at 1262, 1279.

(7) See supra note 1. See Price Waterhouse v. Hopkins, 490 U.S. 228, 251 (1989); Yoshino, supra note 1, at 913.

(9) For a similar conception of identity as a choice rather than as a "discovered" fact, see AMARTYA SEN, THE ARGUMENTATIVE INDIAN: WRITINGS ON INDIAN HISTORY, CULTURE AND IDENTITY 350-51 (2005).

(10) Kimberle Crenshaw, Demarginalizing the Intersection of Race and Sex: A Black Feminist Critique of Antidiscrimination Doctrine, Feminist Theory and Antiracist Politics, 1989 U. CHI. LEGAL F. 139, 140, 149 (1989) [hereinafter Crenshaw, Demarginalizing the Intersection of Race and Sex]; Kimberle Crenshaw, Mapping the Margins: Intersectionality, Identity Politics, and Violence Against Women of Color, 43 SWAN. L. REV. 1241, 1242-44 (1991) [hereinafter Crenshaw, Mapping the Margins].

(11) See Katharine T. Bartlett, Only Girls Wear Barrettes: Dress and Appearance Standards, Community Norms, and Workplace Equality, 92 MICH. L. REV. 2541, 2543-45 (1994) (describing "judicial reliance on community norms").

(12) See, e.g., Robert S. Chang & Jerome McCristal Culp, Jr., After Intersectionality, 71 UMKC L. REV. 485, 485, 489 (2002); Nancy Ehrenreich, Subordination and Symbiosis: Mechanisms of Mutual Support Between Subordinating Systems, 71 UMKC L. REV. 251, 25253, 265-66 (2002); Trina Grillo, Anti-Essentialism and Intersectionality: Tools To Dismantle the Master's House, 10 BERKELEY WOMEN'S L.J. 16, 17-20 (1995); see also supra note 10.

(13) See Darren Lenard Hutchinson, Identity Crisis: "Intersectionality," "Multidimensionality," and the Development of an Adequate Theory of Subordination, 6 MICH. J. RACE & L. 285, 308 (2001) (describing intersectionality as having "destabilized traditional attempts to treat oppressed classes as monolithic groups").

(14) See Clark Freshman, Whatever Happened to Anti-Semitism? How Social Science Theories Identify Discrimination and Promote Coalitions Between "Different" Minorities, 85 CORNELL L. REV. 313, 327 (2000) ("Intersectionality divides existing classifications like race and gender into hybrid categories such as 'women of color' and 'black men.").

(15) See, e.g., Regina Austin, Sapphire Bound!, 1989 WIS. L. REV. 539, 539-40, 545-46, 569 72 (1989) (noting that "[a]nti-racist or anti-sexist scholarship that is overinclusive and abstract is dangerous because it misconceives the often knotty structural nature of the conditions that are its subject," and contextualizing Chambers v. Omaha Girls Club, Inc., 834 F.2d 697 (8th Cir. 1987), with histories of the stereotypes Sapphire, Mammy, and Jezebel); Paulette M. Caldwell, A Hair Piece: Perspectives on the Intersection of Race and Gender, 1991 DUKE L.J. 365, 390-92 (1991) (discussing the centrality of hair in literary accounts of the "[d]egradation of [b]lack [w]omen," and noting Rogers v. Am. Airlines, 527 F. Supp. 229 (S.D.N.Y 1981), a Title VII case in which a plaintiff who was not permitted to wear her hair in braids lost); Dorothy E. Roberts, Punishing Drug Addicts Who Have Babies: Women of Color, Equality, and the Right of Privacy, 104 HARV. L. REV. 1419, 1424 (1991) (presenting an analysis presuming that "Black women experience various forms of oppression simultaneously, as a complex interaction of race, gender, and class that is more than the sum of its parts," and stating that, "[i]t is impossible to isolate any one of the components ... or to separate the experiences that are attributable to one component from experiences attributable to the others.... Their devaluation as mothers ... has its roots in the unique experience of slavery and has been perpetuated by complex social forces.") (citations omitted).

(16) See Roberts, supra note 15, at 1438. The Jezebel stereotype portrays Black women as being "sexually promiscuous." Id.

(17) 490 U.S. 228 (1989).

(18) See id. at 251 (stating that the plaintiff would have been discriminated against for acting masculine and for acting too feminine).

(19) Id.

(20) Rich, supra note 1, at 1185-86.

(21) Id. at 1180.

(22) See id. at 1195-96 (arguing that "judges conducting Title VII inquiries must interrogate aggressively the motives behind policies that disproportionately burden or effectively screen out minority workers because of voluntary race/ethnicity-associated behavior").

(23) Yoshino, supra note 1, at 772.

(24) Id.

(25) See id. at 781.

(26) 42 U.S.C. [section] 2000e-2(a)(1) (2000).

(27) Abigail W. Lloyd, Defining The Human: Are Transgendered People Strangers To The Law?, 20 BERKELEY J. GENDER L. & JUST. 150, 190-91 (2005).

(28) Yoshino, supra note 1, at 772.

(29) Id. at 781.

(30) Id. at 772.

(31) Id. at 781.

(32) E.g., Carbado & Gulati, supra note 1, at 1262, 1279; Rich, supra note 1, at 1180; Kenji Yoshino, Assimilationist Bias in Equal Protection: The Visibility Presumption and the Case of "Don't Ask, Don't Tell," 108 YALE L.J. 485, 500 (1998). (33) See Yoshino, supra note 1, at 931.

(34) Id. at 930-33.

(35) Id. at 933.

(36) Cf. Phillips v. Martin Marietta Corp., 400 U.S. 542, 544 (1971) (Marshall, J., concurring) (refusing to permit "one hiring policy for women and another for men--each having preschool-age children").

(37) See, e.g., Holly D. v. Cal. Inst. of Tech., 339 F.3d 1158, 1169 (9th Cir. 2003) (holding that, under Title VII, when an employee submits to a supervisor's demands for sex and the supervisor then allows the employee to retain her job, gain a promotion, or otherwise benefit from the submission, "a tangible employment action" has been taken because an additional condition of employment has been imposed).

(38) See, e.g., Christine Jolls, Antidiscrimination and Accommodation, 115 HARV. L. REV. 642, 643-45 (2001).

(39) Cf. Rogers v. Am. Airlines, Inc., 527 F. Supp. 229, 231-33 (S.D.N.Y. 1981) (holding that an employer policy prohibiting braided hairstyles did not discriminate on the basis of race or sex under Title VII).

(40) For instance, much gangsta rap includes use of not only "obscene" words, but arguably racist, homophobic, and misogynist lyrics. See generally Gangsta Rap, (last visited Aug. 18, 2005). Of course, the question whether gangsta rap can actually lay claims to validly representing black culture is a matter of great controversy, with commentators taking a range of viewpoints, from the position that it depicts the "reality" of ghetto life, to the position that it simply internalizes racist myths and stereotypes, like "minstrel shows." Id. This Article does not take any position with respect to this debate. Rather, it notes that a plaintiff could create an issue of fact with respect to whether a particular use of the word "bitch" is a performance of racial identity.

(41) Yoshino, supra note 1, at 898-99.

(42) Roberto J. Gonzalez, Cultural Rights and the Immutability Requirement in Disparate Impact Doctrine, 55 STAN. L. REV. 2195, 2206-07 (2003).

(43) See SEN, supra note 9 (disputing characterization of identity as something to be "discovered").

(44) The problems of essentialism and a slippery slope do, however, tell us a great deal about what form the law's approach to these problems should take. Elsewhere, I argue for a rights-based approach, in which certain kinds of identity performance, namely personal appearance choices, are protected as fundamental liberty interests. See Gowri Ramachandran, Freedom of Dress: State and Private Regulation of Clothing, Hair, Makeup, Tattoo, and Piercing Choices (unpublished manuscript), available at In that article, I argue that such an approach is preferable to the traditional equal-treatment rhetoric of antidiscrimination law. Here, however, I focus only on the harm entailed by identity performance demands.

(45) Compare Jefferies v. Harris County Cmty. Action Ass'n, 615 F.2d 1025, 1032, 1034 (5th Cir. 1980) (holding that plaintiffs can state a claim for combined race and sex discrimination under Title VII, and that the use of the word "or" in the statute evidences Congressional intent to permit claims of combined forms of discrimination), with DeGraffenreid v. Gen. Motors Assembly Div., 413 F. Supp. 142, 143, 145 (E.D. Mo. 1976), aff'd in part, rev'd in part, 558 F.2d 480 (8th Cir. 1977) (holding that plaintiffs cannot state such a claim because to do so creates a "super-remedy" beyond the intention of the statute's drafters). See also Virginia Wei, Note, Asian Women and Employment Discrimination: Using Intersectionality Theory to Address Title VII Claims Based on Combined Factors of Race, Gender and National Origin, 37 B.C.L. REV. 771, 775-76 (1996) (describing the ambiguity in Title VII as to whether it applies to discrimination on the basis of multiple factors).

(46) Under the sex-plus rule, plaintiffs must demonstrate that they were discriminated against on the basis of sex and "either an immutable characteristic or the exercise of a fundamental right." Arnett v. Aspin, 846 F. Supp. 1234, 1238, 1241 (E.D. Pa. 1994) (explaining the rule, citing Phillips v. Martin Marietta Corp., 400 U.S. 542, 544 (1971) and its progeny, and applying it as a bar to "sex-plus-age" discrimination); see Phillips, 400 U.S. at 544 (holding that the Court of Appeals erred in holding that section 703(a) of the Civil Rights Act of 1964 permitted "one hiring policy for women and another for men--each having preschool age children"). But, discrimination on the basis of sex and some other factor has been upheld. See, e.g., Fountain v. Safeway Stores, Inc., 555 F.2d 753, 755-56 (9th Cir. 1977) (upholding a policy that required men, but not women, to wear ties); Willingham v. Macon Tel. Publ'g Co., 507 F.2d 1084, 1092 (5th Cir. 1975) (upholding a policy that prohibited men having long hair, but not women).

(47) E.g., Lam v. Univ. of Haw., 40 F.3d 1551, 1562 (9th Cir. 1994) (holding that claims of combined race and sex discrimination require analysis of whether employer discriminated on the basis of both factors combined, rather than a determination of whether persons of the same race or the same sex were discriminated against); Hicks v. Gates Rubber Co., 833 F.2d 1406, 1416 (10th Cir. 1987) (holding that racial slurs should be considered as evidence of a hostile environment within the context of a sexual harassment claim).

(48) Lawrence v. Texas, 539 U.S. 558, 578 (2003) (overturning Bowers v. Hardwick, 478 U.S. 186 (1986), and holding a statute criminalizing consensual same-sex sodomy in the privacy of the home unconstitutional).

(49) See, e.g., N.Y. EXEC. LAW [section] 296(1)(a)-(c) (McKinney 2005).

(50) See, e.g., Mary Jo Wiggins, Foreword: The Future of Intersectionality and Critical Race Feminism, 11 J. CONTEMP. LEGAL ISSUES 677, 677-78 (2001); Hutchinson, supra note 13.

(51) See Ehrenreich, supra note 12, at 265-66; Grillo, supra note 12, at 18; see also Caldwell, supra note 15, at 372, 374.

(52) Crenshaw, Demarginalizing the Intersection of Race and Sex, supra note 10, at 166; Crenshaw, Mapping the Margins, supra note 10, at 1242-44.

(53) Crenshaw, Demarginalizing the Intersection of Race and Sex, supra note 10, at 140, 152.

(54) See Ehrenreich, supra note 12, at 265-66 ("[T]he two categories could not be added together like combining yellow and blue tennis balls in a bucket. Rather, their position at the bottom of racial and gender hierarchies combined to produce a new and different set of concerns and characteristics ... more like combining yellow and blue paint.") (emphasis in original).

(55) See, e.g., Angela P. Harris, Race and Essentialism in Feminist Legal Theory, 42 STAN. L. REV. 581, 594, 608-09 (1990) (focusing on the intersectional experience of African-American women); Mari J. Matsuda, Beside My Sister, Facing the Enemy: Legal Theory Out of Coalition, (43) SWAN. L. REV. 1183, 1189-91 (1991) (analyzing intersectionals' ability to overcome subordination through coalitions); Wiggins, supra note 50, at 677-78.

(56) For example, in the conclusion of one of her articles, Judith Resnik states:</p> <pre> Serious questions, constant within feminism, remain about how to shape such equality demands and about which women will benefit. The category "women"--like the others discussed herein--is neither unitary nor necessarily permanent.... Further, provisions that may benefit one group of women may not serve others of differing classes and races.... Transnational rights advanced in the name of women must also be interrogated to understand how their applications vary. </pre> <p>Judith Resnik, Categorical Federalism: Jurisdiction, Gender, and the Globe, 111 YALE L.J. 619, 677 (2001) (citations omitted).

(57) See sources cited supra note 12.

(58) See, e.g., sources cited supra note 15.

(59) Cf. Mari J. Matsuda, Looking to the Bottom: Critical Legal Studies and Reparations, 22 HARV. C.R.-C.L.L. REV. 323, 335 (1987) (noting the tradition of persons of color appropriating ordinary discourse in order to seek justice, for instance, by "turn[ing] the Bible and the Constitution into texts of liberation").

(60) 490 U.S. 228, 251 (1988).

(61) Id. at 231-32, 257-58.

(62) Id. at 250-51.

(63) Id. at 233-34, 251, 255.

(64) Id. at 235.

(65) Cf. Baehr v. Lewin, 852 P.2d 44, 63 (Haw. 1993) (referring to the reasoning that laws against same-sex marriage formally treat men and women equally by requiring both to marry members of the opposite sex as an "exercise in tortured and conclusory sophistry"); Andrew Koppelman, Is Marriage Inherently Heterosexual?, 42 AM. J. JURIS. 51, 54 (1997) (describing the argument that miscegenation laws formally treated both races equally as "sill[y]").

(66) See Koppelman, supra note 65, at 54-55 (describing same-sex marriage cases prior to Baehr in which this kind of reasoning prevailed, and noting that the dissenting judge in Baehr was persuaded by this argument).

(67) See, e.g., Fountain v. Safeway Stores, Inc., 555 F.2d 753, 755-56 (9th Cir. 1977) (upholding a policy that required men, but not women, to wear ties, and holding that different standards of dress for men and women are permissible); Willingham v. Macon Tel. Publ'g Co., 507 F.2d 1084, 1092 (5th Cir. 1975) (upholding a policy that prohibited men to have long hair, but not women).

(68) See, e.g., Willingham, 507 F.2d at 1091-92; Lanigan v. Bartlett & Co. Grain, 466 F. Supp. 1388, 1391-92 (W.D. Mo. 1979) (holding that the company was within its rights to implement a dress code policy within the executive offices).

(69) See, e.g., Nichols v. Azteca Rest. Enters. Inc., 256 F.3d 864, 875 n.7 (9th Cir. 2001) (noting that different standards of dress between genders may not be discrimination).

(70) Id. at 870.

(71) Id. at 875 n.7 ("We do not imply that all gender-based distinctions are actionable under Title VII. For example, our decision does not imply that there is any violation of Title VII occasioned by reasonable regulations that require male and female employees to conform to different dress and grooming standards."). But see Smith v. City of Salem, 378 F.3d 566, 569-70 (6th Cir. 2004) (holding that a one-day suspension based on gender nonconforming appearance and behavior was actionable under Title VII); Rosa v. Park W. Bank & Trust Co., 214 F.3d 213, 216 (1st Cir. 2000) (noting that '"stereotyped remarks [including statements about dressing more 'femininely'] can certainly be evidence that gender played a part"' (quoting Price Waterhouse v. Hopkins, 490 U.S. 228, 251 (1989)) (emphasis in original)); Higgins v. New Balance Athletic Shoe, Inc., 194 F.3d 252, 261 n.4 (1st Cir. 1999) ("[J]ust as a woman can ground an action on a claim that men discriminated against her because she did not meet stereotyped expectations of femininity, a man can ground a claim on evidence that other men discriminated against him because he did not meet stereotyped expectations of masculinity." (citation omitted)).

(72) Jespersen v. Harrah's Operating Co., 392 F.3d 1076, 1077, 1083 (9th Cir. 2004). But see id. at 1083 (Thomas, J., dissenting).

(73) Mary Ellen Maatman, Choosing Words and Creating Worlds: The Supreme Court's Rhetoric and Its Constitutive Effects on Employment Discrimination Law, 60 U. PITT. L. REV. 1, 21 (1998) (quoting Price Waterhouse, 490 U.S. at 272).

(74) Price Waterhouse, 490 U.S. at 235 (emphasis added).

(75) E.g., Higgins, 194 F.3d at 257 & n.1 (describing harassment of plaintiff including "vulgar and derogatory names.... obscene remarks," mocking, "squirt[ing] him with condiments, snapp[ing] rubber bands at him, and pour[ing] hot cement on him," as well as death threats and violent shaking).

(76) Jespersen, 392 F.3d at 1079, 1083.

(77) See id. at 1082-83.

(78) See id. at 1084 (Thomas, J., dissenting) (citing Oncale v. Sundowner Offshore Servs., 523 U.S. 75, 80 (1998)).

(79) See, e.g., Smith v. City of Salem, 378 F.3d 566, 567-68 (6th Cir. 2004) (holding in favor of the plaintiff only in the sense that the court noted the complaint stated a claim); Rosa v. Park W. Bank & Trust Co., 214 F.3d 213, 214 (1st Cir. 2000).

(80) See Thomas J. Goger, Annotation, Validity and Construction of Statute or Ordinance Respecting Employment of Women in Places Where Intoxicating Liquors Are Sold, 46 A.L.R.3d 369 (1972) (noting and summarizing the results of litigation in a wide variety of states and cities in which women were forbidden from working in places where alcoholic beverages were sold, bartending licenses were limited to men, or restrictions were placed on women bartending, such as requirements that their husbands be the proprietors of the establishment and be present).

(81) Price Waterhouse v. Hopkins, 490 U.S. 228, 251 (1989); Yoshino, supra note 1, at 780.

(82) Yoshino, supra note 1, at 913, 916.

(83) Id. at 916-17.

(84) See id. at 917.

(85) Id. at 905-06 (quoting Price Waterhouse, 490 U.S. at 251).

(86) Id. at 914-15, 918.

(87) Id. at 918.

(88) Id. at 910.

(89) I am indebted to Ian Ayres for this excellent example.

(90) By way of analogy, I point out that many scholars recognize the failure of concepts such as equality, rights, or stare decisis to answer how courts should rule in particular cases. Some argue for abandoning some, or all, of these forms of rhetoric. E.g., Peter Westen, The Empty Idea of Equality, 95 HARV. L. REV. 537, 592, 596 (1982) (arguing for the abandonment of the rhetoric of equality). But see, e.g., ROBIN L. WEST, RE-IMAGINING JUSTICE: PROGRESSIVE INTERPRETATIONS OF FORMAL EQUALITY, RIGHTS, AND THE RULE OF LAW 26-27 (2003) (arguing for a re-imagining of these concepts). One can find these concepts inadequate for structuring the law while believing they are still useful rhetorically.

I believe the catch 22 is not an appropriate legal litmus test for employer behavior, but it is useful as a means of interrogating employer behavior. Elsewhere, I propose what I believe is a workable legal approach to these issues, one that secures worker rights to personal appearance rather than using antidiscrimination law. See generally Ramachandran, supra note 44.

(91) See WEST, supra note 90, at 31; Christine Jolls et al., A Behavioral Approach to Law and Economics, 50 STAN. L. REV. 1471, 1548-50 (1998) (providing an appendix listing specific findings about bounded rationality and applications for these findings in the law); Cass R. Sunstein, Probability Neglect: Emotions, Worst Cases, and Law, 112 YALE L.J. 61, 65, 83 (2002) (illustrating that individuals divert from rational behavior theory when probabilities are small, as well as when outcomes are affect rich).

(92) See, e.g., Jed Rubenfeld, Affirmative Action, 107 YALE L.J. 427, 436-37 (1997) (citing JOHN HART ELY, DEMOCRACY AND DISTRUST 146 (1980)).

(93) See Chang & Culp, Jr., supra note 12, at 485; Crenshaw, Demarginalizing the Intersection of Race and Sex, supra note 10, at 139-40; Crenshaw, Mapping the Margins, supra note 10, at 1253; Ehrenreich, supra note 12, at 252-53; Grillo, supra note 12, at 17, 19-20.

(94) Francisco Valdes, Queers, Sissies, Dykes, and Tomboys: Deconstructing the Conflation of "Sex," "Gender," and "Sexual Orientation" in Euro-American Law and Society, 83 CAL. L. REV. 3, 179 (1995) ("Hopkins therefore prompts some reflection on the reason(s) why this plaintiff, situated so similarly to those in Smith and Strailey, nonetheless escaped conflationary branding as a lesbian...." (first alteration in original)).

(95) Marlon Riggs, Black Macho Revisited: Reflections of a SNAP! Queen, in BROTHER TO BROTHER: NEW WRITINGS BY BLACK GAY MEN 253, 254 (Essex Hemphill ed., 1991) (emphasis added).

(96) Ron Simmons, Tongues Untied: An Interview with Marlon Riggs, in BROTHER TO BROTHER: NEW WRITINGS BY BLACK GAY MEN 189, 190 (Essex Hemphill ed., 1991).

(97) Isaac Julien & Kobena Mercer, True Confessions: A Discourse on Images of Black Male Sexuality, in BROTHER TO BROTHER: NEW WRITINGS BY BLACK GAY MEN 167, 171 (Essex Hemphill ed., 1991).

(98) Kaushalya Bannerji, No Apologies, in A LOTUS OF ANOTHER COLOR: AN UNFOLDING OF THE SOUTH ASIAN GAY AND LESBIAN EXPERIENCE 59, 63 (Rakesh Ratti ed., 1993).

(99) Id. at 62.

(100) Id.


(102) Id. at 24 (quoting Sojourner Truth).

(103) Id. at 25.

(104) Id. at 113 (describing how the Jezebel and Mammy stereotypes served to resolve the dilemma).

(105) Id.

(106) See id. at 95, 96 (describing the historical use of oppression of women into "traditional gender roles" in attempts to prevent the "decline of civilization").

(107) Id. at 117.

(108) Id. at 97.

(109) Id. at 26.

(110) Id. at 117.

(111) Id. at 87.

(112) Id. at 97 (citing MICHAEL HARRINGTON, THE NEW AMERICAN POVERTY 179 (Henry Holt & Co. 1st ed. 1984)).


(114) Id. at 180.

(115) Id.

(116) Id.

(117) Id. at 134.

(118) See supra notes 113-16 and accompanying text.

(119) See supra note 117 and accompanying text.

(120) See, e.g., Angela P. Harris, Race and Essentialism in Feminist Legal Theory, 42 STAN. L. REV. 581, 584 (1990); Matsuda, supra note 59 at 333-42.

(121) See, e.g., Neil Gotanda, A Critique of "Our Constitution Is Color Blind," 44 SWAN. L. REV. 1, 37-42 (1991) (noting the Court's use of formal race, a concept that envisions race and historical Black oppression as unconnected, in cases involving, inter alia, voting rights, affirmative action and jury selection); see also WEST, supra note 90, at 109, 125.

(122) See generally WEST, supra note 90, at 111-19.

(123) See id. at 109-10.

(124) See, e.g., Richard A. Posner, Pragmatism Versus Purposivism in First Amendment Analysis, 54 STAN. L. REV. 737, 738-39, 745 (2002) (rejecting formalism as unpragmatic and endorsing judicial decisionmaking based on "what works," but conceding that "there is a substantial basis for pragmatic anxiety about doctrines that are so loose that they give judges carte blanche to decide cases any which way they want without inviting criticism that they are deviating from the previous course of decisions," and so promoting a pragmatic form of judicial decisionmaking that incorporates stare decisis in order to increase confidence in the judicial system).

(125) Chang & Culp, Jr., supra note 12, at 489 ("When talking about race or gender, it is nonsensical to talk about privilege and subordination as operating uniformly or unidirectionally (as if along a single axis)."); Ehrenreich, supra note 12, at 265 ("[U]nder an additive approach, the situation of women of color is understood as the combined product of sexism ... and racism.... Female writers of color rejected this set of understandings...."); Grillo, supra note 12, at 18 ("[W]omen of color stand at the intersection of the categories of race and gender, and [] their experiences are not simply that of racial oppression plus gender oppression.").

(126) See, e.g., Matsuda, supra note 59, at 328-29; WEST, supra note 90, at 109-10.

(127) See, e.g., Riggs, supra note 95, at 254-55 ("Indeed the representation of Negro faggotry disturbingly parallels and reinforces America's most entrenched racist constructions around African American identity.... Majority representations of both affirm the view that blackness and gayness constitute a fundamental rupture in the order of things....").

(128) MULLINGS, supra note 101, at 111-13.

(129) See SIRCAR, supra note 113, at 133-34 ("She added that being Asian Indian is a handicap because Asian Indian women do not party as much or socialize with coworkers as much.... D, a software consultant ... mentioned some of her White counterparts playing up to their male bosses and being rewarded in terms of juicier assignments.").

(130) See Vicki Schultz, The Sanitized Workplace, 112 YALE L.J. 2061, 2139, 2143-44, 2149-50 (2003) (valuing sex in the workplace and providing examples of sexualized workplaces in which such activity is beneficial to the workers and does not constitute sexual harassment, since such workplaces are gender integrated and the manner in which the workplace is sexualized does not exclude or damage any one gender).

(131) See supra notes 98-100 and accompanying text (describing stereotypes of South Asian women as soft-spoken and docile).

(132) See Crenshaw, Mapping the Margins, supra note 10, at 1243-44.

(133) 1980-2000: Coming Out Black, in BLACK LIKE US: A CENTURY OF LESBIAN, GAY, AND BISEXUAL AFRICAN AMERICAN FICTION 272 (Devon W. Corbado et al. eds., 2002).

(134) Id.

(135) Id.

(136) Rachel F. Moran, The Elusive Nature of Discrimination, 55 STAN. L. REV. 2365, 2394-95 (2003) (reviewing IAN AYRES, PERVASIVE PREJUDICE? UNCONVENTIONAL EVIDENCE OF RACE AND GENDER DISCRIMINATION (2001) and CROSSROADS, DIRECTIONS, AND A NEW CRITICAL RACE THEORY (Francisco Valdes et al. eds., 2002)) (arguing that courts have not allowed Black women to represent either blacks or women, and that Black women experience "double-discrimination").

(137) Many legal scholars and writers in the popular media are increasingly becoming interested in and citing studies based on the "Implicit Attitudes Test," (IAT) a demo of which can be taken at Project Implicit, (last visited Nov. 10, 2005). The IAT helps to unearth unconscious bias and racism. A list of press reports on the results is available at Implicit Association Test, In The Media, (last visited Nov. 9, 2005).

(138) Jespersen v. Harrah's Operating Co., Inc., 392 F.3d 1076, 1082-83 (9th Cir. 2004).

(139) See Schultz, supra note 130, at 2145-51.

(140) See Price Waterhouse v. Hopkins, 490 U.S. 228, 252 (1989).

(141) Cf. Charles R. Lawrence, III, The Id, the Ego, and Equal Protection: Reckoning With Unconscious Racism, 39 STAN. L. REV. 317, 355-56 (proposing a "cultural meaning" test for discrimination rather than a motive-based test of discrimination).

(142) Cf. Ian Ayres, Is Discrimination Elusive?, 55 STAN. L. REV. 2419, 2423-24 (2003) (arguing that proving intersectional discrimination empirically, both quantitatively and non-quantitatively, is exceedingly difficult, because it plays into the device of the discriminator, as well as because of a "degrees of freedom problem").

(143) See infra app. A (graphically representing various intersectional situations).

(144) See Jespersen v. Harrah's Operating Co., Inc., 392 F.3d 1076, 1078, 1083 (9th Cir. 2004) (upholding an employer's requirement that female employees wear make-up against a Title VII challenge, because the employee failed to show that the "policy impose[d] unequal burdens on male and female employees").

(145) 42 U.S.C. [section] 2000e-2(k)(1)(A)(i) (2000).

(146) Roberto Gonzalez has made this point, and has argued that rather than create dangerously essentialist "covering" claims under Title VII's disparate treatment doctrine, we should change disparate impact standards so that they apply to mutable behaviors. Roberto J. Gonzalez, Cultural Rights and the Immutability Requirement in Disparate Impact Doctrine, 55 SWAN. L. REV. 2195, 2216-17 (2003). However, this solution still carries with it the slippery slope problem, given than an employer's only defense to a properly proven disparate impact claim is that the policy or set of policies in question are a "business necessity." Think of a non-harassment policy that disparately impacts evangelical Christian employees, who might be more likely to run afoul of that policy by harassing lesbian and gay coworkers. Cf. Bodett v. CoxCom, Inc., 366 F.3d 736, 740-42 (9th Cir. 2004) (involving a claim by an evangelical Christian supervisor that she was the victim of religious discrimination after being dismissed for repeatedly trying to "convert" a lesbian subordinate). It would be extremely difficult for an employer to claim that its policy is a business necessity.

(147) Ehrenreich, supra note 12, at 275. Therein stating:</p> <pre>

Combining intersectionality insights with a recognition that each

individual sits at the intersection of multiple identities leads to the conclusion that each plays the role of oppressed in some

contexts and oppressor in others.... Thus, each individual is a

complex mix of subordinated and dominant statuses; it is virtually

impossible for someone to always find herself on the bottom of

every social hierarchy. </pre> <p>Id. at 275; see also Grillo, supra note 12, at 18 ("[I]n fact we all stand at multiple intersections of our fragmented legal selves.... In every set of categories there is not only subordination, but also its counterpart, privilege.").

(148) See, e.g., Roberts, supra note 15, at 1424 (arguing that the discrimination faced by Black women cannot be separated into the components of discrimination against women and discrimination against Blacks).

Gowri Ramachandran, The following people have provided extremely helpful advice and comments: Anshul Amar, Ian Ayres, Bill Bratton, Tomiko Brown-Nagin, Devon Carbado, Frank Rudy Cooper, Eric Fleisig-Greene, Mitu Gulati, Jerry Kang, Chuck Lawrence, David Luban, Carrie J. MenkelMeadow, Mari Matsuda, Gary Peller, Janani Ramachandran, Jed Rubenfeld, Michael Seidman, Robin West, and Kenji Yoshino. In addition, I am grateful for comments from students in the "Interdisciplinary Perspectives on Discrimination" seminar at University of Virginia Law School.
COPYRIGHT 2005 Albany Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Ramachandran, Gowri
Publication:Albany Law Review
Date:Dec 22, 2005
Previous Article:Redefining due process analysis: Justice Anthony M. Kennedy and the concept of emergent rights.
Next Article:At last, some clarity: the potential long-term impact of Lingle v. Chevron and the separation of takings and substantive due process.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters