sexual harassment

Can you sue your employer because your co-workers listen to raunchy radio programs?

A federal appeals court is reconsidering its 2008 ruling that you can. The Eleventh Circuit Court of Appeals’ decision in Reeves v. C.H. Robinson Worldwide said you could do so, under the dubious theory that it is “sexual harassment” that’s “based on” your sex. But on May 29, it voted to rehear that case.

U.C.L.A. Law Professor Eugene Volokh criticized the decision on First Amendment grounds, while I criticized the April 2008 ruling as being inconsistent with the language of the discrimination laws and the Eleventh Circuit’s own past rulings, and a threat to the media and freedom of the press in the long run. (Overly-broad interpretations of sexual harassment law are already being used to restrict a wide array of speech, as Professor Volokh, a First Amendment expert and former Supreme Court clerk, has chronicled).

Courts sometimes engage in inconsistent legal reasoning to impose liability and maximize damages in sexual harassment cases, and they sometimes disregard the statutory requirements (see, e.g., 42 USC 1981a) that harassment plaintiffs seeking compensatory damages show that they were harassed based on their sex, and subjected to intentional discrimination.

I don’t like raunchy radio programs, but that doesn’t mean the government should ban listening to them, much less do so under the weak argument that they constitute sex discrimination.

The April 2008 decision in Reeves v. C.H. Robinson Worldwide, 525 F.3d 1139 (11th Cir. 2008), was inconsistent with at least two of the Eleventh Circuit’s own past decisions: Baldwin v. Blue Cross, which defined sexual harassment as being a form of disparate treatment — not disparate impact — and Cross v. Alabama, which said that in the Eleventh Circuit (unlike some other circuits), discriminatory INTENT is required for a Title VII sexual harassment claim.

In Cross v. Alabama, 49 F.3d 1490, 1507-08 (11th Cir. 1995), the Eleventh Circuit held that “the elements of the two causes of action [sexual harassment claims under Title VII and the Fourteenth Amendment] are the same” in the Eleventh Circuit (unlike some other circuits), meaning that a sexual harassment plaintiff “must prove discriminatory motive or purpose.”

The Supreme Court has made clear that the fact that a woman is adversely affected, and that an employer knows this, is not enough to satisfy this exacting standard of showing a discriminatory purpose unless the employer intends to treat the female employee differently: “’Discriminatory purpose,’ however, implies more than intent as volition or intent as awareness of consequences. . .It implies that the decisionmaker . . . selected or reaffirmed a particular course of action at least in part ‘because of,’ not merely ‘in spite of,’ its adverse effects upon an identifiable group.” Personnel Administrator of Massachusetts v. Feeney, 442 U.S. 256, 279 (1979).

Thus, for the plaintiff to prevail, she must show that her co-workers aimed offensive comments at her based on her sex (or broadcast the offensive comments specifically to annoy her based on her sex), not just that she was reasonably offended by them and that they were pervasive.

“Title VII does not prohibit profanity alone, however profane. It does not prohibit harassment alone, however severe and pervasive. Instead, Title VII prohibits discrimination, including harassment that discriminates based on a protected category such as sex. Because a claim of sexual harassment under Title VII is a claim of disparate treatment, in order to prevail a plaintiff must show that similarly situated persons not of her sex were treated differently and better.” Baldwin v. Blue Cross, 480 F.3d 1287, 1301-02 (11th Cir. 2007); see also Lyle v. Warner Bros., 132 P.3d 211 (Cal. 2006) (sexual jokes not aimed at plaintiff did not support her sexual harassment claim).

Even if Title VII reached “disparate impact,” and even if provisions like 42 USC 1981a did not bar compensatory damages in such cases, it would still be a mistake to automatically equate raunchy language or discussions of sex with discrimination based on sex.

Assuming that sexual speech is disproportionately offensive to female employees and thus has a “disparate impact” on them — as many sexual harassment rulings do — raises serious equal-protection problems, since it rests on a gender stereotype. In striking down a statute banning “obscene, profane, indecent, vulgar, or suggestive” communications to women, a court observed that laws “based on ‘old notions’ such as a belief that females should be afforded special protection from ‘rough talk’ because of their perceived ’special sensitivities’ can no longer withstand equal protection scrutiny.” See In re Joseph T., 430 S.E.2d 523, 524 (S.C. 1993). Federal employment laws should not be interpreted broadly when doing so would raise serious constitutional issues. (See NLRB v. Catholic Bishop of Chicago, 440 U.S. 490, 499-501 (1979); Edward J. DeBartolo v. Fla. Gulf Coast Building & Constr. Trades Council, 485 U.S. 568, 575 (1988) (limiting law’s reach to avoid potential free speech problem)).

Moreover, treating raunchy language that offends a single female employee ignores limits contained in “disparate impact” law itself. Disparate impact claims typically require proof that an employer practice systematically excludes female or minority employees, and can’t be based on the impact on just one employee or a small number of employees. See Coe v. Yellow Freight, 646 F.2d 444 (10th Cir. 1981). Sexual harassment cases typically involve just a single plaintiff, who sometimes sues even if other employees of the same gender are perfectly happy with their workplace, or if there are no other employees of her gender in the workplace. In Reeves, the plaintiff seems to have been the only female employee adversely affected by the conduct she recounted.

An employer that fails to prevent sexual or vulgar discussions in the workplace should not be held liable simply because female employees overhear them and claim to be disproportionately offended by them. Some judges seem to grasp this rule, and do not allow sexual harassment claims to be brought just because the plaintiff was offended by sexual or vulgar speech. See, e.g., Lyle v. Warner Bros., 132 P.3d 211 (Cal. 2006); Scusa v. Nestle U.S.A., Inc., 181 F.3d 958 (8th Cir. 1998); Hocevar v. Purdue Frederick Co., 223 F.3d 721, 737 (8th Cir. 2000); Duncan v. Denver, 397 F.3d 1300 (10th Cir. 2005); Butler v. Ysleta Independent School District, 161 F.3d 263, 270-71 (5th Cir. 1998); Pasqua v. Metropolitan Life Ins., 101 F.3d 514, 517 (7th Cir. 1996); Gallant v. Board. of Trustees, 997 F. Supp. 1231, 1232, 1234-35 (N.D. Cal. 1998).

The 2008 ruling in Reeves justified its jettisoning of the discriminatory intent requirement by saying that such intent is not required in racial harassment cases. But even if that were in fact true (and it does not appear to be in some circuits, see Caver v. City of Trenton, 420 F.3d 243 (3d Cir. 2005) (holding that racial harassment plaintiff, just like sexual harassment plaintiff, must show that “(1) he suffered intentional discrimination because of his [race]” and “(2) the discrimination was pervasive and regular”), circuit panels are bound by their own past sexual harassment precedents whether or not they seem consistent with other areas of the law, like racial harassment.

And prior Eleventh Circuit rulings, like Cross v. State of Alabama, 49 F.3d 1490, 1507-08 (11th Cir. 1995), had made very clear that a plaintiff “must prove discriminatory motive or purpose” under BOTH Title VII and the Equal Protection Clause (Section 1983), because “the elements of the two causes of action are the same.” Eleventh Circuit judges had reiterated that understanding over and over again. See, e.g., Mitchell v. Pope, 189 Fed. Appx. 911, 913, 2006 WL 197600111, *1 (11th Cir. July 14, 2006) (”elements of the two causes of action are the same”); Downing v. Bd. of Trustees of Univ. of Alabama, 321 F.3d 1017, 1023 (11th Cir. 2003) (”Cross holds that the elements of a sexual harassment claim under Title VII and the Equal Protection Clause are the same — meaning that the employee must prove that the state actor intended to discriminate because of the employee’s sex”) (opinion later withdrawn for other reasons); Downing, 321 F.3d at 1022 n.9 (In Title VII, “this language is aimed at intentional discrimination, and to prove a violation, a plaintiff must profer either direct or indirect evidence of the employer’s discriminatory intent”).

Moreover, requiring discriminatory intent or purpose is consistent with how courts handle other types of Title VII harassment cases based on the very same language, such as religious harassment cases, where discriminatory treatment and intent are required. See Rivera v. Puerto Rico Aqueduct and Sewers Authority, 331 F.3d 183, 190 (1st Cir. 2003) (rejecting religious harassment claim because of absence of discriminatory purpose; “Rosario asserts that the lyrics of the song are offensive to her, given her deep religious convictions. We do not doubt this is so. But the question is not whether a religious person could find the song offensive; it is whether religious animus prompted Rivera to sing it to her”). The panel did not explain why those cases should be given lesser weight in sexual harassment jurisprudence than racial harassment cases.

Getting rid of the requirement of discriminatory intent, and allowing suits over overheard comments that have a disproportionate impact on one gender or the other, also raises serious First Amendment problems. See Lyle v. Warner Bros., 132 P.3d 211, 300 (Cal. 2006) (Chin, J., concurring) (sexual harassment liability cannot constitutionally be applied to creative speech, such as that involved in books, newspapers, or television, that is “not directed at or about the plaintiff”). Positions on many sexual or gender issues like feminism, gender-based affirmative action, and how to define sexual harassment, are offensive to some listeners, and are sometimes alleged to be disproportionately more so to one gender than the other (there is a political gender-gap). Moreover, the discriminatory intent requirement is one of the few elements of harassment law that can keep it relatively clear and manageable.

“Sexual harassment” under federal case law is a term of art referring to whatever speech or conduct gives rise to a specified state — a “hostile work environment” — not to “harassment” as defined in the dictionary, or anything closely resembling a traditional tort, like a pattern of invasive conduct by one individual towards another. To be sure, a hostile environment can be created by one malicious individual repeatedly pestering a plaintiff. But it can also (if discriminatory intent is not required) be created gradually by many different people each individually making just one offensive statement within earshot of the plaintiff, if their statements cumulatively add up to a hostile environment over time, even if they harbored no ill-will or discriminatory animus towards the plaintiff. The result, as UCLA Law Professor Eugene Volokh notes, is to effectively force employers to suppress a vast array of otherwise-protected speech. If the discriminatory-intent requirement is not enforced, speech such as classical nude paintings and religious articles in employee newsletters would become potential building blocks of a hostile-environment claim. (By “discriminatory intent,” I merely mean gender-based differential treatment; malice is not always required. See UAW v. Johnson Controls, 499 U.S. 187 (1991)).

To be sure, there are other elements of a harassment claim, like the requirement that the plaintiff show that the conduct complained of is “unwelcome” and occurred “based on sex.” But the very court rulings that ignore the intent requirement often have gutted those other elements, so that they frequently add nothing relevant to the requirement of a “hostile work environment.”

For example, people have responded to my past observation that a harassment plaintiff must show discriminatory intent by noting that harassment is only actionable if it is “unwelcome,” and suggesting that the fact that harassment is “unwelcome” somehow makes it intentional even absent discriminatory motivation or targeting of the plaintiff. See Meritor Savings Bank v. Vinson, 477 U.S. 57, 68 (1986) (the essence of “any sexual harassment claim is that” the conduct was “unwelcome”).

But the courts that don’t enforce the requirement of discriminatory intent also often interpret “unwelcomeness” so broadly as to make it meaningless as a limit on harassment claims. They assume that harassment is “unwelcome” whenever a plaintiff is subjectively offended by it, even if the defendant had no way of knowing that, or even if the plaintiff used language similar to the defendant, which could have led the defendant to believe that the language was permissible. See, e.g., Galloway v. General Motors, 78 F.3d 1164 (7th Cir. 1996) (dictum); Gary v. Tyson Foods (W.D. Mo. 1999). This matters because, as Fifth Circuit Judge Edith Jones has observed, a common fact pattern in harassment cases is a plaintiff who sues over comments similar to those she herself has frequently made.

Defined purely subjectively, “welcomeness” thus adds nothing to the separate legal requirement that the plaintiff’s work environment be “subjectively hostile.” See, e.g. Harris v. Forklift Systems, 510 U.S. 17, 21-22 (1993) (work environment must be subjectively hostile for plaintiff to recover); Mendoza v. Borden, 195 F.3d 1238, 1245-1246 (11th Cir. 1999)(same). That violates the principle of tort law that one element of a tort should not be read as being duplicated, or subsumed by, another element of the tort.

An old Eleventh Circuit precedent interpreted unwelcomeness as having an objective component, so that a plaintiff who incited a defendant to say offensive things by saying similarly offensive things of her own could not sue based on them. See Henson v. City of Dundee, 682 F.2d 897, 902 (11th Cir. 1982) (to show unwelcomeness, plaintiff must show that she neither “solicited nor incited” the conduct and that she “regarded the conduct as undesirable or offensive”). Thus, it served a notice function that protected some speech. See, e.g., Brief of Center for Individual Rights and Boston Coalition for Freedom of Expression as Amici Curiae in Support of the Petition for Certiorari in Avis Rent A Car System v. Aguilar, 529 U.S. 1138 (2000) (citing cases and EEOC rulings that interpreted “unwelcome” as having an objective component).

Some other federal appeals courts, like the Eighth Circuit, continue to follow this rule, barring plaintiffs from hypocritically suing the employer for comments made by their co-workers that are similar to those made by the plaintiffs themselves, under the theory that they have objectively welcomed such comments. See, e.g., Scusa v. Nestle U.S.A., Inc., 181 F.3d 958 (8th Cir. 1998); Hocevar v. Purdue Frederick Co., 223 F.3d 721, 737 (8th Cir. 2000).

But some courts (especially state courts) do not, viewing the requirement as unfashionable and outdated. By interpreting “unwelcomess” as purely subjective, they have effectively dispensed with any notice to the accused, who is now deemed a harasser if his speech unintentionally creates a hostile environment.

What is a “hostile environment” can be very vague, as a federal appeals court noted in Pasqua v. Metropolitan Life Ins. Co, 101 F.3d 514 (7th Cir. 1996). Indeed, a Maryland civil rights agency warned that “the legal boundaries” of what is actionable “are so poorly marked,” that “the best course of action is to avoid” any potentially offensive remarks. Reflecting the ambiguity, federal appeals courts differ greatly among themselves as to what conduct they believe is severe or pervasive enough to create a hostile environment. Within the same district, different juries often find starkly similar conduct to either not constitute harassment at all, or to be so patently and egregiously harassing as to warrant punitive damages. I used to work as an attorney at the U.S. Department of Education’s Office for Civil Rights, and there was certainly nothing clear about how we interpreted our sexual harassment guidance.

When someone other than the courts is enforcing a ban on harassment, the courts immediately recognize the disturbing vagueness of the “hostile environment” concept. For example, the Ninth Circuit overturned a public college professor’s discipline for unintentionally creating a hostile environment on vagueness grounds in Cohen v. San Bernardino Valley College, 92 F.3d 968 (9th Cir. 1996). And the Oregon Supreme Court, on state constitutional grounds, overturned a civil-rights agency’s fine on an evangelical Christian employer for unknowingly creating a hostile work environment through religious proselytizing in Meltebeke v. Bureau of Labor and Industries, 903 P.2d 351, 363 (Or. 1995), even though it found that the conduct indeed violated the agency’s hostile-environment harassment rule, because ignoring the employer’s intentions would have a chilling effect on freedom of religion. And in Dambrot v. Central Michigan University, 55 F.3d 1177 (6th Cir. 1995), the court struck down a college’s hostile environment harassment code as unconstitutionally vague, since whether the code was violated turned partly on the subjective reactions of listeners.

But the courts are sometimes blind to the vagueness of the “hostile work environment” concept when they themselves apply harassment law through damages liability — even though the First Amendment vagueness doctrine applies to civil as well as criminal liability, see Bullfrog Films v. Wick, 847 F.2d 502 (9th Cir. 1998), and the Supreme Court has long recognized that the “fear of damage awards” in a lawsuit can chill speech even more than “the fear of prosecution.” See New York Times v. Sullivan, 376 U.S. 264, 277 (1964).