The Yale Law Journal


Hively’s Self-Induced Blindness

13 Jun 2017

abstract. The Seventh Circuit’s en banc decision in Hively v. Ivy Tech Community College has received as much attention for its dueling views of statutory interpretation as for its historic holding: that sexual orientation discrimination is protected under Title VII’s “because of sex” prong. Yet the opinions’ divergent approaches to statutory interpretation end up doing surprisingly little work. Lacking substantive engagement with the ways sexual orientation discrimination helps police gender norms or longstanding debates over how thoroughly Title VII is meant to disrupt those norms, the opinions instead offer an originalism without history, a dynamic interpretation that lacks limits, and a textualism largely divorced from the values Title VII’s text is meant to address. As other courts consider whether to adopt Hively’s important holding, its gender-blind approach to equality law should give way to one that foregrounds the opportunity-limiting sex stereotyping at the heart of anti-LGBT animus.1

Hively v. Ivy Tech Community College2 is an important decision that wants to be an important opinion—four opinions, actually, all competing for space in future casebooks. With its unquestionably historic holding—the first federal court of appeals to declare that sexual orientation discrimination necessarily comprises sex discrimination under Title VII—you might expect the Seventh Circuit’s opinions to be gunning for casebooks on employment discrimination or sexual orientation law. But you would be wrong. These are opinions that want to say something important, first and foremost, about statutory interpretation.3

The irony is that the Hively opinions show how little judges’ views on statutory interpretation matter when they are blind to substance. The opinions cite eight eminent legislation scholars4 without mentioning a single antidiscrimination or gender theorist, legal historian, or gay rights advocate. Perhaps as a result, Hively ends up offering an originalism without history, a dynamic interpretation that lacks a limiting principle, and a textualism largely disengaged from the values Title VII’s text is best understood to promote. What results is a gender-blind approach to equality law closer to that of the conservative anticlassificationists on the Supreme Court than to the Justices who have thus far voted for LGBT rights. We should not forget how few judges were convinced by formalist arguments like Hively’s in the many years of litigation over same-sex marriage.5

As the most important legal success for LGBT rights since the marriage rulings, Hively will hopefully inspire other courts to embrace its result.6 But Hively’s reasoning and influence on other courts might both have been stronger had it acknowledged the decades of scholarship and advocacy that brought us to the point where Hively’s result seemed inevitable.

To make this case, Part I of this Essay describes the formalist debate at the heart of Hively’s majority and dissenting opinions. Part II argues that a more substantive engagement with the sex-specific stereotyping at the heart of sexual orientation discrimination would highlight what is at stake in this debate: gender policing, a (contested) target of Title VII from the time of its passage. Part III turns to Judge Posner’s concurring opinion in Hively and suggests how his common law approach to Title VII’s anti-discrimination provisions could itself learn from those who have actually studied discrimination in the contemporary workplace.

I. the main opinions

Like a number of other recent judicial and administrative opinions and briefs,7 Judge Wood’s opinion for the eight-judge majority in Hively leads with a deceptively simple argument. Title VII prohibits adverse employment actions—like Ivy Tech’s refusal to hire longtime instructor Kimberly Hively for a full-time faculty position—when they are taken against someone “because of such individual’s . . . sex.”8 Hively alleged that she was denied full-time employment because she is a lesbian. Put another way, she alleged that Ivy Tech hires men, but not women, who are attracted to women. But for Hively’s sex, she would have a job.

Joined by two other judges in dissent, Judge Sykes denies the relevance of this comparison. Were Hively a gay man rather than a lesbian, she would still presumably have been denied the job.9 Thus, according to the dissent, no discrimination “because of sex” has occurred. Ivy Tech treats male and female homosexuals equally badly.

How do we decide which comparison is more apt? Here, Judge Sykes appeals to the original public meaning of the statute. It is not “even remotely plausible,” she insists, “that in 1964, when Title VII was adopted, a reasonable person competent in the English language would have understood that a law banning employment discrimination ‘because of sex’ also banned discrimination because of sexual orientation.”10 Reasonable English speakers—in 1964, and now11—understand sex and sexual orientation as different immutable traits that give rise to categorically distinct types of discrimination. (Note that Sykes’ originalism is not necessary to her opinion, since she thinks the plain meaning of Title VII’s text, even today, precludes Hively’s claim.)

Instead of disputing what the dissent calls the traditional notion of sex held by Congress and others in 1964,12 the majority relies on a 1998 opinion by Justice Scalia. In Oncale v. Sundowner Offshore Services, Scalia acknowledged that although male-on-male sexual harassment, the subject of the case, “was assuredly not the principal evil Congress was concerned with when it enacted Title VII, . . . it is ultimately the provisions of our laws rather than the principal concerns of our legislators by which we are governed.”13 Since the dissent’s originalist approach looks to original public meaning, not Congress’s original intent, the passage from Oncale is not exactly responsive. But it does help the majority lay claim to its own brand of textualism: one that decides in the present rather than asking the past how best to understand the words of a statute.

This debate over statutory interpretation has yet to do much work. It has certainly not provided a reason for deciding who Hively’s comparator should be: a man attracted to women (thus maintaining the gender of the worker’s partner) or a gay man (thus holding sexual orientation constant).14

Perhaps recognizing this impasse, the majority opinion—again, like other recent opinions and briefs15—offers two more reasons for finding protection for Hively under Title VII. First, the majority mentions gender stereotyping, as the Supreme Court has long read Title VII to “strike at the entire spectrum of disparate treatment of men and women resulting from sex stereotypes.”16 Hively, according to Judge Wood, “represents the ultimate case of failure to conform to the female stereotype . . . she is not heterosexual.”17

Phrased in this way, the gender stereotyping argument might seem to duplicate the formalist argument already made. If Hively were a man dating a woman, she would conform with gender stereotypes. But since she is a woman dating a woman, she defies them. This sounds like the same argument as before, and it invites the same response: if nonheterosexual men and nonheterosexual women both violate the same gender norm, the norm must not be sex-specific. Thus, it cannot form the basis of a sex discrimination claim. I return to this response in the following Part.

The majority’s final argument offers an “associational theory” of discrimination.18 As Judge Wood writes: “It is now accepted that a person who is discriminated against because of the protected characteristic of one with whom she associates is actually being disadvantaged because of her own traits.”19 Central to this claim is Loving v. Virginia, in which the Supreme Court struck down Virginia’s antimiscegenation laws as unconstitutional race discrimination.20 But care is needed here. Imagine a white male employee married to an African-American woman. He would certainly have a Title VII claim if his boss fired him out of aversion to racial intermingling.21 Call this an associational claim if you like, but it is really just the same comparator claim as before: if the employee had been black rather than white, the boss would not have held his marriage against him.22 The associational approach only adds something when the law protects some groups but not others, and the partner, but not the employee, belongs to the protected group.23

Invoking Loving does not provide a separate route to protection, though it is often presented this way.24 What it does offer, importantly, is a reason for choosing the majority’s comparator over that of the dissent. The dissent, recall, denies that sex discrimination occurred since Ivy Tech acted symmetrically, disadvantaging both women and men who have same-sex partners. This is akin to the argument made by Virginia in Loving, where it insisted that by punishing both white and black people in interracial marriages, it had treated the races equally.25 The Loving Court rejected that approach, holding antimiscegenation laws unconstitutional “even assuming an even-handed state purpose to protect the ‘integrity’ of all races.”26 The analogy in Hively is clear: symmetrical sex discrimination is sex discrimination against both genders, not neither.

There are two reasons why this might be so. First, symmetrical discrimination against lesbians and gay men necessarily takes account of those employees’ gender. This is the approach Judge Flaum takes in his concurrence in Hively.27 At the moment Ivy Tech decided not to hire Hively, it was necessarily aware of, and motivated by, the fact that she is a woman in a relationship with a woman. Ivy Tech is liable because it was insufficiently gender-blind.

The second possibility emerges from the sentence just quoted from Loving.28 It suggests that symmetrical discrimination—punishing both white and black spouses, or both gay men and lesbians—is wrong not because it fails to live up to some color- or gender-blind ideal. Instead, the constitutional problem in Loving stems from the fact that the state sought racial “integrity.”29 Virginia wanted the races to stay in their lanes, maintaining their traditional, even God-ordained, roles in the world.30 Similarly, in cases like Hively, equality concerns arise when employers demand that both men and women stick to traditional gender roles. On this second approach, the problem is not gender consciousness as opposed to blindness; rather, it is gender stereotyping and, ultimately, gendered constraints on opportunity.

The majority opinion hints at this approach when it speaks of employers “policing the boundaries of what . . . behaviors they found acceptable for a woman (or in some cases, for a man).”31 The trouble is, the majority describes Hively as “represent[ing] the ultimate case of failure to conform to the female stereotype” because “she is not heterosexual.”32 Presumably, the majority would describe a man’s nonheterosexuality as a failure to conform to the ultimate male stereotype. But if the normative stereotypes are the same—”Be heterosexual!”—it becomes hard to see how this serves to keep men and women in their separate lanes.

This, I take it, is what Judge Sykes is getting at when she rejects the comparison to Loving. Antimiscegenation laws “use racial classifications toward the end of racial purity and white supremacy,” she writes.33 “Sexual-orientation discrimination, on the other hand, is not inherently sexist. No one argues that sexual-orientation discrimination aims to promote or perpetuate the supremacy of one sex.”34

This last sentence is breathtaking—not least because, understood a particular way, it is correct.

II. not in the opinions

The claim that no one argues that sexual orientation discrimination aims at the supremacy of one sex is true only if the people counted are the Seventh Circuit judges sitting en banc in Hively. If we expand our scope, however, to include nearly anyone who has written seriously about sexual orientation discrimination, gay rights more broadly, or gender theory more broadly still, the claim is astonishingly, flamboyantly wrong.

The fact that no one writing in Hively, at least this time around,35 tied LGBT discrimination to the policing of gender norms and the preservation of male privilege shows why the court’s blindness to scholarship beyond the field of statutory interpretation is significant. For it would be nearly impossible even to glance at the queer and gender theory or antidiscrimination scholarship of the last two decades without encountering the notion that sexual orientation discrimination has something to do with the subordination of women.36 And one could not engage with recent historical scholarship37 on the Civil Rights Act without questioning whether the “traditional notion of sex”—the originally understood meaning of Title VII’s “because of sex” clause—may be more complicated than the formalist opinions in Hively would have us believe.

As early as 1988, Sylvia Law and Andrew Koppelman argued that “contempt for lesbian women and gay men serves primarily to preserve and reinforce the social meaning attached to gender.”38 Judge Rovner’s panel opinion in Hively, now vacated, understood this point well.

Lesbian women and gay men upend our gender paradigms by their very status—causing us to question and casting into doubt antiquated and anachronistic ideas about what roles men and women should play in their relationships. Who is dominant and who is submissive? Who is charged with earning a living and who makes a home? Who is a father and who a mother? In this way the roots of sexual orientation discrimination and gender discrimination wrap around each other inextricably.39

Showing the connection among sexuality, gender stereotypes, and, ultimately, sex discrimination requires disentangling the gender norms the en banc majority conflates under the heading: Be straight.40 As Law argued in 1988 and many others have argued since, these norms are actually differentiated by sex: “Real men are and should be sexually attracted to women, and real women [should] invite and enjoy that attraction.”41 Law’s talk of “real men” and “real women” suggests what is at stake in differentiating the stereotypes applied to each gender. The Hively majority’s conflation does not just open it to the symmetry objection (i.e., that compulsory heterosexuality applies both to men and women and thus cannot comprise sex discrimination);42 it also obscures the way in which homophobia serves more broadly to police men’s and women’s respective spheres, each with its own standards for appearance, affect, activities, occupations, and desires. As Bennett Capers has written, “Women who are heterosexual, feminine, demure, and deferential to men are rewarded, as are men who are heterosexual, masculine, competitive, and protective of women.”43 Forcing men and women into stereotyped, gender-specific boxes of this sort has long been seen to violate Title VII.

The Supreme Court emphasized this in Price Waterhouse v. Hopkins in 1989 when it claimed that Congress, by passing Title VII, “intended to strike at the entire spectrum of disparate treatment of men and women resulting from sex stereotypes.”44Men who do not want women (while remaining sexually “inviolable” themselves),45 and women who fail to make themselves available to men, find themselves on this spectrum of stereotype violators, along with “effeminate” men and “mannish” or “butch” women, whatever their sexuality. As I have argued46 and the Hively panel opinion discussed at length,47 courts have for some time now accepted gender nonconformity claims from gay and lesbian plaintiffs under Title VII, but only if the claims were based on plaintiffs’ “gay” appearance or affect and not on the employer’s knowledge of their sexual orientation. As most of the judges in Hively recognize, this doctrinal result is untenable.48 It makes no sense to protect employees because they look or sound gay, but not because they are gay.49

The obvious solution is to take Price Waterhouse to its logical conclusion—to protect known, not just seen or heard, violations of gender norms related to sexual orientation.50. Admittedly, while Price Waterhouse has long prohibited employers from enforcing outmoded norms about how men versus women are meant to appear and behave, not all gendered differences are outlawed, as cases about uniforms, hairstyles, and makeup standards show.51 But this is why it is important to emphasize how certain gender norms substantively constrain men and women in separate spheres, with separate expectations and opportunities.52 Among these are expectations about women’s sexual availability and men’s impenetrability—stereotyping that the presence and relationships of gays, lesbians, and bisexuals tend to disrupt. This, rather than any linguistic claim about the meaning of “sex,” is what ties protection against sexual orientation discrimination to Title VII’s goal of dismantling constraining gender roles and hierarchies, in the workplace and beyond.53

But what about the dissent’s originalist retort that Title VII, as publicly understood at the time it was enacted, had no such sweeping goals—that the law simply made “it unlawful for an employer ‘to discriminate against women because they are women and against men because they are men’”?54 The problem with this response is that it uncritically assumes the very history on which its originalism depends. As Cary Franklin has compellingly demonstrated, the “traditional concept” of sex assumed by the dissent is a legal fiction, invented well after 1964. In actuality, “the legislative debate over Title VII’s sex provision emphasized . . . that [sex discrimination] was understood as a means of enforcing conventional sex and family roles.”55 Leading and influential advocates at the time argued that Title VII was not meant to “eradicate all formal sex classifications from the law, but to invalidate employment practices that pressed women into traditional roles.”56 Employers and legislators were concerned that the law “would upend traditional gender norms and sexual conventions, and disrupt forms of regulation that defined what it meant to be a man or a woman.”57 Franklin’s argument—and more recently, Vicki Schultz’s argument as well58—is that the meaning of discrimination “because of sex” is and has always been a contested one, where the extent to which gender roles and stereotypes should be disrupted by law has always been among the things at stake.

By offering an originalism that is blind to history, the dissent misses this point. But so too does the textualist majority, which fails to observe fully the social realities that give Title VII’s words their meaning.59 What results is yet another form of blindness in Hively, as its formalist arguments end up turning Title VII into a mandate for gender blindness in the workplace.60

Some might see this as a strategic move, made with an eye to the self-proclaimed colorblind Justices on the Supreme Court. But if that is the case, the Hively majority should not be over-optimistic. For another antidiscrimination scholar that they might have cited, Suzanne Goldberg, has documented the almost total failure of formalist sex-discrimination arguments in gay rights litigation, especially the marriage cases, in recent years.61 It is unclear why these arguments might suddenly start doing better now in the context of Title VII.

III. judge posner’s opinion

I have so far ignored Judge Posner’s candid and provocative concurring opinion—as the majority and dissenting opinions in Hively largely do as well. But Posner will not be ignored; aside from Hively’s historic holding, it is his distinctive approach to statutory interpretation that has garnered the most attention.62 Posner’s arguments are often orthogonal to those of both the majority opinion (which he joins) and the dissent, hence the separate discussion here. The discussion is merited, however, because of the one thing Posner’s concurring opinion shares with the others: a blindness to the sex-specific ways that gender stereotypes involving sexual orientation actually operate in the contemporary world.

Focused instead on theories of statutory interpretation, Posner begins his opinion by describing three: original meaning, which he (unlike the dissent) equates with “the meaning intended by the legislators”;63 unexpressed intent, which looks to the “spirit” of the law (and does not seem to correspond to any of the opinions in Hively); and Posner’s preferred method, a dynamic form of interpretation that he calls “judicial interpretive updating.”64

Avoiding “statutory obsolescence,” Posner argues, requires that judges sometimes impose meaning on statutes that they did not originally bear.65 It justifies taking “an admittedly loose ‘interpretation’”66 of words like ‘sex’ in Title VII—one that accounts, in this case, for the newfound recognition that “there are significant numbers” of gays and lesbians, and that their “sexual orientation is not evil and does not threaten our society.”67

The newfound recognition Judge Posner describes is undoubtedly his own. When he acknowledges the time it has taken “our courts and our society” to realize “that discrimination based on a woman’s failure to fulfill stereotypical gender roles is . . . a form of sex discrimination,” or that “homosexuality is nothing worse than failing to fulfill stereotypical gender roles,” Posner may well be recalling a 2003 concurring opinion where he wrote that “case law has gone off the tracks in the matter of ‘sex stereotyping’” when it is interpreted to imply “a federally protected right for male workers to wear nail polish and dresses and speak in falsetto and mince about in high heels.”68

Posner is more explicitly autobiographical when he describes the early 1960s as a time in his life when he had never “met a male homosexual,” and had “met a lesbian . . . ‘only in the pages of À la recherche du temps perdu.’”69 Posner’s anecdote will hardly reassure those who worry that his approach to interpretation may be overly idiosyncratic and inadequately constrained. After all, we have to wonder: did Posner just skip the pages in Proust about Charlus and Jupien to get to the lesbian parts?70

In fact, these worries suggest a missed opportunity in Posner’s opinion. Near the start, Posner analogizes Title VII to the Sherman Antitrust Act, which has long been interpreted dynamically “in conformity to the modern, not the nineteenth century, understanding of the relevant economics.”71 Posner might have taken his analogy more seriously. For judges do not just impose their own autobiographical experiences on the Sherman Act, striking down a proposed merger because their cable bill has grown over time. As Posner rightly notes, judges draw on the work of contemporary economists in seeking (or updating or creating) the meaning of the Sherman Act for our time. If Posner’s analogy is to hold, judges interpreting Title VII should presumably reach beyond their own experience by looking to those who have spent their careers studying the dynamics of sexual orientation, gender-based stereotyping, and subordination. But these are the sources that Judge Posner, like his fellow judges in Hively, chose not to see.

IV. conclusion

As other courts consider whether to join the Seventh Circuit’s historic holding, it will be incumbent on advocates to ensure that they see the gender policing that lies at the heart of anti-LGBT bias. If judges perceive that, no particular theory of statutory interpretation will be necessary to find protection under current law. Suitably informed originalists,72 textualists, purposivists, or even “judicial interpretive updaters” might all see in sexual orientation discrimination the very kind of sex-specific, opportunity-constraining stereotyping that Title VII has long been rightly understood to prohibit.

Brian Soucek is an Acting Professor of Law and Martin Luther King, Jr. Hall Research Scholar at the University of California, Davis School of Law. Thanks to Anya Bernstein, Jessica Clarke, Chris Elmendorf, and Courtney Joslin for helpful conversations and disagreements about Hively, and to Meenakshi Krishnan and her Yale Law Journal colleagues for their unusually quick and skillful editing.

Preferred Citation: Brian Soucek, Hively’s Self-Induced Blindness, 127 Yale L.J. F. 115 (2017),