A practical and philosophical proposal for sexual harassment law.

In Harm's Way: The Pornography Civil Rights Hearings

edited by Catharine A. MacKinnon and Andrea Dworkin

(Harvard, 496 pp., $24.95)

Heterophobia: Sexual Harassment and the Future of Feminism

by Daphne Patai

(Rowman & Littlefield, 288 pp., $22.95)

I.

In February, Yale Law School sponsored a conference to celebrate the twentieth anniversary of the publication of Catharine MacKinnon's Sexual Harassment of Working Women. The Paula Jones trial seemed still likely to proceed on schedule, and there was a tincture of defensiveness in the air as conference participants dismissed the growing chorus of criticism that harassment law was losing its moorings. Andrea Dworkin thoughtfully ridiculed the critics of sexual harassment law as "millions of men who want to have a young woman in the workplace to suck their cock." Jane Larson knowingly declared that "more than one person may have come to orgasm in that White House closet," and insisted that "sex harassment law benefits women who seek sex at work and those who resist." And Frederick Schauer, the Frank Stanton Professor of the First Amendment at Harvard's Kennedy School of Government, dismissed as "frivolous" concerns that current harassment law might inhibit free speech. All this certitude was a sure sign that all was not well with the cause.

There were useful presentations at the conference, notably on the manner in which sexual harassment law operates in the cases that actually go to trial. Cass R. Sunstein noted that in the nearly 70 reported cases, there was no correlation between the factors that ought to justify higher or lower jury awards--such as coerced sex, bodily contact, or harassment of other employees- -and the damages actually awarded. What was missing from the proceedings, however, was a critical examination of the first principles of sexual harassment law. There was remarkably little discussion, for example, about the extent to which the liability rules of modern harassment law have transformed corporate workplaces and universities, creating strong incentives for prudent employers to restrict far more speech and behavior than the law actually forbids. And more abstractly, there is the urgent and indelicate question of the relationship of sexual harassment law to liberalism. For sexual harassment law makes certain assumptions about legal personhood, and about the objectivity of law, that seem profoundly inconsistent with the liberal ideal.

The truth is that, practically and philosophically, something has gone terribly wrong. As the Supreme Court prepares to decide three important cases that could redefine the liability rules in sexual harassment cases, it is worth asking whether or not the costs of the existing liability regime outweigh the benefits. More ambitiously, the time has come to resurrect a vision of gender-blindness that the current harassment regime has spurned. In contemporary American life, gender-blindness may seem even more impracticable than color-blindness; but in the face of empirical evidence suggesting that men and women, in the aggregate, perceive the same situations in intractably different ways, the need for a legal standard that refuses to honor these differences, and promote them into the last word on the subject, is increasingly urgent in a democracy founded on notions of legal neutrality and equal citizenship.

II.

Most people are surprised to learn that sexual harassment law does not impose liability on sexual harassers. Instead it puts the full weight of responsibility on their employers. And owing to the incentives created by this liability regime, prudent companies have little choice but to restrict a great deal of sexual expression that no jury would ultimately condemn. The law has transformed inquisitions into the emotional lives of employees into an ordinary matter of corporate self-interest.

In 1986, the Supreme Court held that employers can reduce the danger of being held liable for sexual harassment by establishing procedures " calculated to encourage victims of harassment to come forward." Since then, almost 75 percent of companies with more than one hundred employees have adopted anti-sexual harassment policies. Most of those policies look very much like the "Sample Antiharassment Policy" reproduced in Barbara Lindemann and David Kadue's Sexual Harassment in Employment Law, which first appeared in 1992. The model policy begins by quoting the EEOC's definition of sexual harassment: "Unwelcome sexual advances, requests for sexual favors, and other physical, verbal, or visual conduct based on sex constitute sexual harassment when (1) submission to the conduct is an explicit or implicit term or condition of employment, (2) submission to or rejection of the conduct is used as the basis for an employment decision, or (3) the conduct has the purpose or effect of unreasonably interfering with an individual's work performance or creating an intimidating, hostile, or offensive working environment." And the model policy goes on to say that "sexual harassment may include explicit sexual propositions, sexual innuendo, suggestive comments, sexually oriented kidding' or teasing,' practical jokes,' jokes about genderspecific traits, foul or obscene language or gestures, display of foul or obscene printed or visual material, and physical contact such as patting, pinching, or brushing against another's body."

By itself, of course, a suggestive comment or a practical joke from a boorish supervisor or coworker wouldn't be sufficiently "severe or pervasive" to be illegal under the Supreme Court's definition of harassment. But a hostile environment can be created by nothing more egregious than a pattern of comments or jokes from different employees; and so, to protect themselves from the risk of ruinous liability, companies face increasing pressure to launch formal investigations in response to relatively trivial offenses. "If you are in management, now is the time to refine your sexual harassment policy so that it is in the spirit of your organization's values rather than written to the letter of the law," writes Rita Risser in a report in 1996 by Fair Measures Management Law Consulting Group. "Your policy should go beyond what the law forbids . If you set your standards too low, one mistake by one supervisor could make you the next landmark case. Also, the EEOC accepts claims for conduct that clearly is not illegal. Since it's costly to respond to such claims, it's in an organization's best interest to minimize them."

The Sample Antiharassment Policy goes on to provide a grievance mechanism for employees to report harassment complaints. "All ABC employees are responsible to help assure that we avoid harassment. If you feel that you have experienced or witnessed harassment, you are to notify immediately Mr. -- ---- or Ms. ------ in the Department of Human Resources... The Company's policy is to investigate all such complaints thoroughly and promptly. To the fullest extent practicable, the Company will keep complaints and the terms of their resolution confidential."

The fact that harassment policies allow complaints by offended third parties who think that they have witnessed harassment, regardless of the feelings of the purported victim, exemplifies the stereotypes that the law now embodies. Imagine, for example, the case of a male supervisor who engages in well-intentioned but suggestive teasing with a female employee. She isn't offended, and she doesn't believe that his remarks interfere with her ability to do her job. Under most company harassment policies, if a coworker overheard the conversation and felt offended, she would have a duty to notify the Department of Human Resources and might have a cause of action on her own.

The company would then be compelled to launch a full-blown investigation, interviewing the alleged harasser and the putative victim, and all the other employees who had come into contact with either of them, about intimate details of their relationships. Even if the woman who was the object of the contested remarks continued to insist that she did not feel victimized, the model policy wouldn't permit the company to drop the matter without allowing the brutal investigative mechanisms to grind to their conclusion.

As the quasi-legal procedures took their toll, the alleged harasser and the alleged victim might feel more ill-treated by the invasive investigation than by anything that had been said--the harasser, because he suddenly found himself forced to defend his career over remarks that no one had ever told him were offensive, and the victim, because she did not appreciate being cast as a victim against her will.

Beset by pressures from all sides, the purported victim and the purported harasser might justifiably feel that they were being used as symbols to ventilate the grievances and the frustrations of their colleagues. Yet none of this would matter. Despite the obvious costs of harassment policy--the disruption of workplace relationships, the harm to reputations, and the invasions of privacy--even a liberal employer would not have the option of resolving the dispute quietly, by taking the supervisor aside and telling him to watch his language.

The scenario that I've described is a machine without mercy. It does not empower individual men and women; it demeans them, and robs them of their autonomy. And it is the ineluctable consequence of the liability rules currently under examination by the Supreme Court.

Companies are generally held "strictly liable" for cases of "quid pro quo" harassment, in which supervisors fire or demote or punish employees who refuse to have sex with them. Strict liability means that the company management is held responsible even if it did not know that the harassment was occurring, and even if it has adopted a well-publicized policy designed to prevent harassment. When supervisors flamboyantly abuse the authority that the company has delegated to them, the argument goes, their acts should be imputed to the company as a whole.

In cases involving "hostile environment" harassment, courts tend to impose a lower liability standard. The general rule is that companies should be held responsible only when they are negligent--that is, when they actually knew about the harassment and failed to do anything to stop it, or when they should have known about it but failed to adopt policies designed to prevent it. The effect of the negligence standard has been to encourage companies to adopt anti-sexual harassment policies to shield themselves from liability. But lower courts aren't sure whether or not the adoption of a policy automatically gets a company off the hook, and so they are increasingly inclined to ask whether the policy was effective in deciding the company's ultimate liability. This only increases the pressure on companies to root out as much sexual expression as possible.

So the defenders of harassment law must be wrong when they deny that harassment law has provided incentives for companies to regulate speech and behavior to a degree that would have been unimaginable when MacKinnon published her pathbreaking book two decades ago. The question is not whether the current harassment regime has led to excesses; it is whether we should accept those excesses as a transitional phase in the transformation of workplace norms that has followed the integration of women into the workplace.

In a provocative presentation at the Yale conference, Robert Post suggested that the corporate workplace is a "managerial" sphere in which social relations are organized around principles of efficiency. For this reason, he argued, citizens should be willing to accept greater restrictions on their autonomy and their expression in the workplace than they would tolerate in the public sphere, which is ideally a space of self-governance.

But surely this Taylorite vision of the modern workplace, rooted in principles of industrial democracy in the 1920s, is hard to accept today. As e-mail, modems, and PCs break down the boundaries between work and home, there are increasingly few private or public spaces for citizens to express themselves autonomously. Moreover, the range of workplaces regulated by modern harassment law--which includes not only widget factories but newspapers, theaters, and lifeguards on beaches--is too diverse to be captured by the managerial model. Without the freedom to test controversial ideas with colleagues, journalists (to take an example) cannot properly do their job; and actors (to take another example) are not exactly devoted to efficiency above all.

As harassment law has developed, it seems ill-equipped to make the fine distinctions among workplaces that Post imagines. Surely public universities and high schools should be regulated by less restrictive behavior codes than corporate businesses. The purpose of a university is to promote freedom of thought, not to increase productivity; and holding schools to the same standards as corporations arguably poses a direct threat to the experience of teaching and learning. The Supreme Court is now debating whether liability rules should be different for universities and corporate workplaces, but political pressures to adopt harassment codes are already extending the restrictions that originated in corporations into the academic sphere.

In her powerful new book, Daphne Patai argues that anti-harassment policies, in universities at least, have produced "not greater justice, not an absence of discrimination against women, but a climate that is inhospitable to all human beings." Patai includes a close reading of the leading harassment manual for colleges and universities, Sexual Harassment on Campus: A Guide for Administrators, Faculty, and Students, edited by Bernice R. Sandler and Robert J. Shoop. Its first chapter, entitled "What Is Sexual Harassment," enumerates a list of indignities, extending from "sexual innuendos, comments, or bantering," through "humor or jokes about sex or females in general"; it also includes "touching a person," "giving a neck or shoulder massage," " leering or ogling, such as elevator eyes,'" "calling women hot stuff,' cutie pie,' etc.," "sexual graffiti," and "laughing at or not taking seriously someone who experiences sexual harassment."

Sandler and Shoop also include a chapter on "student to student harassment," which they think is on the rise. The attempt to stamp out peer harassment, in public schools as well as public universities, is a useful window on the excesses of the current liability regime in harassment law. Children, after all, are not adults, and they can't be expected to conform their behavior to adult norms. Teasing, poking, grabbing, and clumsy horseplay are a natural part of growing up; and if publicly funded schools were held liable for failing to root out all sexual experimentation between students, they would be vulnerable for damage awards far greater than the federal grants that created the liability in the first place. They would also transform themselves into cruelly stunting environments.

The question of how responsible schools should be for peer harassment is an open legal question; but a few months ago Judge Richard A. Posner suggested in an appellate opinion that even a negligence standard would be too demanding. In an age in which parents are suing teachers for attempting to clip their children's fingernails, forcing schools to take Draconian measures to banish all horseplay from the classroom would be as likely to expose a school to suits as the failure to protect the potential victims in its care. For this reason, Posner proposed that schools should not be liable for peer harassment unless they were deliberately or recklessly indifferent to it: unless, in other words, they actually knew about hostile or offensive conduct likely to interfere with a student's education, and made no efforts to stop it.

It is easy to imagine the hazards of holding schools liable for sex discrimination when their students grab each other; but the question of a school's liability when teachers harass students, which is now before the Supreme Court, raises similarly troubling questions. The facts of the Supreme Court case are stark. A Texas social studies teacher named Frank Waldrop seduced one of his ninth-grade students, a girl named Alida Gebser, and had sex with her for a year. As soon as he was discovered, he was fired; later, he was jailed for sexual assault.

The school district doesn't dispute that the conduct was criminal; it says that since the girl never told anyone about her ordeal, including her parents, school officials should not be liable for conduct that they couldn't have been expected to discover. Gebser's parents counter that the school should have known that one of its teachers was abusing his student; and since the school failed to promulgate a well-publicized anti-sexual harassment policy, it should have to pay a lot of money.

The Gebser case suggests that well-publicized harassment policies are unlikely to deter the most egregious sexual abusers. Alida Gebser did not complain because she was confused and traumatized by her ordeal; but the procedures were not wanting. Later she testified that she was afraid that she would lose Waldrop as a teacher if she told anyone about their relationship.

All this calls into question the argument, made by the Clinton administration and women's advocacy groups, that sexual harassment in schools is "eminently foreseeable." The kind of monitoring that would be necessary to root out secret relationships such as the one in the Gebser case would disrupt far more than secret sexual relationships between teachers and students. To avoid liability, schools that accept public funds would have a powerful legal incentive to adopt speech codes to regulate the behavior of their students and teachers, even though the Supreme Court held in 1992 that speech codes adopted by legislatures are presumptively unconstitutional. In trying to navigate the legal shoals, schools would suffer a serious threat to academic freedom, suppressing a great deal of innocent speech for the dubious purpose of holding schools liable for conduct that is already illegal under federal and state law.

The Supreme Court might decide to treat schools and universities differently than other workplaces, on the grounds that free expression is central to the idea of a university, and the costs of monitoring the behavior of teachers and students would chill the learning process. Yet the workplace at issue in the second Supreme Court case involves lifeguards on Boca Raton Beach, and this mise-en-scene is similarly hard to construe as a "managerial" sphere.

Beth Ann Faragher says that she was subject to a series of tribulations during the five years that she worked as a lifeguard in Boca Raton to put herself through college. Her supervisors, she says, invited her to the unisex shower with them, used vulgar language in her presence, complimented her body, and grabbed her on occasion; and one of them tried to kiss her after she agreed to go on a date. Yet Faragher never told the offending parties that she found the conduct offensive during her five years on the job. Nor did she complain to anyone in the city management, though she did tell another supervisor, who thought she was consulting him as a friend and failed to report it to anyone else.

At what level must a woman complain before a public employer will be held liable for harassment? Like Alida Gebser's parents, Beth Faragher says that the city of Boca Raton should be held liable, because it had an anti-sexual harassment policy but failed to publicize it properly. It is hard to believe, however, that a better-publicized policy would have encouraged Faragher to file a complaint, since she refused to do so even when urged by a female coworker, who was filing a complaint of her own.

In its Supreme Court brief, the Clinton administration argues that employees should not be expected to complain in hostile environment cases because they may fear retaliation from their supervisors, even if there is no explicit threat. "When the employer's practices and the supervisor's conduct reasonably have led an employee to fear that adverse employment consequences will result if she resists or complains," writes the Solicitor General, the employer should be liable, whether or not it has adopted a harassment policy. Yet the "hostile environment" doctrine is an amorphous one, according to which a man's joke can be a woman's lawsuit, and so it seems dangerous to create a strict liability standard that allows women to sue for boorish behavior that they never indicated was unwelcome.

The administration is taking a similarly paternalistic position in the third Supreme Court case, Burlington Industries v. Ellerth. This case has been compared to the one brought by Paula Jones. Kimberly Ellerth says that her boss made sexually suggestive comments, told dirty jokes, commented on her breasts and legs, and once, while leaving a hotel bar after a business trip, said, "You know, Kim, I could make your life very hard or very easy at Burlington." Ellerth understood this as a quid pro quo threat.

But she ignored it, and she was promoted anyway. Two months later, without mentioning anything about harassment, Ellerth resigned, after other bosses told her that customers and colleagues had complained about her work. Three weeks later, she accused her supervisor of harassment for the first time. Ellerth says that she never filed a formal complaint or invoked the company's harassment policies because she was afraid of losing her job. Arguing, as lawyers say, in the alternative, she also says that she quit because the harassment became "unbearable." But if she knew that she was going to quit, why didn't she complain?

The technical question in the Ellerth case is what the liability standard should be for harassment cases that involve unexecuted threats but no tangible job injuries. In ordinary quid pro quo cases, again, the company is liable whether or not it knew about the harassment, on the theory that firing is an official act that companies should be expected to monitor carefully. But, as Posner argued in his appellate opinion in the Ellerth case, it is infeasible to expect employers to stamp out all consensual affairs in the workplace between supervisors and employees:

Romantic encounters, including romantic encounters between supervisors and supervised, are a fact of the workplace. Title VII does not purport to forbid them, and would be quixotic if it did. Many happy marriages have grown out of such encounters... The words, the gestures, the other behaviors that differentiate the fully consensual relationship from the coercive relationship will often be invisible to the supervisor's supervisor. The yielding to a threat will look no different from the yielding to a lawful proposal. It is only when the threat is carried out that the abusive supervisor does something, such as firing the supervised employee, that the employer will know about and should monitor. It is facile to suggest that employers are quite capable of monitoring a supervisor's actions affecting the work environment. Large companies have thousands of supervisory employees. Are they all to be put under video surveillance? Subjected to periodic lie detector tests? Trailed on business trips by company spies?

For these reasons, Posner proposes that companies should be strictly liable for their supervisor's threats only when the threats are actually carried out; and all other advances should be governed by a negligence standard, which means that the companies may (or may not) be able to avoid liability by adopting a harassment policy.

Behind the legalisms about liability, the three Supreme Court cases raise a single question, and it pertains as much to social policy as to law: Should a woman have to make clear that a man's behavior is offensive before the man is punished? Some of the most interesting empirical data at the Yale conference were presented by Louise Fitzgerald, a social psychologist at the University of Illinois, who suggested that men and women often perceive the same social interactions very differently. "Men are more likely to perceive sexual motives and intentions," Fitzgerald said, pointing to one of her interview subjects who insisted that he was being propositioned at a lunch counter by a woman who had accidentally stepped on his foot.

From this gap in perception, Fitzgerald concludes that it is unrealistic to expect women to complain about behavior that they find offensive. " Psychological literature says that the real victims do nothing," she said. " One in five women directly tell men to stop; only five percent complain to their employer." According to Fitzgerald, this doesn't mean that women are passive in the face of sexual insult. Her subjects show a range of behavior: avoiding or appeasing the offending man, or remaining silent and refusing to give him the satisfaction of a response. Since "coping with abuse is a psychological process," Fitzgerald concluded, the law is wrong to insist that the only valid complaint is a formal complaint.

Fitzgerald proposed instead that certain acts should be considered " harassing per se," such as calling women by names traditionally reserved for body parts, whether or not a woman complains. In such cases, the burden should shift onto the man to prove that the words were welcome. But this conclusion seems exactly backward. Fitzgerald is correct that "men in general see sex where it isn't there and insist that it is there," but in light of this gap in perception, shouldn't the liability rules be set up to encourage women to make their own very different perspective clear, by objecting to behavior that they find objectionable? Otherwise, boorish men in the workplace may be unable to change their behavior until it is too late, because no one ever told them that their behavior was unacceptable. (The best data suggest that the most effective way to stop harassment is to ask the harasser to stop.) Or to put it differently, an essentialist view of gender differences--for that is what Fitzgerald is peddling, even if the essentialism comes in social-scientific camouflage--makes it even more imperative that we think hard about the roots of the sexual harassment law that is now loose in the land.

III.

It may seem surprising that the Supreme Court has not yet made clear whether or not women have to complain about offensive speech for liability to be imposed. But really the confusion is not surprising, for it reflects a deeper confusion at the heart of the legal and philosophical definition of sexual harassment itself. Since the Supreme Court has continued to skirt the question of why, precisely, harassment is a form of discrimination, the law has evolved in ways that only gender essentialists can find convincing.

At the beginning of the Supreme Court argument in the Texas schoolgirl case, Chief Justice Rehnquist asked a simple question. "Was there some showing of discrimination?" he asked, referring to allegations that the ninth-grade teacher had an affair with his student. "We're assuming harassment and discrimination are synonymous," the girl's lawyer revealingly replied. But Rehnquist was not convinced. "I think the statute says you have to discriminate on the basis of sex; you have to treat students differently." " Here, this student was singled out," the lawyer repeated. "Because of sex?" Rehnquist asked again.

The exchange went to the heart of the analytical weakness of harassment law. It seems odd to think of a teacher's seduction of a ninth-grader as a form of sex discrimination. Statutory rape, certainly. But discrimination on the basis of gender? Not in the way that we ordinarily understand the term. To be sure, Rehnquist's ingenuousness was a little disingenuous. It was Rehnquist who, in 1986, wrote the opinion for the Court in the Meritor case, declaring unequivocally that a bank supervisor who coerced one of his employees into repeatedly having sex was practicing a form of sex discrimination. "Without question, when a supervisor sexually harasses a subordinate because of the subordinate's sex, that supervisor discriminate s ' on the basis of sex," he wrote.

So how, precisely, did the Court come to view erotic fixation as form of gender discrimination? Discrimination usually implies some form of contempt for a class of people being singled out for disadvantageous treatment as a consequence of their shared characteristics. Unwanted advances, by contrast, often involve a man's attraction to a particular woman because of her unique characteristics. Early judicial rulings in the 1970s refused to recognize sexual advances as a form of gender discrimination, on the grounds that they were motivated by personal proclivities rather than by gender animus. Desire may be rampant, but it is not general. A man does not hit on a gender.

Unless, of course, you believe that a woman is a gender; and that there is no significant distinction between a class and a member of a class, and every member of a class stands for the class, and is essentially and definitionally fixed in her identity by her membership in the class. Enter Catharine MacKinnon. It was she who provided the theoretical arguments for considering unwanted advances a form of discrimination. In Sexual Harassment of Working Women, MacKinnon compared two mutually inconsistent, competing theories of liability. The first, which she called the difference approach, asks the following question: "How can you tell that this happened because one is a woman, rather than to a person who just happens to be a woman?" According to MacKinnon, "The basic answer, which presupposes sex comparability, is: a man in her position would not be or was not so treated."

The difference approach is unsatisfying for several reasons. First, it requires courts to engage in an oddly subjunctive thought experiment, a kind of transvestite counterfactual: How would the complainant have been treated if his or her sex were changed? But sexual attraction, again, is particularized, even unique. It is hard to imagine how a judge is supposed to guess whether someone would have been treated differently if she were a different person. And in workplaces where there are few men or few women, the thought experiment is even more arcane: earlier this year, in the Oncale case involving same-sex harassment, the justices were asked to imagine whether Oncale would have been treated differently if he were a woman, even though there were no women on the oil rig where the roustabout worked.

Indeed, the difference approach presents a particular difficulty for those who hold a MacKinnonite analysis of gender. It requires courts to imagine how women would have been treated if they were men, when MacKinnon believes that the genders really cannot be compared in meaningful ways, because women are socially subordinate to men. "To take the differences approach requires temporary suspension of the fact that the sexes are substantively unequal, not just different, a fact which calls into question the appropriateness of presuming equality in order to measure disparity," MacKinnon observes. Although it might be reasonable to compare the sexual harassment of men and women in a world in which the sexes were equal, "in this society, it is not, because of that social inequality which discrimination law exists to eliminate."

MacKinnon explored the difference approach as a litigation strategy, but she preferred an alternative, mutually inconsistent, and far more radical approach to the question of how sexual advances could be considered sex discrimination. She called it the "inequality approach." In her words, " practices which express and reinforce the social inequality of women to men are clear cases of sex-based discrimination in the inequality approach."

This inequality is not merely social or political. It is a much deeper and more indelible disadvantage. Sexuality itself, MacKinnon has famously argued, is the lynchpin of gender inequality. "The male sexual role ... centers on aggressive intrusion on those with less power"--namely, women. "A feminist theory of sexuality," therefore, "locates sexuality within a theory of gender inequality, meaning the social hierarchy of men over women." Anatomy is destiny; or rather, law.

In her less guarded rhetoric, MacKinnon seems to suggest that even apparently consensual heterosexual sex may reinforce the social inequality of women to men, and thus constitute sex discrimination, at least when it takes place between men and women whose social status is not precisely equal. "Is ordinary sexuality, under conditions of gender inequality, to be presumed healthy?" MacKinnon asks in a crucial passage:

What if inequality is built into the social conceptions of male and female sexuality, of masculinity and femininity, of sexiness and heterosexual attractiveness? Incidents of sexual harassment suggest that male sexual desire itself may be aroused by female vulnerability... Examination of sexual harassment, precisely because the episodes appear commonplace, forces one to confront the fact that sexual intercourse normally occurs between economic (as well as physical) unequals. In this context, the apparent legal requirement that violations of women's sexuality appear out of the ordinary before they will be punished helps prevent some from defining the ordinary conditions of their own consent.

MacKinnon's inequality approach, in other words, narrows the definition of consensual sex in a way that dramatically restricts the sphere in which consensual sexual relations can occur. " I t cannot be assumed that if the woman cares about the man, the sex is not coerced," she writes. When pressed to explain whether or not she believes that women can freely consent to sex with more powerful men, MacKinnon's writing becomes even more opaque than usual. "The courts are clear, if others are not, that working women cannot sue for having mutual sexual interactions," she wrote in an elusive piece a few months ago in The New York Times. "Courts also recognize that sexual coercion can be situational: power differences can be a form of force. Sex under conditions of extreme inequality can be coerced and exploitative; sexual compliance can be coerced. But sex cannot be harassment at work by law unless the woman or man really didn't want to have it."

Although MacKinnon held out the possibility, in that article, that sex, under some circumstances, might be consensual, she has argued elsewhere that in a world defined by gender inequality--which is to say, by gender--neither the law nor the accused man nor the woman herself can know a woman's will, because the law defines consent from the man's perspective. Thus, MacKinnon has emphasized "the appearance of rape as similar to normal heterosexual encounters" in a world defined by the male viewpoint. As she rather crisply taught, "Man fucks woman; subject verb object."

This is not only a terrible pessimism about erotic life. It is also legally and historically troublesome. In an essay called "Sex and Guilt," recently published in the University of Virginia Law Review, Anne Coughlin challenges MacKinnon's view that rape law, by requiring women to offer physical resistance to prove their lack of consent, promotes the sexual agency of men at the expense of that of women. MacKinnon's error, Coughlin argues, is to assume that American law ordinarily treats sexual intercourse as legal. In fact, the roots of rape law extend to the puritanical codes of the late seventeenth century, in which consensual sex outside marriage was presumptively illegal.

For most of American history, judges confronted with an allegation of sexual misconduct had to decide whether the encounter involved a rape, for which the man alone was to blame; or fornication or adultery, for which both the man and the woman shared criminal responsibility; or marital intercourse, for which neither participant would be punished. Rather than marking the boundary between sex and rape, Coughlin argues, the woman's lack of consent was what distinguished the man's crime (rape) from the couple's crime (fornication or adultery). Since women had an incentive to claim they had not consented to sex to avoid criminal liability for fornication or adultery, courts had every reason to put the woman on trial, to find out whether or not she was looking for an excuse to avoid guilt.

MacKinnon, ironically, does not assume that nonmarital sex is presumptively legal. By suggesting that sex between unequals cannot ever be truly consensual--and that the inequality in sex is owed to an essential incommensurability of men and women--MacKinnon has helped to resurrect a world that very much resembles the seventeenth-century Puritanism from which modern rape law evolved.

MacKinnon's argument that "sex under conditions of extreme inequality can be coerced and exploitative" masquerades as a vision of radical egalitarianism; but its intellectual and legal consequence is to promote and to preserve a rigid social hierarchy. By pressuring companies to ban even consensual affairs between those on different rungs of the corporate ladder, for fear of liability, MacKinnon has dramatically restricted the possibility for romantic interactions between people of different social classes, and in the process she has reinforced the class distinctions themselves. She is a sexual determinist, and determinists are not the most reliable champions of progress.

MacKinnon thinks ahistorically, in terms of fixities. Specifically, she thinks ahistorically about America, and about its great democratizing (and disorienting) achievement of social mobility. The American workplace has traditionally been a laboratory of social mobility, as men and women from very different backgrounds meet each other and improve their station. From Bill Gates on down, executives often fall in love with employees and, through marriage, transform their social status. Is this love only another instrument of oppression?

Indeed, in an essay on Marxism and feminism, MacKinnon herself recognizes this dynamic. "From a feminist perspective, a woman's class position, whether or not she works for wages, is as much or more set through her relation first to her father, then to her husband," MacKinnon writes. "Through relations with men, women have considerable class mobility, down as well as up." And yet the perverse effect of the MacKinnonite rule--that only romances between social equals are permissible--is to entrench existing class divisions by preventing women and men lower down on the social scale from marrying those above them. The promise of America is embodied in the Whitmanian individualist, unconstrained by class limitations, and free to imagine the world anew and to act on his (or her) imaginings; but MacKinnon's vision is closer to the hierarchical corporatism of Babbitry, in which anxious middle managers are imprisoned by their class limitations, and afraid to accept dinner invitations from those above or below them.

If one were looking for modern societies that most closely approximate MacKinnon's neo-Puritanical vision of gender relations, the likeliest candidates would be the extreme Islamic theocracies, in which the private space for sexual relations beyond the reach of legal and religious regulation is exceedingly narrow. In such societies, marriages are arranged, and class mixing is rare because prospective wives are chosen (by male marriage guardians) from among the groom's social class. Women are secluded and protected from unwanted sexual attention by a strict dress code: the injunction for women to cover all but their hands and face is designed to prevent them from provoking sexual excitement in men.

The constitutions of several Islamic states proclaim that preserving female honor is a duty, and they require the state to prohibit offenses to public decency, including anything that might be objected to as immodest or sexually suggestive. Extramarital sexual relations must be negotiated with legalistic attention. As Geraldine Brooks has described, for example, Shiite Muslims execute a temporary marriage contract called sigheh, which allows unmarried couples to date or to have sex for a mutually agreed upon amount of time, free from the prying of revolutionary zealots. Outside of sigheh, unrelated men and women cannot be alone together, touch each other, or appear uncovered before each other. Women in general are seen not as distinctive individuals but as passive, fungible beings, who must preserve morality at all costs.

Islamic law, of course, secludes and protects women as a means of subordinating them to men, while MacKinnon seeks to seclude and to protect women as a means of liberating them from male subordination. For the Islamic clerics, the power of the patriarchy is prescriptive; for MacKinnon, it is descriptive. But both the Islamic and MacKinnonite visions of gender agree that the private sphere, in which men and women can encounter each other free from regulations, should be very small, and the public sphere, in which sexual relations are intricately regulated, should be very large; and that male domination and female submission are inherent in the very nature of gender identity. Inadvertently evoking the Islamic injunction to preserve female honor, MacKinnon sees sexual harassment laws, like restrictions on pornography, as a form of protection for female dignity.

IV.

In the recently published hearings on her bill to ban pornography in Minneapolis and Indianapolis as a form of sex discrimination, MacKinnon stresses that her arguments "that defined sexual harassment as a form of discrimination on the basis of sex" are based on "the same view of sex discrimination that underlines this ordinance." The root meaning of pornography, MacKinnon noted during the Indianapolis hearings, is "the graphic depiction of whores... In pornography, women are graphically depicted as whores by nature, that is, defined by our status as sexual chattel." It is the depiction of women in subordinate positions, MacKinnon argues, that makes pornography and harassment forms of sex discrimination: " pornography as we define it makes the inequality of the sexes sexual, the way that it makes sexy, the way that it eroticizes putting women in an inferior position."

Courts and citizens have had no trouble understanding that MacKinnon's efforts to regulate pornography as a form of sex discrimination violate the core protections of the First Amendment. Her Minneapolis ordinance defined pornography as "the sexually explicit subordination of women, graphically depicted, whether in pictures or in words, that also includes one or more of the following: women are presented dehumanized as sexual objects, things or commodities ... women are presented in postures of sexual submission ... women are presented as whores by nature." "This is thought control," Judge Frank Easterbrook noted in his opinion in 1985 striking down the Indianapolis ordinance.

Speech treating women in the approved way--in sexual encounters "premised on equality"--is lawful no matter how sexually explicit. Speech treating women in the disapproved way--as submissive in matters sexual or as enjoying humiliation--is unlawful no matter how significant the literary, artistic or political qualities of the work taken as a whole. The state may not ordain preferred viewpoints in this way.

What is surprising is that judges have been so much slower to grasp that MacKinnon's view of sexual harassment law offends the First Amendment for precisely the same reason that her vision of pornography does. The problem is not that the MacKinnonite vision of harassment and pornography permits the banning of speech merely because it is offensive. "Somebody can be real offended by something and if it isn't described by this statute, they can't do anything about it under this statute," MacKinnon noted during the Indianapolis hearings. "They can also not be offended by it, and they won't be the people to bring an action under this law probably... In other words, it doesn't have anything to do with subjective feeling."

Instead, the MacKinnonite efforts to ban harassment and pornography are designed to protect the reputation of women in general, regardless of whether or not any particular woman is offended, and they achieve their purposes by forbidding the verbal denigration of women in some terms but not in others. In this sense, MacKinnon's vision of harassment and pornography very much resembles efforts to restrict group libel, which the Supreme Court now seems to recognize as an unconstitutional attempt to regulate speech at the center of the First Amendment's protections. So the MacKinnonite vision of harassment and pornography violates the central principles of free expression, at least in the context of public employers who are bound by the First Amendment. (Whether the First Amendment is violated when the government pressures private employers to adopt speech codes in order to avoid liability under Title VII is a more complicated question.)

But the MacKinnonite vision is troubling in a deeper sense, which has to do with MacKinnon's conception of legal personhood. She believes that men and women are defined by their sex rather than by their individual personalities and choices. Although she is too much of a post-foundationalist to talk in terms of essences, she is, for all intents and purposes, a gender essentialist. In her view, gender is constitutive of personhood, which is to say that all men are oppressors and all women are victims, regardless of whether or not they actually feel oppressive or victimized.

To abstract away from gender, and to construct a universalist theory of equality, MacKinnon believes, is a concession to the patriarchy. It is remarkable, in retrospect, that MacKinnon has largely succeeded in embedding her vision of gender essentialism into American law, for her vision represents a direct assault on the liberal vision of gender-blindness. And that vision is the foundation on which the Supreme Court's jurisprudence of gender discrimination was originally constructed. It was championed by Ruth Bader Ginsburg as an advocate in the 1970s, and as a Justice in the Virginia Military Institute case in 1996. "Generalizations about the way women are,' estimates of what is appropriate for most women, no longer justify denying opportunity to women whose talent and capacity place them outside the average description," Justice Ginsburg wrote, in a rebuke to Virginia's essentialist claim that men learn best in adversity and women in harmony.

Ginsburg has not yet acknowledged, in her sexual harassment opinions, the extent to which the MacKinnonite vision of gender equality is a direct repudiation of her own. And so, as the Court prepares to hand down three important harassment decisions, it is worth asking what a gender-blind understanding of harassment law would look like. In its recent opinions, the Supreme Court has intoned rather plaintively that harassment must be "because of sex" in order to be actionable. "The critical issue, Title VII's text indicates, is whether members of one sex are exposed to disadvantageous terms or conditions of employment to which members of the other sex are not exposed, " Justice Ginsburg wrote in 1993. But the Court can't make the analytical problem go away merely by incarnating the magic words "because of sex." How, precisely, can the law be reformed to focus on sex discrimination, as opposed to sexual expression?

In a widely publicized proposal for reform, Vicki Schultz argued recently in the Yale Law Journal that sexual harassment law has erred in emphasizing sexual advances, rather than gender-based hostility, as the core of an actionable offense. By accepting the MacKinnonite notion that sexuality is the lynchpin of inequality, Schultz argues, courts have "neglected equally pernicious, nonsexual forms of gender-based misconduct in the workplace-- particularly conduct that denigrates women's competence and thereby preserves work along gendered lines." Rather than defining harassment as "unwelcome sexual advances, requests for sexual favors, and other verbal or physical conduct of a sexual nature," as the EEOC guidelines do, Schultz would prohibit "conduct the purpose or effect of which is to make it more difficult for someone to do their job based on sex."

There is much to admire in Schultz's effort to reorient harassment law away from its misguided focus on sexual expression; and Schultz has been cast as an advocate of narrowing the scope of harassment liability. Alas, this is a misrepresentation of her proposal. In fact, Schultz argues the opposite. "We have not conceptualized the problem in sufficiently broad terms," she writes. Schultz would expand the scope of admissible evidence in harassment cases, stressing that conduct that demeans the person as a professional and makes it more difficult for women to do the job may take the form of sexual advances or other conduct of a sexual nature, but need not do so. Nonsexual hazing, work sabotage, or failure to provide informational training would count toward establishing a hostile work environment for women. "In the context of a workplace with longstanding inequality," she writes, "a potentially stray remark' about women ... may assume heightened causal significance."

In Schultz's view, when a supervisor demeans the professional abilities of a subordinate, the company should be automatically liable, rather than being able to take refuge in an effective anti-harassment policy. This would create an incentive for companies to police the speech of their supervisors far more dramatically than current law requires. Schultz also argues that a harassment complaint, in which a woman alleges that her competence has been undermined, should open the window on a broad inquiry into the treatment of women in the corporate workplace more generally. The criteria for hiring, assignment, and promotion would all come under scrutiny. The only way that a company could be confident that it could avoid liability for the remarks of its employees, Schultz suggests, would be to achieve some kind of proportional representation of women at all levels of the corporate hierarchy.

Schultz's ultimate goal, gender proportionality and gender parity, is a far more ambitious endeavor even than MacKinnon's effort to eliminate pornography and to narrow the scope of sexual expression in the workplace. Schultz might also suppress more speech at the core of the First Amendment than MacKinnon. Imagine, for example, a San Francisco firefighter who is appalled by a consent decree in which the Fire Department agrees to hire women in proportion to their numbers in the population as a whole, even though the applicant pool of women who want to be firefighters is less than 20 percent. What if the angry male firefighter, questioning the overall ability of women hired under this scheme, said something like this: "This new group of women that we've been forced to hire just isn't up to the job." Under Schultz's regime, such comments might expose the Fire Department to ruinous damages, and would increase the legal pressures to achieve gender parity, even though the comment is a fine example of core political speech.

Schultz's attempt to refocus harassment law on gender discrimination rather than sexual expression deserves respect; but she, like MacKinnon, wants to use harassment law as an engine for the wholesale restructuring of the American workplace. In both Schultz's and MacKinnon's accounts, workers cannot be protected from sexualized or demeaning comments unless the workplace itself is transformed. This is a long way from the classically liberal vision of Ginsburg, who sought to guarantee equal access for individual women to choose professions on the basis of their individual talents and preferences, instead of using sex discrimination law to generalize about the appropriate choices and desires of women as a whole.

V.

What, then, would harassment law look like if it were reformulated in gender-blind terms? Title VII refers to discrimination in the terms and the conditions of employment, and this would certainly include firing a woman or a group of women because of their sex--that is, because the employer does not like women in general, or likes them too much and wants to protect them from the indignities of the workplace. The existing law of quid pro quo harassment, in other words, could largely be preserved in a gender-blind world.

It is arguable, as courts held in the 1970s, that a supervisor who threatens to fire a woman because he is attracted to her, and is uncomfortable working with her after he has been rebuffed, is discriminating on the basis of her rejection of his sexual advance, and not on the basis of her sex. It would clearly be sex discrimination, however, if all women in the workplace were forced to endure the ordeal of being threatened with losing their jobs unless they succumbed to sexual advances, because the terms and the conditions of employment would then be different for women than for men. And the analysis is at least arguably similar if a supervisor only threatens to fire that subset of women that he is attracted to, even if it includes only a single woman.

Hostile environment harassment is another matter. A concerted campaign to drive all women, or some women, from the workplace by hassling, teasing, and abusing them would change the terms and the conditions of employment in a way that should be actionable under Title VII; but the law cannot forbid comments that only some women might find offensive without generalizing about the sensibilities of women in a way that offends the gender-blind construction of the idea of equality. For this reason, several commentators have suggested refining the hostile environment test in a way that imposes liability only on speech whose purpose is to annoy a woman or a group of women in particular, rather than on speech that supposedly has a disparate impact on women in general. Thus Eugene Volokh argues that speech, to qualify as harassment, must be directed at a particular listener. A group called Feminists for Free Expression would require "a pattern of targeted and/or intentional verbal abuse." Cathy Young suggests that "hostile environment" claims should be based on actual hostility; she would narrow the EEOC formula to conduct that has the "purpose and effect" (rather than "purpose or effect") of " unreasonably interfering with an individual's work performance or creating an intimidating, hostile, or offensive working environment."

Where does that leave what Mark Hager calls "mashing," or sexual overtures, as opposed to "bashing," or derogatory and demeaning treatment of women? In a provocative article in the Connecticut Law Review, Hager argues that sexual overtures among nonsupervisory coworkers should be regulated through tort law rather than through sex discrimination law. This represents a major shift in legal paradigm, and a welcome one. The restoration of liberal sanity to the discussion of sexual harassment and the law may indeed require that the sex discrimination model be exchanged for the tort model.

Wrongful discrimination implies some kind of contempt for a category of people, Hager argues, but harassing sexual overtures are a different matter. They mostly represent a combination of erotic fixation and insufficient moral restraint. In addition to offending common sense, the construal of erotic fixation as "discrimination" obscures the moral harm involved when sexual advances in the workplace really are egregious and really do go too far. "The harasser's wrong consists of contempt for the dignity and autonomy of another human being," Hager wisely observes, "denigrating her intrinsic worth and treating her as a mere means of gratification. The victim's harm consists of having her dignity disparaged and her autonomy impeded. Hence, harassment is inherently tortious. Its essence is of non-consensual personal invasion." There is no inconsistency, in other words, between President Clinton's abstract support for women's rights and his behavior as an appalling cad.

Genuinely malicious advances, Hager argues, could be punished under an existing branch of tort law that compensates victims who have suffered " intentional infliction of emotional distress." Employers would be liable for a pattern of directed harassment by a supervisor only if they manifested " deliberate indifference" to the offensive conduct. This sort of legal standard would dismantle the inquisitorial machine that the sex discrimination version of sexual harassment law has constructed. Courts would be encouraged to weigh the harms of harassment against the burdens of preventing them. As for harassment by coworkers, Hager proposes that employers should only be liable when they are deliberately indifferent to a pattern of targeted abuse by more than one worker, joined by at least one superior.

To be actionable as a tort, a sexual advance would have to be "outrageous" and the distress that it provokes would have to be "severe," which means that many indignities that are now illegal under discrimination law would no longer be legally regulated. This represents a shrinkage in the scope of sex harassment law; but only the tort paradigm, I think, can vindicate the promise of a gender-blind ideal that treats women and men as autonomous individuals, rather than as prisoners of gender stereotypes and sex-based generalizations. The tort model would encourage individual responsibility by placing the blame for inappropriate conduct on the perpetrator, rather than the employer, and would increase the incentives for women to protect their own dignity rather than passively relying on employers to protect it for them. It would empower individual women and men by allowing them to decide whether or not offensive comments interfered with their ability to do the job, rather than using them as helpless symbols in the gender wars as a whole. It would end the unconscionable invasions of privacy that are now a regular feature of the workplace, in which employers are forced to engage in hideous and paternalistic investigations of consensual affairs between employees and supervisors, for fear of liability. And given the extensive evidence that men and women perceive the same situations differently, a tort paradigm would increase the possibility of understanding between the sexes by encouraging women or men to complain when they feel offended.

If the costs of the current harassment regime were limited to disrupting the careers of a small group of unfortunate employees who have failed to adjust quickly enough to the transformation in social norms that govern interactions between men and women at work, then it would be easy enough for society to bear them. All of us should try to behave like ladies and gentlemen in every sphere of our lives; and at this stage in our gender politics, anyone who is reckless enough to treat colleagues, employees, students, and fellow citizens with less than Victorian respect deserves, in some measure, the ignominy that will follow. But the real costs of harassment law are not simply the careers that it disrupts and the workplaces that it fills with confusion and fear. The costs are, more generally, to the liberal dispensation in America.

Constructing legal rules based on stereotypes about the essential natures of men and women is the central evil that Ginsburg's ideal of gender- blindness was designed to banish. The law, Ginsburg rightly insisted, can never provide a complete description of the person. The legal person is an abstraction. In this act of abstraction, however, lies the possibility of neutrality. The denuding of the individual of his or her particular traits or characteristics, so as to arrive at a standpoint from which all may be viewed impartially, is not the debasement of an identity but the construction of an identity, which is the identity of citizenship. It is especially important now, in this golden age of difference in America, in which universalism is on the run and perspectivism is a reality to which politicians and professors gladly pander, that we appreciate that equality is a positing of sameness. The United States today is a proudly splintered polity; but the sameness that liberalism stubbornly imputes to all men and all women is a sublime sameness.

For the law, certainly, it is the very condition of objectivity. But then, these are not the best of times for objectivity either. In the gender-blind view, however, objectivity is not an illusion or a disguise for power, and equality is something that all men and women have inalienably in common. It is a lowest common denominator that is also a highest common denominator. In our madly sexualized culture, to be sure, the ideal of gender-blindness may seem especially elusive. But if it is not possible to look beyond the gender (and the race and the class) of an individual, it is not possible to deliberate and meet as citizens on equal terms.

Gender-blindness was a tremendous legal and philosophical achievement, perhaps the greatest achievement of legal liberalism since the sexual revolution. Yet no sooner was it enshrined in law than it suffered a radical assault from the gender essentialism that threatens everything it stands for. Two decades ago, when women were beginning to enter the workplace in meaningful numbers, MacKinnonite generalizations about the subordination of women to men may have had more of a foundation in reality, and they may have served as a useful spur. But those generalizations are neither accurate nor harmless any longer; and philosophically they were always a betrayal of liberalism. Today the victims of patriarchy have been joined by the victims of essentialism. This has to stop. We must make ourselves blinder if we are to see more.

Why Conservatives Are Defending Al Franken

By Jeet Heer

The Death Of A Cartoonist

By Cathy Young

In Defense of Gender-Blindness

By Jeffrey Rosen

Page 1