Reassessing Gender Neutrality

Abstract

Since the 1970s, advocates have used the term gender neutral to press for legal change in contexts ranging from employment discrimination to marriage equality to public restroom access. Drawing on analyses of all Supreme Court cases, federal courts of appeals cases, and Supreme Court amicus briefs in which the terms gender neutral/neutrality, sex neutral/neutrality, or sexually neutral/sexual neutrality appear, this study examines how US courts have defined gender neutrality and what the scope and limits of its legal application have been. We find that the courts have defined gender neutrality narrowly as facial neutrality, but nonetheless that this limited understanding has transformed some areas of the law, even if it has had little impact on others. Our analysis confirms earlier feminist skepticism about the sufficiency of gender neutrality to guarantee equality but also points to areas in which the law has yet to exploit the idea's significant potential to address discrimination on the basis of sex, sexual orientation, and gender identity.

The 1970s mark a critical turning point in the history of US sex equality jurisprudence. In Reed v. Reed (1971), the Supreme Court for the first time held a sex-based classification in violation of the Equal Protection Clause, declaring unconstitutional an Idaho law granting automatic preference to husbands over wives in the administration of family estates. In the years since this landmark ruling, gender neutrality has emerged as a guiding principle in sex discrimination cases (Appleton 2005; Colker 1987). According to O'Connor, “the existence of comparable or superior sex-neutral alternatives has been a powerful reason to reject a sex-based classification” (Tuan Anh Nguyen v. I.N.S. 2001: 82). As the first systematic study of the term gender neutrality in the law, this article offers new insight into the meaning and usage of a key concept in modern sex equality jurisprudence. Our analysis provides the basis for a critical reassessment of its impact and charts a course for its future development in sex, sexual orientation, and gender identity discrimination cases.¹

Incorporation of the principle of gender neutrality into sex equality jurisprudence has transformed family, employment, and sexual violence law (Appleton 2005; Case 1999; Goldscheid 2014). In the family law context, the courts have invalidated provisions that allow former wives—but not husbands—to collect alimony (Orr v. Orr 1979) and that limit benefits associated with the provider role to men (Stanton v. Stanton 1975; Weinberger v. Wiesenfeld 1975). They have likewise declared unconstitutional laws that limit the right to manage community property to men (Kirchberg v. Feenstra 1981) or that deny unmarried fathers (but not mothers) the right to care for a child without a hearing on parental fitness (Stanley v. Illinois 1972). The courts have invalidated laws containing sex-based conditions of employment, such as a California law prohibiting women from being employed as bartenders, a decision cited with approval by the Supreme Court (Sail'er Inn v. Kirby 1971). The Supreme Court has further held that, under Title VII of the Civil Rights Act of 1964, an employer may not refuse to hire women with preschool-age children while hiring similarly situated men (Phillips v. Martin Marietta Corp. 1971) and that the military may not deny benefits to military husbands that are granted to military wives (Frontiero v. Richardson 1973). Finally, the First Circuit has invalidated laws that make it a felony for a man to have sexual intercourse with a consenting underage girl without making it a crime for adult women to have sexual intercourse with underage boys (Meloon v. Helgemoe 1977).

Despite the significant impact of gender neutrality in the law, feminist scholars today generally regard the principle as suspect. Historically, demands for “equality under the law” have been a centerpiece of the struggle for women's rights (Grimké 1988). But as women have achieved greater legal parity with men, feminists have evinced doubt about formal legal equality as an antidote to gender-based subordination. Since the 1980s, the prevailing view of gender neutrality is that it not only fails to guarantee equality, but that it actually works to confer legitimacy on an inequitable status quo (Baer 2008; Fineman 1983; MacKinnon 1987; Weitzman 1985; Williams 1989). In 2005, legal scholar Martha Fineman defined feminism itself as a theory that “challenges the assertions and assumptions of gender-neutrality and objectivity in received disciplinary knowledge” (Fineman 2005). Commentators also have noted the persistent refusal of the courts to acknowledge the relevance of the principle of gender neutrality in cases involving discrimination on the basis of sexual orientation, such as prohibitions on same-sex marriage, or in cases of discrimination against transgender people. These limitations help explain why the term gender neutrality rarely receives more than passing reference in accounts of modern sex equality jurisprudence, despite novel and sweeping application of the principle since the 1970s.

An unfavorable consensus on gender neutrality has congealed in the absence of systematic research examining how the term actually has been defined and applied in the law. This study—and the grounded knowledge it provides—is crucial to assessing the past uses and limits of gender neutrality, as well as to identifying potential opportunities to expand its application in the future. We begin by tracing feminist debates surrounding the principle of gender neutrality in the 1970s and 1980s, explaining how a concept once championed by feminist legal critics fell into disrepute. Next, we present an analysis of an original sample of hundreds of federal courts of appeals and Supreme Court cases in which the term gender neutral (or synonyms) appears. Our quantitative analyses reveal that courts have predominantly defined gender neutrality narrowly as facial neutrality and generally limited its application to traditional sex discrimination claims. We then contextualize our findings drawing on our qualitative analysis and the secondary literature. We explore how the development of disparate impact doctrine in the courts during the 1970s and judicial refusals to apply gender neutrality in sexual orientation and gender identity discrimination cases have perpetuated its narrow scope. We conclude with some reflections on how the principle of gender neutrality could be utilized in the future to address discrimination on the basis of sex, sexual orientation, and gender identity.

1. ASPIRATIONS AND ASPERSIONS: GENDER NEUTRALITY IN THE COURTS

Feminist legal advocates in the 1970s argued that laws enforcing traditional gender roles violate the Fourteenth Amendment guarantee of equal protection (Franklin 2009). Driven by the conviction that “laws that steer men out of traditionally female roles effectively require women to assume those roles” (Franklin 2009: 26), then-civil rights lawyer Ruth Bader Ginsburg, as head of the American Civil Liberties Union's (ACLU's) Women's Rights Project, led a series of legal challenges on behalf of male plaintiffs who had been denied benefits made available only to women. In one instance, a man was denied “mother's benefits” following the death of his high-earning wife (Weinberger v. Wiesenfeld 1975). During this same period, feminist family law reformers sought to address inequalities within marriage by demanding the removal of sex-specific regulations (such as lower minimum age requirements for women than men) and the replacement of the sex-specific terms husband and wife with the gender-neutral term spouse.

In “Gender and the Constitution,” Ginsburg (1975: 2) explored the potential for constitutional jurisprudence to act as a “stimulus … toward a society in which members of both sexes feel free to develop their full potential as human individuals.” Taking inspiration from the nineteenth century writings of Harriet Taylor Mill and John Stuart Mill, Ginsburg argued that “the legal subordination of one sex to the other…ought to be replaced by a principle of perfect equality, admitting no power or privilege on the one side, nor disability on the other” (1975: 2). With this goal in mind, Ginsburg proposed that courts should insure that gender is irrelevant in “determining the legal rights of men or women” (1975: 23). Ginsburg elaborated: “As in the case of discrimination against racial and ethnic minorities, the ultimate goal with respect to sex-based discrimination should be a system of genuine neutrality” (1975: 28–29, emphasis added).

Ginsburg's use of the term “genuine” gestures at a more robust vision of gender neutrality than one that merely prohibits explicit sex-based legal classifications. Ginsburg's account of gender neutrality is an example of what we call thicker gender neutrality, drawing on philosopher Williams's distinction between “thin” and “thick” ethical concepts (2012). Gender neutrality as facial neutrality is a thin conception of neutrality that does not move beyond a formal level of evaluation.² Thicker conceptions of gender neutrality take additional factors into account, such as intent or impact, or rely on a substantive view of justice in assessing claims. Consider, for example, an employment policy that limits eligibility for certain jobs to veterans. Such a policy is facially neutral in that it does not explicitly disqualify or disadvantage women candidates. Yet, given the historical underrepresentation of women in the armed forces, a thicker conception of gender neutrality might lead to the rejection of such a policy. Scholars have not yet systematically examined the extent to which judges have defined gender neutrality as facial neutrality or, alternatively, adopted a thicker conception.

In 1976, the term gender neutral made its first appearance in a Supreme Court decision, in a dissent filed in the closely watched case General Electric Company v. Gilbert (1976).³ Disputing the “supposed gender-neutrality” of employee sickness and accident benefits plans that excluded coverage for pregnancy-related conditions, Justice Brennan declared the Majority's conclusion that the law did not discriminate to rest on “simplistic and misleading” reasoning (429). Later that same year, the term gender neutral once again appeared in a Supreme Court decision, this time in a Majority opinion striking down an Oklahoma ordinance that set a minimum age of 18 for women and 21 for men to purchase 3.2 percent beer, on the grounds that laws should operate in a “gender-neutral fashion” (Craig v. Boren 1976).

Understandably, such early cases did little to convince skeptics that the principle of gender neutrality would advance constitutional sex equality jurisprudence beyond its existing limits. On the contrary, the Gilbert (1976) decision stood as a troubling indication that the courts might selectively invoke the principle of gender neutrality to justify ignoring sex-based disadvantages; the Craig (1976) decision raised the possibility that the principle would be used primarily to grant men equal access to the few remaining benefits reserved exclusively for women.

By the 1980s, legal feminists emerged as the most vociferous critics of the principle of gender neutrality (Fineman 1983; MacKinnon 1987; Weitzman 1985). In the essay “Difference and Dominance: On Sex Discrimination,” MacKinnon assailed the masculine bias inherent in the “sameness” approach to sex equality jurisprudence—an approach, MacKinnon argued, that is premised on the idea that a woman deserves the same rights and opportunities as a man, but only insofar as she proves herself capable of acting like a man (1987). MacKinnon (1987: 33) argued that this version of the sameness approach “is termed gender neutrality doctrinally” and is “considered formal equality” (see also Fineman 1983).

In assessing the principle of gender neutrality, MacKinnon focused on its application to laws treating men and women as distinct classes. In MacKinnon's view, “as applied, the sameness standard has mostly gotten men the benefit of those few things women have historically had—for all the good they did us” (1987: 35). As evidence, MacKinnon noted that “almost every sex discrimination case that has been won at the Supreme Court level has been brought by a man” (1987: 35). Of course, it is possible that sex discrimination cases involving men plaintiffs are more likely to make it to the Supreme Court in the first place—instead of being settled at the lower courts—because they are more controversial than cases involving women plaintiffs. Nonetheless, MacKinnon dismissed those women who had benefited from application of the principle of gender neutrality as atypical in that they “have been able to construct a biography that somewhat approximates the male norm, at least on paper” (1987: 37). MacKinnon thereby granted that the sameness approach had enabled some women to gain “some access” to professional and educational opportunities but trivialized the cases women had won as instances involving laws based on the most blatant stereotypes, such as the notion that men are inherently more capable than women in financial matters. Referring to the Reed (1971) decision, MacKinnon (1987: 37) quipped: “Imaginary sex differences I will concede the doctrine can handle.”

While feminists have been the most prominent critics of the doctrine of gender neutrality, the term gender neutral has earned the unusual distinction of eliciting equally vehement denunciations from anti-feminists. Consider, for example, the debate over passage of the Equal Rights Amendment (ERA). Since the 1970s, opponents have tried to undermine support for ratification by warning that the ERA would lead to a “totally gender-neutral society” in which recognition of same-sex marriage and “unisex” bathrooms will be compelled by law (Schlafly 1994). These outcomes have been realized even without the passage of the ERA—confirming opponents' predictions about the far-reaching implications of legal incorporation of gender neutrality as a guiding ideal, and indicating that gender neutrality has exerted significant influence within the law even without an explicit legal mandate to do so (Mansbridge 1986).

Even those commentators who seem to endorse the underlying idea of gender neutrality generally refrain from using the term itself (Case 1999; Williams 1989). This may be attributable in part to contemporaneous analyses of race neutrality or “colorblindness” as masking—rather than counteracting—injustice. As legal scholar Kimberlé Crenshaw has argued: “Racial hierarchy cannot be cured by the move to facial race-neutrality in the laws that structure the economic, political, and social lives of Black people” (Crenshaw 1988; see also Harris 1993; Mayeri 2011; Williams 1992). Of course, to insist on facial race (or gender) neutrality is not to imply that this is sufficient to insure justice, only necessary.

To be sure, some commentators have viewed gender neutrality more favorably, if not entirely enthusiastically. In the mid-1980s, Wendy Williams acknowledged the significant harms wrought by “covert as well as overt gender sorting laws,” but concluded that the solution lay in addressing the “disparate effect” of facially neutral laws so as “to squeeze the male tilt out of a purportedly neutral legal structure and thus substitute genuine for merely formal gender neutrality” (1984: 331). Others contend that the case against gender neutrality relies on a caricature of the feminist equality ideal as a demand merely for “sameness” of treatment. For example, Joan Williams has argued that MacKinnon's critique of gender neutrality rests on a “misconception” about the “traditional feminist ideal,” which, properly understood, aims not merely for “gender blindness,” but rather, to restrict the state from enforcing any particular vision of gender roles (1989: 836). Feminist efforts to vindicate the ideal of gender neutrality have never gained critical traction, however, proving a poor match to the rhetorically powerful—if empirically unsubstantiated—case against gender neutrality.

More recent scholarship invites broader reconsideration of the conventional wisdom that gender neutrality is inherently ineffectual. For example, Cary Franklin's compelling reconstruction of Justice Ruth Bader Ginsburg's equal protection jurisprudence elucidates the surprisingly radical implications of Ginsburg's application of the anti-stereotyping principle, and thereby suggests that anticlassification doctrine—commonly regarded as a less exacting approach to equality analysis than antisubordination approaches—in fact may have had more radical implications than has been previously recognized. In an overview of the development of sex discrimination law since the passage of Title VII, Vicki Schultz praises the “gender-neutral, expansive approach” adopted in the Family and Medical Leave Act (FMLA) of 1993, arguing that this law advances the.

feminist…dream of a world in which pregnant women and all women would assume their rightful place alongside men and all other employees—the sick and able-bodied, the parents and the childfree, the caretakers and the carefree, women and men of all races, ethnicities, religions, and walks of life—and together they would create workplaces that met fundamentally human needs to address life's triumphs and travails” (Schultz 2015: 1117, emphasis in the original).

In this way, Schultz offers an alternative to the typical conception of gender neutrality as a principle that promotes superficial over substantive review of laws by reframing it as an ideal that promotes inclusivity over selectivity.

Building on recent work that questions canonical views of the implications and limitations of modern sex equality jurisprudence, this study offers the first in-depth, empirically-grounded consideration of the principle of gender neutrality in the law. Our analysis reveals that the legal impact of gender neutrality has been widely underestimated among legal scholars and that the principle remains significantly underutilized in the law.

2. ACCOUNTING FOR GENDER NEUTRALITY

Despite widespread criticism of the principle of gender neutrality, the courts have incorporated the idea into sex equality jurisprudence since the latter part of the 1970s. We ask: how have judges (in majoriy, concurring, and dissenting opinions) used the term gender neutral? Specifically, we consider whether judges have used gender neutral merely to mean facially gender neutral, or have instead adopted “thicker” understandings of gender neutrality, as defined above.

We further consider the legal context of cases employing the term gender neutral, focusing on the type of discrimination at issue: discrimination against women or men as a class, sexual orientation discrimination, or gender identity discrimination. As an analysis based specifically on the term gender neutral, it is beyond the scope of this study to account more broadly for the development of disparate impact analysis or other legal approaches to sex discrimination. Instead, focusing on the term gender neutral enables us to trace the development of one approach to thinking about sex discrimination, while considering the distinctive strengths and limitations of the principle of gender neutrality for addressing gender inequality.

2.1 Sampling

This study draws on analyses of all Supreme Court and federal courts of appeals cases—including the majority, concurring, and dissenting opinions—and amicus briefs filed in Supreme Court cases in which the terms gender neutral/neutrality, sex neutral/neutrality, or sexually neutral/sexual neutrality appear from the first mention of any of these terms (in 1974) to December 31, 2016, when we conducted the most recent search, using WestlawNext. (We henceforth refer to “the term gender neutral” as shorthand for all permutations.) By identifying all cases that include these search terms, we are able to provide a comprehensive assessment of the use of these specific terms within these courts. By examining how judges use the term gender neutral, this project follows the tradition of cultural sociological research projects that interrogate how societies—including institutions such as courts, legislatures, and the news media—socially construct specific concepts and associated terms (DiBennardo 2018; Saguy 2003). As with the terms sexual harassment and sexually violent predator, for example, there is disagreement about what gender neutrality does—and should—mean. Outside of the courts, the growing and varied use of the term gender neutral suggests that one could potentially use this term to advance a host of feminist and lesbian, gay, bisexual, transgender, and queer (LGBTQ)-rights goals (Saguy and Williams 2019). This leads us to consider the extent to which judges have put the term to those ends and what this might suggest for future application of the concept, inside the courts and beyond.

Our approach systematically examines usages of the term gender neutral in the Supreme Court and federal courts of appeals. This focus cannot tell us how the term gender neutral is used in case law generally. Given that appellate courts set precedent for lower courts, however, the way they have defined gender neutrality is expected to shape lower court interpretations of the term. This follows the doctrine of “stare decisis,” which compels lower courts to defer to the outcome and reasoning established by higher courts. Powerful dissents articulate important alternative perspectives and can influence future rulings and inspire new legislation (Ginsburg 2010). Our approach also cannot address to the question of whether other terms (beyond the synonyms we specify) have been used in its place as functional equivalents. Two possible synonyms we intentionally did not use as search terms are neutral (without gender as a qualifier) and facially neutral. Searching for all cases containing neutral produces a large number of cases that do not concern sex or gender but rather address neutrality in regard to other issues, from “race neutrality” to “net neutrality.” Given that one of our empirical questions concerns the extent to which judges use gender neutral to mean facially neutral or alternatively adopt a thicker conception of gender neutrality, we did not include “facial neutrality” as a search term. We intended to avoid the initial assumption that facial neutrality is used as a synonym for gender neutrality so that we could investigate whether, and to what extent, this is the case.

Our data include 27 Supreme Court and 488 appellate court cases, including majority, concurring, and dissenting opinions. Ten of our appellate court cases were heard twice. We treat each pair of cases as a single case. Three of the cases in our Supreme Court sample are also included as part of our appellate court sample, including Kirchberg v. Feenstra (1981), Arizona Governing Committee for Tax Deferred Annuity and Deferred Compensation Plans v. Norris (1983), and Price Waterhouse v. Hopkins (1989). Given that previous research has shown that justices often incorporate language from certain amicus briefs that they believe will enhance their ability to make effective law and policy (Collins et al. 2015) and that amicus briefs may significantly affect the likelihood of dissenting opinions (Collins 2008), we also qualitatively analyze 143 amicus briefs collectively filed in 67 Supreme Court cases. Forty of these Supreme Court cases are not in our Supreme Court sample since they did not use any of our search terms. Of these, nine are in our courts of appeals sample.

2.2 Coding

With a team of coders—made up of one law student and several undergraduate students in addition to the authors—we recorded the court level, case citation, date (of the most recent case if it was heard twice at the appellate level), year, and date of earlier hearing if applicable, for each case. We created dichotomous variables to code for whether the case included two appellate cases that were merged, whether it was a Supreme Court case, or whether it was an appellate court case. For accountability purposes, we included a text box on the coding form where coders recorded their initials.

We coded the court cases—including majority, concurring and dissenting opinions—for several substantive dichotomous variables, coding 1 when the element was present anywhere in the case and 0 when it was not. Below we describe the variables analyzed for this paper. To determine how the term gender neutral was defined, we coded each case for whether it used the term gender neutral to mean facial gender neutrality and whether it used the term to mean thicker gender neutrality. The Supreme Court case Orr v. Orr provides an example of gender neutrality used to mean facial neutrality: “Whereas here, the State's compensatory and ameliorative purposes are as well served by a gender-neutral classification as one that gender classifies and therefore carries with it the baggage of sexual stereotypes, the State cannot be permitted to classify on the basis of sex” (1979: 283). In this context, gender neutral means the opposite of sex-based.

In contrast, the dissent in Personnel Administration of Massachusetts v. Feeney insists that a “facially neutral” policy “that so severely restricts public employment opportunities for women cannot reasonably be thought gender-neutral” (1979: 99). We coded this usage as one for thicker gender neutrality. Note that in our coding schema, the facial gender neutrality and thicker gender neutrality codes are not mutually exclusive. A given case could use the term gender neutrality at one point to mean facial neutrality and at another to mean genuine neutrality (or neither). It is also important to emphasize that we coded for whether a given case invoked a particular conception of gender neutrality, not whether the decision hinged on this particular conception.

Five variables concerned the type of discrimination at issue in a case, including whether it discussed: (1) discrimination against women, (2) discrimination against men, (3) discrimination on the basis of sexual orientation, (4) discrimination against transgender people, or (5) no discrimination mentioned. We combined the first and second variables to create a variable that indicates sex discrimination as traditionally understood, that is, as discrimination against women or men as a class. We double coded about 10 percent of the cases. The Krippendorf Alpha (Krippendorff 2004), a measure of intercoder reliability that controls for the likelihood of agreeing by chance, was 91 percent for our variables as a whole. To enable further qualitative analysis of the data, we included a text box that prompted the coder to describe the legal question at stake and another prompting the coder to comment on the case. A third text box instructed the coder to copy and paste all usages of the search terms in the case.

In addition to our analysis of the cases, a law student research assistant recorded the citation for the Supreme Court case in which an amicus brief was filed, the number of amicus briefs filed for each Supreme Court case, whether the search term was used in the Supreme Court opinion itself (i.e., whether the Supreme Court case was also in our Supreme Court sample), whether the search term was used in an earlier court of appeals opinion (i.e., whether the court of appeals case was in our courts of appeals sample), a brief explanation—when relevant—for why the Supreme Court case did not use any of our search terms, and the party filing the brief. We then qualitatively analyzed the 143 amicus briefs focusing on how the parties writing these briefs used the term gender neutral. We identified instances in which an amicus brief employed the term gender neutral differently than the Supreme Court majority, concurring, or dissenting opinions, including instances in which an amicus brief—but not the final decision—defined gender neutral as thicker gender neutrality. Our analysis of amicus briefs provides a glimpse of additional possible usages of the term and contexts in which it could be invoked that were not recognized in higher court decisions.

3. LIMITING GENDER NEUTRALITY

As we report below, the United States Supreme Court and federal courts of appeals generally have adopted a narrow definition of gender neutrality, predominantly using it to mean facial neutrality as opposed to following a thicker gender-neutrality approach. At the same time, courts overwhelmingly have limited the application of the principle of gender neutrality to traditional sex discrimination claims. Courts have not recognized the relevance of the principle of gender neutrality in adjudicating sexual orientation or gender identity discrimination claims, despite advocacy (as reflected in our analyses of the amicus briefs) urging them to do so.

3.1 Predominantly Facial Neutrality

In our sample, the term gender neutral is used overwhelmingly to mean facially neutral. Of the 27 Supreme Court cases in our sample, all include instances of the term gender neutral used to mean facially neutral, whereas only two cases—General Electric Company v. Gilbert (1976) and Personnel Administration of Massachusetts v. Feeney (1979)—also include usages of gender neutral to indicate a thicker conception of the principle. In Gilbert, the dissenting opinion insists that if one takes into consideration the “historical backdrop of General Electric's employment practices,” the majority's “assumption that General Electric engaged in a gender-neutral risk-assignment process” is “purely fanciful” (1979: 148). Likewise, in Feeney, the dissenting opinion argues that “although neutral in form, the statute [giving veterans priority in hiring for all state employment] is anything but neutral in application. It inescapably reserves a major sector of public employment to ‘an already established class which, as a matter of historical fact, is 98% male’” (1979: 284).

Only eight out of a total of 488 Appellate Court cases—less than 2 percent—in our sample include an instance of a thicker conception of gender neutrality anywhere in the decision. In Concrete Works of Colorado, Inc. v. City and County of Denver (2003), for example, the court rejected the idea that a prequalification related to the size and experience of businesses competing for Denver transportation projects was truly gender neutral, even though it conceded that it was “neutral on its face.” The court ruled that “experience and size are not race-and gender-neutral variables” since “M/WBE [Minority and Women-Owned Business Enterprise] construction firms are generally smaller and less experienced because of industry discrimination” (2003: 981). This case is unusual in our sample; rather than merely acknowledging that facial neutrality does not guarantee a lack of discrimination, the court takes the extra step of (re)defining gender neutrality as requiring a more searching standard.

While the thicker conception of gender neutrality appears in only a small number of cases in our sample, it has been advanced by various amici, indicating a recognition—among at least some advocates—of the potential impact of adopting a more substantive conception of gender neutrality in the law. In Personnel Administration of Massachusetts v. Feeney (1979), discussed above, the Supreme Court considered whether a hiring practice that gave preference to veterans was discriminatory because it disproportionately disadvantaged women applicants, who were less likely to have served in the military. In its opinion, the Majority used the term gender neutrality to mean facial neutrality, holding that a statute with a disparate impact on women is valid absent evidence of a discriminatory purpose. But an amicus brief submitted by several national organizations, including the National Organization for Women and the NOW Legal Defense and Education Fund, articulated a different view of gender neutrality—one that, as we saw above, was echoed in the dissenting opinion:

At the purely semantic level, the Massachusetts statute is, perhaps, not gender-based; that is, the preference is not expressly granted to “men” only. But neither is the statute gender-neutral in defining the preferred group in terms of criteria that men and women are equally capable of satisfying under law [citation omitted]. The legal impediments to women qualifying as “veterans” must be read into the term itself. If the statute is neutral in form, it is not in fact. By operation of law, it is gender-based. (Personnel Administration of Massachusetts v. Feeney 1979: 5–6, emphasis added)

In this statement, amici do not urge the court to adopt additional criteria beyond gender neutrality to determine whether a policy discriminates. Rather, they contend that the neutrality principle itself requires more than consideration of a law's form; its impact must also be taken into account. We rarely encountered this more robust understanding of the neutrality concept in our sample. Nonetheless, as we discuss later, this understanding has the potential to make a significant impact in the law.

3.2 Majority of Claims Concern Traditional Sex Discrimination

In addition to generally defining gender neutrality as facial neutrality, the court cases in our sample overwhelmingly address instances of traditional sex discrimination—that is, claims of discrimination against women or men. Strikingly, court opinions —whether majority, concurring, or dissenting—rarely mention gender neutrality when considering sexual orientation or gender identity discrimination claims. All of the cases in our Supreme Court sample concern sex discrimination as traditionally understood, that is, as discrimination against women or men as a class (see Figure 1). In contrast, only one case in our sample discusses discrimination on the basis of sexual orientation and none discuss gender identity discrimination. Among courts of appeals cases, 385 out of 488 cases (79 percent) concern traditional sex discrimination (see Figure 2). In contrast, only three cases mention discrimination on the basis of sexual orientation and two mention discrimination on the basis of gender identity (less than 1 percent).

Figure 1. Type of Discrimination at Stake, Supreme Court Sample (N = 27). [Color figure can be viewed at wileyonlinelibrary.com]

Figure 2. Type of Discrimination at Stake, Appellate Court Sample (N = 488). [Color figure can be viewed at wileyonlinelibrary.com]

4. UNDERSTANDING GENDER NEUTRALITY IN THE LAW

Our qualitative analyses of the cases in our sample—read alongside the secondary literature—help contextualize why courts have interpreted and applied the principle of gender neutrality narrowly. At the same time, this analysis highlights debates about prevailing judicial understandings of the meaning and scope of the principle. We now turn to these issues.

4.1 Delimiting Gender Neutrality

Our finding that the US Supreme Court and federal courts of appeals have used the term gender neutral predominantly to mean facially neutral reflects the broader influence of the anticlassification principle in sex discrimination jurisprudence since the 1970s (Balkin and Siegel 2003). At the start of that decade, sex-based classifications were commonplace. As Chamallas has observed, “gender determined not only who received alimony (only women) or who was eligible for the draft (only men) but also virtually every facet of life subject to legal regulation (2013: 35; see also Baumgardner and Richards 2000). This rapidly changed, however, once the Court began to question laws that explicitly classified on the basis of sex. By applying heightened scrutiny to sex-based classifications, many common practices suddenly faced a constitutional standard they could not meet. In 1976, the Court declared that states would be required to “realign their substantive laws in a gender-neutral fashion” (Craig v. Boren 1976: 199). The next year, Justice Stevens emphasized in a concurring opinion that “where … the State's compensatory and ameliorative purposes are as well served by a gender-neutral classification as one that gender classifies and therefore carries with it the baggage of sexual stereotypes, the State cannot be permitted to classify on the basis of sex” (Califano v. Goldfarb 1977: 283).

Still, as we have seen, the dominant approach is to treat gender neutrality as facial neutrality. Indicative of this, as a synonym for gender neutrality, the term facial neutrality appears frequently in our sample. In contrast, the term genuine neutrality—used by Ginsburg in a law review article in the 1970s, and which indicates a thicker conception of gender neutrality—appears nowhere in our sample.⁴ This is similar to how the courts have curtailed efforts to combat explicit (or de jure) racial segregation in schools, while tolerating de facto segregation (Milliken v. Bradley 418 US 717 1974). Judicial reluctance to adopt a more robust conception of neutrality may speak generally to the role that judges see themselves as playing—as responsible for bringing about social change or upholding laws narrowly defined (Gibson 1978; Segal and Cover 1989; George and Epstein 1992). The narrow interpretation of gender neutrality may also reflect the demographic composition of the courts, which have been dominated by white, heterosexual men, who—some but not all studies suggest—are less likely than women or African American judges to support sex discrimination plaintiffs (Beiner 2011). Perhaps partly in response to this, the nomination of people who would bring greater demographic diversity to the Supreme Court—including Sonia Sotomayor and Elena Kagan—have been met with considerable resistance (Beiner 2011).

The absence of the term genuine neutrality in our sample may also be an indication that judges have sought to address the limitations of a facial neutrality standard primarily by undertaking disparate impact analysis, rather than by elaborating thicker conceptualizations of gender neutrality, such as the genuine neutrality ideal. Disparate impact claims provide a way to challenge facially neutral laws that disproportionately exclude members of a protected class. The Supreme Court first recognized a disparate impact cause of action in Griggs v. Duke Power Co. (1971), a Title VII case in which the Court held that an employer cannot use selection criteria with a disparate racial impact unless the criteria are necessary to perform the job. Incorporating disparate impact analysis into sex discrimination law was thought to have several advantages, including smoking out intentional discrimination masked by facially neutral policies, uncovering unconscious bias, and revealing structural discrimination—such as when an employer acting without bias adopts a standard selecting for traits whose allocation is shaped by past discrimination (Siegel 2015: 657).

At the time Griggs (1971) was decided, no clear distinction was made between statutory and constitutional equality standards, and the courts treated disparate impact alone as evidence of an unconstitutional purpose (Siegel 2015: 661). But in Washington v. Davis (1976) and then in Personnel Administration of Massachusetts v. Feeney (1979), the Supreme Court declined to extend this statutory framework to constitutional violations. Instead, in the equal protection context, the Court henceforth required a showing of discriminatory purpose to advance a discrimination claim. But because discriminatory purpose has proven exceedingly difficult to establish, the result is that any facially neutral law is virtually guaranteed to pass constitutional muster (Case 2010: 1474; Mayeri 2008: 1854).

In arguing for the need to extend disparate impact analysis to constitutional sex discrimination claims, commentators and advocates alike have bypassed the question of whether the principle of gender neutrality itself might be defined more robustly and applied more vigorously. In our sample, 17 percent of the cases mention the term disparate impact—considerably more than the proportion employing a thicker conception of gender neutrality. This may indicate that an insistence on the need for disparate impact analysis in antidiscrimination cases has displaced arguments for thicker conceptions of gender neutrality. Assessing the extent to which courts are using disparate impact analysis in place of adopting a thicker conception of gender neutrality would require detailed analysis of all discrimination cases, which is well beyond the scope of this project. Future work should further examine the extent to which disparate impact analysis serves the same purpose in legal decisions as would a thicker conception of gender neutrality.

4.2 Debating Facial Neutrality

While we found few examples of courts using the term gender neutral to indicate something more than facial neutrality, we did find acknowledgment in decisions that meeting the standard of facial neutrality is not sufficient to guarantee equal treatment. In Miller v. Albright the court placed the term gender neutral in scare quotes to underscore its recognition that a facially neutral policy would nonetheless “disfavor” unmarried men “in practical operation” (1998: 436). In other cases, courts have indicated awareness that facial neutrality can too readily be used to mask an invidious intent. For example, in E.E.O.C. v. Farmer Bros. Co., the Court found it “troubling” that an employer “assumes the mere appearance of gender neutrality negates the district court's finding that [the employer] engaged in gender discrimination” (1994: 900). In highlighting the superficiality of facial neutrality as a standard, some courts have echoed concerns raised earlier by feminist commentators.

As we have seen, judges generally have defined gender neutrality narrowly to mean facial neutrality. At the same time, the concept of facial neutrality itself has been defined narrowly. Those laws and policies that use the terms male and female, man and woman, and husband and wife have been readily declared not gender neutral. But others have been deemed facially neutral, even when their impact is evidently limited to members of one sex—as is the case with pregnancy-related policies. In a 2012 dissent, Justice Ginsburg “revisited” the nearly forty-year-old Geduldig (1974) decision in which the Court declared that “discrimination on the basis of pregnancy is not discrimination on the basis of sex” (Coleman v. Court of Appeals of Maryland 2012: 51). Citing an earlier dissent in a 1993 case, in which Justice Stevens declared it “simply false” that “a classification based on pregnancy is gender neutral,” Justice Ginsburg maintained in Coleman that “pregnancy discrimination is inevitably sex discrimination” (2012: 56). Justice Ginsburg's continued insistence on this point indicates that the question of what it means to be facially neutral has yet to be definitively resolved (Balkin and Siegel 2003).

4.2.1 Limited Application Beyond Traditional Sex Discrimination Cases

Gender neutrality has been widely affirmed as a guiding principle in traditional sex discrimination cases (Appleton 2005: 18). We find, however, that courts generally have not invoked the principle in cases involving sexual orientation or gender identity discrimination claims (Currah et al. 2006). This finding supports the more general observation that attention to gender has been largely “missing” in sexual orientation and gender identity discrimination jurisprudence (see Appleton 2005; Case 2016).

The only Supreme Court case in our sample in which gay rights was discussed is Bowers v. Hardwick (1986), which considered the constitutionality of a Georgia statute criminalizing all acts of sodomy—whether between same or different sex partners. In upholding the law, the Majority seemed to regard it as self-evident that a law is facially neutral if it does not explicitly reference sex or sexual orientation. Others disagreed. As Halley (1993: 1741) observed, “all the dissenters and virtually every academic commentator on the case have noted … [that] Michael Hardwick challenged a gender neutral sodomy statute on its face.” In other words, Hardwick and several commentators rejected the idea that a statute banning sodomy can be deemed facially gender neutral simply because it does not explicitly mention homosexuals as a class. As with pregnancy discrimination cases, the Bowers decision and its aftermath point to ongoing contestation over the meaning of facial neutrality itself.

The Bowers decision was overturned by Lawrence v. Texas (2003a), a case that is not in our sample because the decision does not use the term gender neutral. In contrast to Bowers, the Lawrence case concerned a statute that applied specifically to same-sex partners. While the Supreme Court chose to sidestep the question of whether sodomy bans are gender neutral in its decision, four amici submitted in the case did not. The NOW Legal Defense and Education Fund advocated that laws prohibiting same-sex sodomy cannot be gender neutral because the sex of the parties involved is key to determining the criminality of the act (Lawrence v. Texas 2003b). In contrast, other organizations submitted briefs insisting that the statutory prohibition against sodomy is gender neutral. The Concerned Women for America argued: “The Texas statute at issue here, of course, is entirely gender-neutral, applying equally to same-sex ‘deviant sexual intercourse’ by men and by women and without regard to sexual orientation” (Lawrence v. Texas 2003c: 21). These amici demonstrate that legal advocates on both sides recognized the relevance of the principle of gender neutrality to a sexual orientation discrimination claim, even if the courts generally have not.

In marriage equality cases as well, appellate courts have assiduously avoided the issue of whether same-sex marriage bans violate the principle of gender neutrality. In 1993, the Supreme Court of Hawaii declared the state's same-sex marriage ban to be facially discriminatory. “On its face and as applied … [the same-sex marriage ban] regulates access to the marital status and its concomitant rights and benefits on the basis of the applicants' sex. As such, [the ban] establishes a sex-based classification” (Baehr v. Lewin 1993: 64, emphasis added).

In contrast, the Supreme Court has chosen to bypass sex discrimination claims in its marriage equality decisions, even as amici urged the Court to confront the sex classifications created by same-sex marriage bans. Notably, two amici in United States v. Windsor (2013) and ten in Obergefell v. Hodges (2015a) specifically reference gender neutrality. The National Women's Law Center and law professors associated with the Williams Institute at UCLA School of Law issued an amici curiae brief in Windsor arguing that the Defense of Marriage Act (“DOMA”) discriminates on the basis of sexual orientation because it is based on overbroad gender stereotypes about the preferences and capabilities of men and women:

Laws relating to marriage have been almost wholly gender-neutral apart from their frequent exclusion of same-sex couples (citing Appleton 2005). Men and women entering marriage today have the liberty to decide for themselves the responsibilities each will shoulder as parents or wage earners or family decision-makers regardless of whether these responsibilities conform to or depart from traditional arrangements. (United States v. Windsor 2013: 18)

In other words, the argument against same-sex marriage has become less tenable as the sexist foundation for different-sex marriages has eroded.

Several amici in the Obergefell (2015a) case advanced similar arguments. An amicus brief submitted by 74 family law scholars noted that arguments against same-sex marriage are based on outdated gender stereotypes about marriage and parenting: “Today, both parents are equally responsible for the care and support of their children, and, upon separation or divorce, the standards for child custody determinations are gender-neutral” (Obergefell v. Hodges 2015b: 22). A group of legal scholars submitted a brief arguing that laws restricting the right to marry on the basis of gender fail intermediate scrutiny in part because, while states can deny adoption rights to individual couples deemed unsuitable for raising children through “gender-neutral” mechanisms, they cannot categorically declare same-sex couples inherently unfit for parenting (Obergefell v. Hodges 2015c: 29). These amici demonstrate that numerous advocates seeking recognition of same-sex marriage have invoked the principle of gender neutrality. While these arguments failed to win recognition in Supreme Court rulings on marriage equality (Appleton 2005; Case 2010; 2016), their appearance in amicus briefs from elite organizations indicates the plausibility of these claims (Collins et al. 2015).

Despite the fact that the Supreme Court has steadfastly ignored the principle of gender neutrality in sexual orientation and gender identity discrimination cases, the steady “gender neutralization” of family law in particular has had significant implications for same-sex couples, and for transgender and gender nonconforming parents (Appleton 2005: 19). As noted above, the US Supreme Court decision in Obergefell v. Hodges nowhere mentions the term gender neutral. Still, just hours after the ruling was announced, the Supreme Court of Ohio issued an order stating that “all references to husband, wife, father, mother, parent, spouse, and other terms that express familial relationship” in the state legal code henceforth would be “construed as gender neutral” (In re Admin. Actions, dated June 26, 2015). In 2017, the Uniform Parentage Act (UPA) was given a gender-neutral overhaul to remove gendered presumptions surrounding definitions of marriage and family; the UPA has been enacted in three states with two more considering it in 2019 (Uniform Law Commission 2018). In these ways, the principle of gender neutrality has influenced LGBTQ rights law significantly, if indirectly—and it can be expected to continue to do so in the future (Nejaime 2016: 1211).

5. THE FUTURE OF GENDER NEUTRALITY

Legal advocates have pressed the courts to adopt a thicker conceptualization of gender neutrality and to extend its scope of application beyond traditional sex discrimination cases. Doing so would produce meaningful changes in laws and policies related to issues ranging from pregnancy discrimination to LGBTQ family law. At the same time, there are indications of other efforts to rethink gender neutrality that have even more radical implications. We consider two such efforts here.

The cases in our sample reveal that when a sex-based classification is challenged, there may be more than one possible gender-neutral alternative. Consider Coleman v. Court of Appeals of Maryland (2012). One of the questions the Court discussed in this case was whether the self-care leaves provision of the 1993 FMLA originally was intended to address gender discrimination in the workplace. The majority held that, because FMLA's self-care provision is gender neutral on its face and has been used by both men and women, it could not plausibly be construed as an antisex discrimination measure. In a strongly worded dissent, Justice Ginsburg explained that the FMLA was crafted in gender-neutral terms not because legislators were indifferent to the problem of sex discrimination, but precisely because their central goal was to address unequal treatment. The authors of the FMLA recognized that even gender-neutral parental leave policies can disadvantage women, Justice Ginsburg explained, if employers prefer to hire men assuming they will be less likely than women to use parental leave provisions. To account for this risk, the FMLA provides for “self-care” leave—a broad category of leave that includes, but is not limited to, time off from work for conditions related to childbirth and child-rearing. In this way, the FMLA self-care leaves provision positions pregnancy and parenting on par with most other conditions that lead employees (including men) to request leave.

Elaborating the logic underlying the FMLA leave policy, Justice Ginsburg cited legal scholar Wendy Williams, who, in a well-known 1984 law review article (Williams 1984), argued against the creation of special legal protections for pregnancy-related conditions. In Williams's view, the goal of sex equality was better served by expanding worker-protection policies for all employees, and including pregnancy among other covered conditions. As Williams explained, gender-neutral policies can serve to “overcome the definition of the prototypical worker as male and to promote an integrated—and androgynous—prototype” (1984: 363). In citing Williams, Justice Ginsburg's Coleman dissent suggests that judicial inquiry is obliged to move beyond the simple question of whether or not a law is gender neutral to consider how exactly gender neutrality is achieved in a particular situation.

Increasingly, it is becoming clear that how gender neutrality is defined matters. Underscoring this point, researchers recently analyzed the effects of gender-neutral “tenure clock stop” policies adopted by faculty in economics departments. These policies were expected to increase tenure rates for women. Instead, researchers found that the gender-neutral policies “substantially reduced female tenure rates while substantially increasing male tenure rates,” likely because men who took the leave published more during their leaves than did women, who—on average—assume a disproportionate share of childcare responsibilities (Antecol et al. 2018: 2420). The study differentiates between gender-neutral policies that “extend equal benefits to new mothers and fathers” and “female-only” policies that “are only available to women” (Antecol et al. 2018: 2422). Yet, one could also envision gender-neutral policies that take into account differences in childcare responsibilities, by, for instance, distinguishing between primary and secondary caretakers. Similarly, some have sought to shift attention from discussions of who can use a given public facility as currently configured to redesigning restrooms to improve privacy, safety, and convenience so that the issue of who else is using it would become less salient (Sanders and Stryker 2016). Within the law, too, such considerations warrant closer attention.

Another possible reconceptualization of the principle of gender neutrality is suggested by popular usages of the term gender neutral (Saguy and Williams 2019). While courts generally have not mentioned gender neutrality in decisions concerning gender identity discrimination, prominent LGBTQ rights advocates have adopted the term gender neutral as a way to describe places (such as restrooms) and practices (such as pronoun usage) that recognize that gender identity does not always match assigned sex and that relieves the obligation to declare a fixed gender identity (Meadow 2010; Cruz 2002). Harkening back to Williams's concept of “deinstitutionalizing gender,” these efforts raise important questions about the need for the state to assign sex/gender identities in the first place (1989). The conceptualization of gender neutrality as the deinstitutionalization of gender would be a powerful tool for countering the proliferation of so-called “bathroom bills” that force people to use restrooms that correspond to their legal sex, regardless of gender. Gender neutrality as the deinstitutionalization of gender would also enable efforts to remove gender markers altogether from official identity documents, including birth certificates, social security cards, driver's licenses, and marriage licenses (Davis 2017). Reconceptualized in this way, gender neutrality would be an even more powerful tool for combatting sex discrimination, while providing a further basis for extending application of the principle to sexual orientation discrimination and gender identity discrimination claims.

6. CONCLUSIONS

Empirical analysis of Supreme Court and courts of appeals cases using the term gender neutral demonstrates that, as legal feminists warned, the courts have defined gender neutrality narrowly to mean facial neutrality. At the same time, our analysis reveals that—contrary to the predictions of feminist skeptics—the courts have applied this narrow principle widely in traditional sex discrimination cases, rather than selectively or inconsistently. These findings suggest that the extent to which facial neutrality has promoted gender equality may be underappreciated. We also believe that the limited number of cases in our sample concerning sexual orientation and gender identity discrimination claims means that there is room to further develop the concept.

Antifeminist activists in the 1970s warned that a gender-neutral society would be forced to reject everything from traditional understandings of mothering to prohibitions on same-sex marriage to segregated restrooms. In contrast, feminist legal assessments of gender neutrality in the same period ignored its applicability to sexual orientation and gender identity discrimination claims. Our analysis suggests that this omission was a missed opportunity to challenge sex-based classifications in policies that maintain heteronormative and cisgender privilege. Outside of the courts, gender activists have adopted the term gender neutral to frame demands for more egalitarian and inclusive gender practices (Saguy and Williams 2019). It would be an unfortunate irony if the legacy of earlier determinations of the inherent limitations of the principle of gender neutrality were to lead potential beneficiaries to underestimate its untapped potential in the law, and beyond.