If you have been feeling closely scrutinized at work lately, you are not alone. The most recent study of electronic monitoring in the workplace, conducted earlier this year by the American Management Association, found that more than half of the large U.S. firms surveyed monitor the Internet activity of their employees. Two-thirds monitor e-mail messages, computer files, or telephone conversations, up from 35 percent three years ago. The companies gave several reasons for these intrusions, including concerns about employee performance and productivity. But there was another, perhaps less obvious, reason: sexual harassment. "Employees who are unwittingly exposed to offensive graphic material on colleagues' computer screens may charge a hostile workplace environment," the AMA survey notes. They may do so because the Supreme Court's definition of sexual harassment is vague. It includes not only sexual extortion--this is known as "quid pro quo" harassment, as in, " Sleep with me or you're fired"--but also a more ambiguous category known as " hostile environment" harassment. It's hard to know in advance what kind of environment a jury might consider hostile, and employers don't want to find out the hard way. So they protect themselves by monitoring, and sometimes punishing, e-mailing and Web-browsing that employees thought was private.
By itself, of course, a single sexually explicit e-mail message or practical joke isn't sufficiently "severe or pervasive" to render the environment hostile, according to the Supreme Court. But a pattern of comments or jokes from different employees can create a hostile environment. As a result, even the most liberal employers are adopting draconian monitoring policies. "These systems may not be used to create, transmit, or receive any offensive or disruptive messages," reads the Internet policy recently adopted by The Village Voice. "Among those that are considered offensive are any messages that contain sexual implications or jokes; messages that comment offensively on a person's race, sex, age, sexual orientation, national origin, disability, or religion; messages that defame others; and messages that invade a person's privacy.'' In an eloquent denunciation of the policy, columnist Nat Hentoff asked, "As for that last phrase, doesn't the Voice now do precisely that?"
The well-intentioned campaign against sexual harassment has threatened privacy offline as well. On several prominent occasions, harassment law has been used to scrutinize consensual sexual behavior. Merely by accusing Bill Clinton of an unwanted advance, Paula Jones was able to violate not only his privacy but also that of Monica Lewinsky, who was forced to describe her own consensual sexual activities under oath. Lewinsky's biography, Monica's Story, reports that "Monica felt a brooding sense of outrage that Paula Jones's freedom to sue the President for money should take precedence over her right to privacy."
Something has gone wrong in the law of sexual harassment. A jurisprudence originally designed to protect privacy and dignity is inadvertently invading privacy and dignity. There is, however, a solution that would retain the important benefits of harassment law while avoiding its debilitating side effects. It lies in abandoning the hostile-environment test and returning to the text of the Civil Rights Act of 1964, which pledges that Americans should not have decisions about their livelihood turn on their gender. Any speech or conduct that affects an employee's professional status because of sex is, and clearly should be, forbidden by law. But speech and conduct that some employees may find offensive but that has no tangible consequences for employment should be recognized for what it is: not a form of gender discrimination but an invasion of the offended person's privacy. By reconceiving harassment, the law could protect people's legitimate rights without imposing a cure as dangerous as the disease.
How can an offensive comment or joke or glance in the workplace invade someone's privacy? In a book from 1963 called Behavior in Public Places, the sociologist Erving Goffman described the subtle and complicated range of expressions, glances, and signals we use either to make ourselves accessible to other people in face-to-face interactions or to maintain boundaries of reserve. Based in part on studies of etiquette manuals, Goffman argued that all of us rely instinctively on a richly calibrated range of "involvement shields" that regulate our encounters with strangers, acquaintances, and intimate friends. In different social settings, we maintain boundaries of reticence that other people are forbidden to cross without mutually negotiated consent.
In America today, sexual-harassment codes have become an imperfect substitute for etiquette manuals in regulating face-to-face interactions. Consider the following list of potential sexual-harassment infractions at a typical state university:
Making sexual innuendoes; turning work discussions to sexual topics; telling sexual jokes or stories; asking about sexual preferences or history; asking personal questions about a person's social or sexual life; making sexual comments about a person's clothing, body, or looks; repeatedly asking out a person who is not interested; telling lies or spreading rumors about a person's sex life. Looking a person up and down (elevator eyes); blocking a person's path; displaying sexual or derogatory comments about men/women on coffee mugs, clothing, and so on; making facial expressions such as winking, throwing kisses; making sexual gestures with hands and/or body movements; giving letters, gifts, and/or materials of a sexual nature; invading a person's body space by standing closer than appropriate or necessary.
Many of the liberties on this remarkable list, when considered in isolation, are better conceived as invasions of privacy than as examples of gender discrimination. A single case of "elevator eyes," for example, in which one person lasciviously looks another up and down, is hard to think of as a form of discrimination that limits a woman's professional opportunities on the basis of gender. But it is by no means absurd to conceive of leering as an invasion of privacy. Indeed, Goffman emphasizes the particular difficulty of maintaining civil inattention in small spaces such as railway cars or, as it happens, elevators. He quotes a charming etiquette guide from the 1950s by Cornelia Otis Skinner, called Where to Look, which identifies the modern elevator as a particularly trying example of a "where to look situation," in which "any mutual exchange of glance on the part of the occupants would add almost a touch of lewdness to such already over-cozy sardine formation."
To be looked up and down by strangers can be an indignity, although in most cases a minor one. To be looked up and down by your boss may be a more serious indignity. The problem is that American law has traditionally been reluctant to punish offenses against individual honor or dignity: this is why the early architects of harassment law had to define workplace offenses as gender discrimination. "Our system," Louis Brandeis and Samuel D. Warren wrote wistfully in 1890, in their famous article on the right to privacy, " does not afford a remedy even for mental suffering which results from mere contumely and insult, from an intentional and unwarranted violation of the 'honor' of another."
Other legal systems, whose privacy protections are less rooted in ideas of private property, have less trouble describing the injury that results when people are observed against their will. Jewish law, for example, has developed a remarkable doctrine around the concept of hezzek re'iyyah, which means "the injury caused by seeing." Jewish law protects neighbors not only from unwanted observation but also from the possibility of being observed. Thus, if your neighbor constructs a window that overlooks your home or courtyard, you are entitled to an injunction that not only prohibits him from observing you through the window but also orders the window to be removed. From its earliest days, Jewish law recognized that it is not only surveillance itself but also uncertainty about whether or not we are being observed that inhibits us from speaking and acting freely in private places. " Even the smallest intrusion by the unwanted gaze into private space causes damage," medieval authorities wrote, "because the injury caused by seeing cannot be measured."
This notion of the unwanted gaze helps us understand why some breaches of social boundaries that are currently considered sexual harassment are better defined as invasions of privacy. By treating a woman merely as a sex object and allowing a single aspect of her identity--her appearance--to eclipse all other aspects, an unwanted gaze wrenches her out of context in the same way Monica Lewinsky was wrenched out of context by the exposure of her bookstore receipts and undeleted e-mail messages. In the 1970s, feminist film critics wrote about the objectifying quality of the "male gaze" in popular movies, which, in their view, reduced women to eroticized body parts fit for male ogling. These critics are correct when they describe the indignity of being looked at in a way that substitutes a part of a woman's body for the whole of her personality. But that indignity is more precisely described not as gender discrimination but as an invasion of privacy.
In arguing that certain sexual advances or gazes may invade privacy, I don't mean to deny that they might also be viewed as a form of gender discrimination. In the workplace, even relatively trivial forms of sexualization may impede women's progress and make it harder, in subtle ways, for them to function as professionals. (An offense against dignity can also become an offense against equality when it occurs in a professional context.) But, as Lewinsky's protests show, views about consensual sexual behavior and expression are tremendously varied, and forcing people to justify their own sexual choices in public, as the hostile-environment test often does, can be appalling. Treating this behavior as an invasion of privacy rather than as gender discrimination would protect private spaces for negotiating legitimately different views of the good life, freeing people from the constant burden of justifying their differences.
How did offenses against personal dignity in the workplace become conceived as sex discrimination rather than as invasions of privacy? The paradox arose in a strain of feminist theory that stressed the connection between male desire and male dominance. In Sexual Harassment of Working Women: A Case of Sex Discrimination, published in 1979, Catharine MacKinnon insisted that some sexual expression might be viewed as inherently discriminatory, because heterosexual relations were the means by which male domination and female subordination were enforced. The great innovation of MacKinnon's book was to identify a new category of sexual harassment that didn't require evidence of physical contact, explicit threats, or tangible job consequences. She described this as situations where sexualized comments or displays in the workplace made the "work environment unbearable."
A year after the publication of MacKinnon's book, the Equal Employment Opportunity Commission issued guidelines defining sexual harassment to include not only quid pro quo threats but also "verbal or physical conduct of a sexual nature" that has "the purpose or effect of unreasonably interfering with an individual's work performance or creating an intimidating, hostile, or offensive working environment." And in 1986, in the landmark Meritor case, the Supreme Court unanimously endorsed the EEOC's hostile-environment test. The case involved a bank teller whose supervisor repeatedly pressured her for sex and eventually raped her. The Supreme Court could have said the supervisor's constant sexual demands clearly changed the terms and conditions of the teller's employment because of sex--an obvious violation of the text of Title VII of the 1964 Civil Rights Act. Instead, Chief Justice William Rehnquist departed from the text of Title VII, holding that the supervisor had discriminated by creating a hostile and offensive working environment. In cases like Roe v. Wade, in which the Supreme Court declared a constitutional right to privacy, the justices used privacy as a clumsy metaphor for rights of reproductive choice that the Court would later reconceive as an aspect of gender equality. By contrast, in Meritor, the justices said it was a form of gender discrimination to commit offenses against personal dignity in the workplace that are better conceived as invasions of privacy.
The hostile-environment test transformed the workplace by blurring the boundaries between the public and private spheres. By allowing women (or men) to complain about any sexually oriented speech or conduct they found hostile or offensive, the new test allowed aggrieved coworkers to object to overheard jokes, e-mail messages, suggestive pictures, or even consensual flirtation on the part of their colleagues, even if the conduct wasn't considered offensive by the men and women directly involved. And, unlike quid pro quo harassment, a hostile environment doesn't require proof of economic or even psychological injury. While under quid pro quo harassment only women who suffered tangible harm could sue, under the hostile-environment theory, the listener's perception of the "hostility" or "offensiveness" of the speech became the test of legal harm. This ambiguity is why prudent employers began to monitor private speech and conduct to avoid any incident that might conceivably give offense.
What would the world look like if the hostile-environment test were abandoned? Quid pro quo harassment would still be prohibited. Title VII makes it illegal "to discharge any individual, or otherwise to discriminate against any individual with respect to his compensation, terms, conditions, or privileges of employment, because of such individual's ... sex." The courts long ago abandoned the idea that there was any difference under Title VII between discrimination based on gender and discrimination based on sex. And there is no doubt that implicit or explicit threats to fire someone for failure to submit to sexual demands is a form of discrimination in the terms of employment "because of ... sex."
Similarly, pervasive gender-based animosity, or differential treatment of women based on stereotypical views about their abilities, would clearly change the terms of employment and thus constitute discrimination based on gender. The often overlooked second sentence of the relevant section of Title VII makes it illegal to "limit, segregate, or classify" employees in a way that tends to "deprive any individual employee of employment opportunities or otherwise adversely affect his or her status as an employee because of ... sex." This clearly covers the gauntlets, seemingly more common in blue-collar professions, in which women endure a daily pattern of sexist taunts and abuse. It also covers subtler manifestations of sexist treatment, in which women's professional opportunities are limited and their professional abilities belittled because of the discriminatory attitudes of male employers.
That leaves only one category of sexual harassment that might be less closely regulated if the hostile-environment test were abandoned: sexual speech and conduct that can't plausibly be characterized as sex discrimination. This includes certain unwanted advances, suggestive looks and gestures, sexual joking and teasing, and the display of sexually explicit material--in cases where these behaviors carry no threat of retaliation, don't change the terms of employment, and don't deprive an individual of employment opportunities because of sex but are nevertheless potentially offensive. There is clearly room for vigorous debate about how much speech and conduct this category includes. Sexual harassment is highly dependent on context: a series of unwanted requests for dates from a supervisor might change the terms of employment in a discriminatory way, while a series of unwanted requests for dates from a coworker might not. Similarly, sexist taunts from a supervisor might affect the professional status of a female employee in a way that vulgar jokes about that same woman by two male colleagues might not. But if the hostile-environment test were abandoned, this debate about which speech and conduct changes the terms of employment would be guided by the text of Title VII itself. And focusing on whether someone was discriminated against rather than whether someone was offended would make employers less likely to ban private communications out of fear that a third party might take offense.
A new generation of feminist scholars is beginning to recognize the analytical awkwardness of conceiving hostile-environment sexual harassment exclusively as a form of gender discrimination. In a powerful article last year in The Georgetown Law Journal, Rosa Ehrenreich argued that sexual harassment should be understood as an injury to dignity. The harms that hostile-environment sexual harassment inflicts, Ehrenreich argued, are indignity and humiliation, and they therefore are better understood not simply as gender discrimination but as something like what the law calls a tort--a civil injury inflicted by one individual upon another.
Tort law is the area of our legal system that gives people remedies against those who have transgressed social norms in especially outrageous ways. But in harassment cases that don't involve professional consequences, what is the social norm being transgressed? It is, in my view, privacy: in particular, the ability to control the conditions under which we make ourselves socially accessible to others. In these cases, privacy law is better equipped than gender-discrimination law to distinguish truly egregious violations, which should be illegal, from merely offensive behavior, which should not. The most intentionally hurtful behavior can be regulated by the intentional torts: battery, assault, false imprisonment, and outrage. But many incidents of sexual harassment are not clearly intended to humiliate, and they look more like boorish and insensitive advances pressed despite ambiguous warnings. In this gray area, invasion-of-privacy law is especially clarifying. One of the torts of invasion of privacy forbids intrusions on seclusion: "One who intentionally intrudes, physically or otherwise, upon the solitude or seclusion of another or his private affairs or concerns is subject to liability to the other for invasion of his privacy, if the intrusion would be highly offensive to a reasonable person."
The intrusion-on-seclusion tort effectively identifies the injury that results from unwanted advances and other trespasses on personal honor. Courts in intrusion-on-seclusion cases, like courts in hostile-environment harassment cases, have not insisted on proof of serious psychological injury or mental distress. This makes little sense if harassment is conceived of as a workplace injury, which usually requires proof of adverse professional consequences, but it makes perfect sense if harassment is conceived of as an offense against dignity. Alabama is not generally regarded as being at the vanguard of civil rights; but, in a far-reaching decision, the Alabama Supreme Court held in 1983 that a series of repeated sexual demands by a supervisor might constitute an intrusion on seclusion. Rejecting the argument that such an intrusion has to be a physical invasion of a private space, like the home, the court held that, even in the workplace, there is an "emotional sanctum" that is entitled to the same expectations of privacy as our physical environment.
Using invasion-of-privacy law, rather than the hostile-environment test, to regulate offensive speech that has no professional consequences would retain the benefits of current harassment law while avoiding its inadvertent side effects. It would encourage companies to intervene in serious cases, but it would reduce the pressure they now feel to monitor private speech and conduct that wouldn't ordinarily be brought to their attention. That's partly because courts have been more reluctant to hold businesses (as opposed to the offending individuals) liable in privacy suits than in hostile-environment suits. In Alabama, employers are only liable for intentional torts committed by their employees if they know about the conduct, know it constitutes an illegal invasion of privacy, and fail to stop it.
The intrusion-on-seclusion tort has also proved more effective than the hostile-environment test in distinguishing serious violations of privacy from unfortunate misunderstandings. The requirement that the indignity must cause " outrage or cause mental suffering, shame, or humiliation to a person of ordinary sensibilities" sets the bar high enough so that relatively trivial indignities are no longer actionable. And invasion-of-privacy law focuses on speech targeted at a particular woman that has the purpose or effect of humiliating her--as opposed to speech that inadvertently offends third parties.
Most important of all, using tort law, rather than the hostile-environment test, to regulate offensive but nondiscriminatory speech and conduct would make it easier for employers to reconstruct private spaces in which employees could express themselves without fear of being observed. As Goffman has noted, individuals, like actors in a theater, need backstage areas where they can take off their public masks, collect themselves, and relieve the tensions that are an inevitable part of public performance. By definition, a backstage area is a place of retreat from social norms--a place where workers can be lustful, sloppy, indiscreet, or playful, form intimate bonds, and indulge in behavior that would be inappropriate if practiced in more formal areas of the workplace. In a company where men vastly outnumber women, a place where men relax among themselves--making vulgar jokes, talking about women, and engaging in the ordinary rituals of male bonding--may be perceived by women as threatening and discriminatory. When women are in the majority, the backstage area may seem inhospitable to men. But, although one sex's backstage behavior may be another's hostile environment, social remedies are better than legal ones when the offense is not severe and the terms of employment are not clearly affected. The law is a blunderbuss rather than a scalpel, and the effort to provide legal remedies for relatively minor offenses, as the hostile-environment test now does, may inadvertently destroy the backstage enclaves that Goffman shows to be so important.
Consider how the law of privacy, as opposed to the law of gender discrimination, would have responded to the case of Bill Clinton and Paula Jones. On public streets and in the workplace, we all rightly expect we won't be bothered by sexual overtures unless we've clearly indicated, by our words, gestures, and glances, that the overtures will be reciprocated. Certain spaces--hotel bars, cocktail lounges, even some supermarkets--have traditionally been seen as "open places" where sexual advances are considered permissible rather than an invasion of privacy. The hotel room to which Jones allegedly repaired after allegedly being told by a state trooper that she made Governor Clinton's "knees knock" might be seen in the same vein. But, even in a hotel room, a sexual advance can invade the recipient's expectations of privacy when it is disproportionate to the kind of attention she has invited. Jones may have gone to the hotel room prepared to flirt with Clinton, but by all indications she was alarmed and discomfited when, without warning, he allegedly exposed himself to her. If Clinton did what Jones accused him of doing, he insulted her and invaded her privacy.
Furthermore, if Jones had sued Clinton for invasion of privacy rather than sex discrimination, the collateral invasions of the privacy of innocent third parties, such as Lewinsky, might have largely been avoided. In a hypothetical invasion-of-privacy suit, the discovery process would have focused on Clinton's treatment of Jones rather than his treatment of all the women to whom he had made previous sexual advances. Because the invasion-of-privacy tort, unlike the law of sexual harassment, focuses on particular encounters rather than patterns of behavior, a sensitive judge would have more discretion to exclude invasive evidence without making it impossible for Jones to argue her case. And, whether or not Jones ultimately won, an invasion-of-privacy suit would have focused on the indignity she suffered, which was obvious, rather than its effect on her employment status, which was negligible.
There are many acceptable responses to inappropriate speech and conduct in the workplace, from talking quietly to the offender to applying formal discipline. Indeed, in a pluralistic society, different employers should be allowed to protect privacy in different ways: the military should be free to monitor e-mail and forbid fraternization; universities might regulate conduct between professors and students but refuse to regulate speech under any circumstances; and different corporations could strike different balances, depending on the nature of their product and the kind of employees they hope to attract. This flexibility, in which privacy is enforced primarily through social disapproval backed by voluntary choices made by employers, is something today's sexual-harassment law refuses to allow.
Anyone who has joined the workforce in the past generation has benefited from the heightened sensitivity to sexual harassment, sensitivity that has helped men and women work together with greater civility, equality, and mutual respect. The integration of women into the workforce is the most important social change of the postwar era; and the transformation of social norms so that people who inevitably find themselves attracted to each other can interact as professionals is no small achievement. And, if the strong medicine of harassment law has restored a Victorian caution to the way ladies and gentlemen interact at work, that is hardly a bad thing.
But, while the hostile-environment test may have been necessary when white- collar workplaces were more sexually segregated and less sexually sophisticated than they are today, its inadvertent side effects now threaten the boundaries between the public and private spheres. It is wrong, in a liberal society, that sexual expression without tangible employment consequences--overheard jokes, private e-mail, and consensual affairs between colleagues--should be monitored by employers and punished by the state. When speech or conduct changes the conditions of employment, the answer is anti- discrimination law. But, in less extreme cases, invasion-of-privacy law is better suited than the hostile environment test to distinguish speech that is merely offensive from speech invasive enough to be illegal. By returning to the text of the 1964 Civil Rights Act, the courts could help rebuild private enclaves in public places for people to relax, to reveal different sides of themselves in different contexts, to misjudge each other--in short, to be human.