When is sex because of sex? The causation problem in sexual harassment law.

• Author: Schwartz, David S.

"[S]exual harassment is ordinarily based on sex. What else could it be based on?"

--Judge Stephen Reinhardt, in Nichols v. Frank (1)

INTRODUCTION

In 1991, Lois Robinson, a female welder who had worked in a mostly male shipyard for over a decade, won a judgment against her employer for sexual harassment. (2) Robinson proved that the work atmosphere was infested with pornographic pictures and graffiti, which constituted a "visual assault on the sensibilities of female workers." (3) Male co-workers frequently made graphic sexual comments in front of Robinson, referring to this plentitude of pornography. (4) They put misogynistic sexual graffiti on the walls in her working areas and made her the object of sexual comments. (5) Robinson did not allege that any of this abusive sexual conduct was in the nature of a sexual "overture," or that the harassers had a particular sexual interest in her. Yet the court had no trouble concluding that the harassment was "because of sex": "[S]exual behavior directed at women will raise the inference that the harassment is based on their sex." (6)

In July 1997, in Doe v. City of Belleville, (7) an appellate court ruled that a sixteen-year-old boy, whose male co-workers had subjected him to sexual epithets, grabbed his testicles, and even threatened him with rape, had sufficiently established "discrimination because of sex" to proceed to trial on his sexual harassment claim under Title VII. (8) According to the court, there would have been no question that this was sexual harassment had the plaintiff been a woman, and the court found it clear that this too was a case of "discrimination because of sex," notwithstanding that the harassers and the victim were both male. (9) In a striking example of plain good sense, the court stated: "Frankly, we find it hard to think of a situation in which someone intentionally grabs another's testicles for reasons entirely unrelated to that person's gender." (10)

Robinson and Doe differ in the obvious respect that the former involved male-on-female and the latter male-on-male, or "same sex," harassment. Yet the cases have a common theme and bookend what was an important advance in sexual harassment doctrine: the recognition that harassing sexual conduct may be motivated by something other than sexual attraction toward the victim. Instead, courts were coming increasingly to the understanding that "why the harassment was perpetrated (sexual interest? misogyny? personal vendetta? Misguided humor? boredom?) is beside the point." (11) This notion was embodied in what might be called the "sex per se" rule: that whatever other conduct might constitute sexual harassment, and whatever other elements might be required to prove actionable sexual harassment, sexual conduct per se established the "causation" element necessary under Title VII to prove that the conduct was "because of sex." (12)

In the year following Doe, the law of sexual harassment underwent major development in both doctrine and academic theory. Four leading theoretical articles on sexual harassment by progressive "second generation" feminist scholars (13) "sought to reconceptualize the wrong of sexual harassment so as to correct conspicuous errors and set the claim on a sound future course." (14) As the last of these was being published, at the end of the 1997-1998 Term, the U.S. Supreme Court issued three opinions on workplace sexual harassment. (15) One of these, Oncale v. Sundowner Offshore Services, Inc., (16) held that a Title VII plaintiff could state a claim for sexual harassment notwithstanding that the harasser was of the same sex. (17) The coincidence of these Supreme Court decisions and the related works of scholarship suggested "the arrival of a jurisprudential moment characterized by reinvigorated theorizing about the appropriate legal response to sexual harassment." (18) One federal appellate court paid a noteworthy tribute to these articles, which "reexamine the theoretical underpinnings of sexual harassment law," and pronounced that, due to the conjunction of this scholarship and the Supreme Court rulings, "[w]e are witnesses to the birth of a second generation of sexual harassment law." (19)

Ironically, in the aftermath of these significant developments it is no longer clear--as it was prior to 1998--that a man or a woman who experiences severe or pervasive sexual epithets, crotch-grabbing, or even threatened rape has experienced "discrimination because of sex." In attempting to reconceptualize sexual harassment, both the second generation theorists and the Court reopened the question of causation and disturbed, if not rejected, the unquestioned assumption that sexual conduct in the workplace is per se "because of sex."

Second generation sexual harassment scholarship is characterized by efforts to solve various problems coming under the label "essentialism," which, in the context of sex discrimination, is the troublesome notion that womanhood is a monolithic concept, and that there is a single problem of sex discrimination that can be solved by a unified solution. (20) For several years, feminist scholars have sought to advance beyond the classic antisubordination theory that Catharine MacKinnon articulated in Sexual Harassment of Working Women. (21) While acknowledging the great debt owed to MacKinnon and sharing her central idea that sexual harassment perpetuates the subordination of women in the workplace, the second generation theorists move away from MacKinnon's essentialism. In particular, these second generation scholars diverge from her underlying thesis that male sexual conduct inherently subordinates women--that male sexuality "is always, already sexist." (22)

In Oncale, a unanimous Supreme Court cast doubt on the sex per se rule, stating that "[w]e have never held that workplace harassment, even harassment between men and women, is automatically discrimination because of sex merely because the words used have sexual content or connotations." (23) In so doing, the Supreme Court reexamined the question of causation, but conspicuously declined to adopt the feminist theorists' antisubordination commitments. While Oncale, on the surface, was a victory for the plaintiff, the Court went out of its way to avoid endorsing a theory of harassment on account of gender nonconformity and, in so doing, narrowed the avenues by which female plaintiffs in the most common sexual harassment cases can prove causation. At the same time, the Oncale Court embraced a sexual desire theory of causation that can be turned against gays and lesbians.

The second generation theorists, in effect, had invited the courts to reopen the question of causation in sexual harassment cases in order to solve the problems of essentialism: to embrace a prohibition on harassment on account of gender nonconformity, to broaden the understanding of harassment to include nonsexual but subordinating conduct, and to eliminate paternalistic "anti-sex" attitudes that surface in occasional harassment decisions. The Oncale decision joins the second generation theorists in breaking down an unexamined but useful consensus that was beginning to emerge among courts and the older theories of sexual harassment; namely, that sexual conduct is per se because of sex. But it does so without giving us the world the second generation theorists were looking for: proof of causation may now be more difficult for the H. Does and the Lois Robinsons. Thus, while it may be that the courts have accepted the invitation, they seem determined to behave badly at the party.

It should come as no surprise that the Court in Oncale failed to pursue the antisubordination agenda once it reopened the question of the meaning of discrimination "because of sex." In this, Oncale exposes a problem in the antisubordination theories of sexual harassment themselves. Courts have always taken seriously the question of "causation" in employment discrimination law: the element of proof that an otherwise lawful workplace act is "because of" sex, race, or the other prohibited grounds. But recent antisubordination theorists may have paid insufficient attention to causation, particularly to the need to define "because of sex" in a way that bridges their theories to the doctrine in the courts.

In Part I of this Article, I examine what I call the "sex per se" rule. To trace its development in sexual harassment caselaw and explain the doctrinal function that it has fulfilled, I first examine the issue of causation in employment discrimination law. In cases of intentional discrimination, courts have always required some proof that the alleged discriminatory conduct was "because of" race/sex/etc. This "causation" element, which refers to proof of a conscious discriminatory intent, serves a gatekeeping function to distinguish discriminatory from presumptively lawful conduct. In harassment cases involving sexual conduct, however, the sex per se rule has emerged as an evidentiary shortcut, relieving plaintiffs of the burden of proving the harasser's motivation in targeting the plaintiff.

In Part II, I analyze the Oncale decision and its potential consequences. Although the plaintiff "won" in the Supreme Court, the implications in the opinion for this plaintiff on remand placed him in more of a losing position. Moreover, the ambiguous quality of the opinion extends to sexual harassment law more generally. While the opinion seems on one reading to open new opportunities for including broader categories of sexual harassment claims and claimants, it simultaneously creates the likelihood of retrenchment. The opinion seems to do away with the sex per se shortcut to proof of causation, thereby making it more difficult for many plaintiffs to prove that the harassment was "because of" sex. At the same time, Oncale fails to extend the protection of Title VII antiharassment doctrine to gays and lesbians. In important respects, Oncale looks more like a "reverse...