Cyber Sexual Harassment: Comments on Citron and Franks

This post is part of an eleven-part series entitled Cyber Civil Rights.  Click here for a PDF version of the entire Cyber Civil Rights series.  Click here for a PDF version of this post.

By Nancy Ehrenreich

Introduction

One of the most interesting challenges for lawyers and law professors, of course, is creating new language and concepts to capture new injuries as they arise.  Naturally, the impulse is to explore existing doctrine for appropriate analogies, and there are many interesting possibilities for the injury of online sexual harassment, as Professors Citron’s and Franks’ scholarship reveals.

The obvious place to start is certainly sexual harassment law, as many participants at this symposium noted.[1]  I agree with Professor Citron that there are many historical similarities between discussions of cyber harassment today and the initial debates about sexual harassment in the workplace that occurred during the 1980s.  Harassment is being trivialized now in quite similar ways to how it was then, and the arguments for seeing such behavior as non-actionable private crudeness rather than civil rights violations are familiar as well.  Courts and commentators in those early days routinely dismissed harassment at work as harmless flirting, and would-be plaintiffs were often exhorted to seek work elsewhere if they didn’t like the sexually charged atmosphere that some workplaces “happened” to have.  So, the recycling of such attitudes into arguments such as “if you don’t like the atmosphere, stay off the website” is certainly not surprising.

In the sections that follow, I’ll comment upon several aspects of the problem of cyber sexual harassment and the presenters’ thoughts on how to solve it, including: (1) the overstated benefits of Internet freedom; (2) the nature of the harm of cyber sexual harassment and possible solutions; (3) conflicting liberties, the pornography parallel, and access issues; and (4) the real danger of web regulation: censorship of political dissent.

I. The Overstated Benefits of Internet Freedom

What I appreciate most about Professor Citron’s Cyber Civil Rights is the way that it gives the lie to the myth of neutrality on the net—the idea that free speech exists currently on the Internet and must be preserved at all costs.  Professor Franks picks up on that idea as well, with her concept of “cyberspace idealism”: “the view of cyberspace as a utopian realm of the mind where all can participate equally, free from social, historical, and physical restraints.”  The online dynamics identified by both of these commentators starkly reveal a truth as old as the American Legal Realists: Liberties often conflict; one person’s freedom often compromises another’s.  In this context, this means that granting unfettered freedom to harassers silences those whom they harass, foreclosing access to the net for those individuals and depriving them of an incredibly valuable societal resource.  “If she wants to work, she can just put up with the behavior,” was the refrain in the 1980s—until activists revealed it as an argument in favor of preserving male privilege.  To say that the web is a sacrosanct arena that can’t be regulated is to sound the same refrain.

But I’d like to push the challenge to net naïveté a bit further—by questioning the popular image of the Internet (implicitly endorsed at times by Professor Franks) as a place where individuals can “escape the physical constraints imposed on [their] identity, and . . . exert control over [their] representation.”[3]  As Professor Franks rightly notes, such freedom (if it existed) would be especially beneficial to marginalized groups, whose bodies are so often deployed as metonyms for their otherness.  And it seems plausible that individuals might find it liberating in some sense to be able to perform a different identity on the web than they are able to perform in “real life”—whether it’s women masquerading as men or men as women; people of color as white or vice versa; sexual minorities as straight or the other way around.

But the benefits of such identity fluidity on the net may be more illusory than real.  And the notion that one can somehow escape subordination when in the virtual world sounds a familiar theme that has repeatedly been proven wrong: the naïve faith in technology’s ability to overcome social realities of power and privilege. (Recall, for example, the early love affair that feminists had with reproductive technologies or the more recent refrain that (as yet uninvented) technological advances will solve the problem of global warming.)  The unfortunate fact is that, because communication and language are socially constructed, categories of meaning, of similarity and difference, of value and lack of value, transcend the material context, continuing to structure behavior and limit possibilities in virtual space.

It may be beneficial, for example, for a man to pretend to be a woman while on the web, and thereby experiment with expressing his feminine side without risking negative repercussions.  But, wouldn’t it be more beneficial if he could openly express his femininity without disguising himself?  Is it liberating for a woman to be able to masquerade as a man in order to be taken more seriously on a blog? Is that any different than, say, a gay person being closeted at work?

The Internet can’t escape culture any more than any other technology can.  While it might be a more subversive place if, for example, “handles” were incapable of being sex-identified (through randomly assigning numbers or the like), even then dominant ontologies would prevail.  Because dominant classes operate as unmarked categories (“man” is the unmarked generic word for humans, etc), the default assumption about any particular individual communicating on the web would likely be that he or she occupied one or more dominant positions (was male, white, etc.).  Studies have shown that such defaulting to the unmarked category is common in human communication—for example, people often assume an unidentified actor is white, male, and heterosexual.  Thus, socially constructed categories would likely perpetuate existing hierarchies (rendering members of marginalized groups invisible) even if handles were completely detached from identity.

Beyond that, class membership would also likely be ascribed to people based on their online personalities, behavior, communication styles, or other traits—just as race and sex are ascribed to particular physical bodies even when those bodies are ambiguous (the phenomenon highlighted by the “Pat” character on Saturday Night Live).  As many law professors may remember from the days of handwritten law school exams, even something as innocuous as handwriting carries gendered meaning.  How many of us found ourselves (against our better instincts) assuming that the author of a particular bluebook was male or female just because of the handwriting?

The categories in which we think and with which we order our social world transcend the physicality of that world.  This is not to say that there aren’t liberating aspects to the web.  But they have been seriously overstated, skewing the debate about how much is gained and how much is lost from regulation of the Internet to prevent harms of sexual (as well as racial and other) harassment.

II. The Harm of Cyber Sexual Harassment and Possible Solutions

One harm caused by the types of harassment that Professors Citron and Franks describe is that, as Professor Citron notes, such abuse imposes a tax on Internet access for the vulnerable groups against whom it is directed (women, people of color, and sexual minorities, to name a few).  Before Catharine MacKinnon so trenchantly described the connection between sexuality and power (and even after, unfortunately), analysts and decision makers failed to appreciate that sexual demands in a context of unequal power (such as between employer and employee) result in discriminatory conditions of work.  Race- and sexuality-based harassment likewise impose additional conditions of work that dominant-group members don’t have to endure.  Similarly, when women and other marginalized groups are subjected to harassment on the Internet, being subjected to such conduct becomes a condition of access to this crucially important societal resource. While the disparities in power are not as clear on the web as in the employer/employee context, an analogy can be drawn to coworker harassment—which of course is illegal under the Title VII “hostile environment” rubric.  Just as an employer who fails to address coworker harassment at work makes being subjected to such treatment a term or condition of employment, so a website manager who fails to control harassment by “co-users” of a website makes being subjected to such treatment a term or condition of access to that site.  And, just as “go find another job” is no longer an acceptable response to a plaintiff’s complaint under Title VII, “go to another website” should be considered an unacceptable response to complaints about limited access due to uncontrolled harassment on a site.

Thus, for me, a potential solution that warrants consideration is simply a separate statute that prohibits discrimination based on race, sex, religion, sexuality, etc., in the provision of Internet services.  Unlike Professor Franks, I see cyber harassment as having significant effects in cyberspace itself—as well as, of course, in the other arenas she identifies such as work or school.  There is just one Internet, and it is undeniable that lack of access to it imposes significant social, economic, work, travel, and other disabilities on an individual.  It’s not just that cyber harassment affects the victim’s ability to work or attend school. It also limits an individual’s access to the Internet itself.  As Professor Citron convincingly argues, harassment on the web is a civil rights issue.  I would add that the web itself has become so important to human thriving that, like employment, housing, and educational institutions, Internet services are something that all individuals should be able to access without being subjected to discrimination based on subordinated status.

A civil rights statute directly focused on Internet access, therefore, would address the impacts with which Professor Franks is concerned, but would also protect against a far wider range of impacts that could result when cyber harassment deters or limits individuals’ access to the Internet (for example, by causing them to shut down their email account or social networking page).[5]  This statutory approach (which I lack the space to fully explore here) would not limit potential liability to conduct that has “effects in a space traditionally protected by sexual harassment law,” as Professor Franks proposes,[6] but instead would recognize other effects as equally deserving of remedy.

Of course, a statute prohibiting discrimination in the provision of Internet services would require providers to gather identity information about web-users and bloggers in order to police their sites—raising privacy issues usefully addressed by Professor Franks.  The major confidentiality concerns that compiling such information raises in my mind, however, are not the violation of privacy rights of harassers (or those whom providers might perceive to be harassers), but rather the violation of privacy rights of political dissidents.  Just as telecommunications companies willingly turned over confidential information about phone users to the government in violation of federal law after 9/11, the most serious risk of Internet providers’ identity-gathering would be the risk that providers would similarly cave to government pressure in times of political tension.[7]

III. Conflicting Liberties, the Pornography Parallel, and Access to the Internet

Another analogy that the discussion during this symposium evoked in my mind is pornography—in particular the pornography debates of the 1980s.  The parallels between those debates and the current cyber harassment discussion are striking.  Just as Internet analysts have probed such quandaries as whether it is possible to be harmed by a “virtual” sexual attack, so pornography analysts two decades ago pondered whether people can be harmed by “fictional” film or photographic depictions of the sexual torture of women. Some readers may recall Linda Marchiano, known under her film name of Linda Lovelace and the “star” of the infamous porn film, Deep Throat.  Marchiano was coerced at gunpoint into performing in the film, including being forced to smile so that she would look like she was enjoying the acts in which she was forced to engage.  How different, one might ask, is the harm done to her by dissemination of that film from the harm done to a woman whose avatar is raped online, and made to look like she/it enjoyed the attack (as described by Professor Franks in her article)?  Moreover, separate from questions of the harms to the particular person depicted in each case, one might also ask whether “mere” depictions of abusive behaviors help perpetuate social inequities in the broader society.  Does the widespread, unsanctioned, and highly visible abuse of women and other vulnerable groups on the Internet, for example, contribute to the devaluation of and sense of license towards such groups?  When balanced against the (limited) benefits of unfettered net “free speech”,[4] does such harm justify restriction of harassment on the web?

As these questions suggest, many of the free speech issues raised during the pornography debates are also relevant here, and the answers provided by anti-pornography activists have resonance for the cyber harassment debates of today.  Important questions raised by those activists are worth considering in this new context, questions such as: Is online harassment primarily speech?  Is making a false Facebook page conduct? Does, or should, freedom of speech extend to protection from non-state violators of one’s speech? Is communication on the Internet (as Professor Citron suggests) a zero sum game in which one person’s (or group’s) speech is sacrificed when another’s is protected?

Professor Citron’s insightful argument that allowing harassing speech on the web silences other speech tracks precisely the argument made by anti-pornography activists in the 80s.  And even though the latter’s efforts did not ultimately succeed, their analyses provide useful starting points for some of the debates about cyber harassment today.  The zero sum argument, for example, reveals the incoherence of the notion of state nonintervention in the arena of private speech.  If some silencing of speech will occur under any regulatory regime, then the question becomes: whose speech should the government constrain and whose should it protect?  The illusion of a perfectly free marketplace of ideas falls to the same fate as the Lochnerian illusion of a perfectly free economic market.  And the substantive nature of purportedly neutral state “nonintervention” is revealed as a free speech subsidy for the powerful.

In fact, this line of argument may have even more traction in the cyber harassment context than it did in the pornography context, because, unlike in the porn debate where the contention was that women were being silenced through the delegitimation of their views, here the argument would be that they are actually physically silenced by being intimidated into withdrawing from the virtual “marketplace of ideas” altogether.  The harm to their ability to communicate is more direct, and the silencing more total.

IV. Censorship of Political Dissent on the Web

Finally, it’s worth mentioning the real potential dangers of government regulation of the Internet in the current era—dangers that few defenders of Internet freedom mention, or perhaps are even aware of.  While the web may have the Wild West image of a free speech frontier, in reality it is already vulnerable to censorship.  Web sites carrying messages of political dissent that are perceived as threatening to the existing power structure can and have been closed down, silencing those voices.[5]  Perhaps the most famous example of political censorship on the web is Yahoo’s “assisting the Chinese government in sending four dissidents . . . to prison for terms of up to 10 years.”  Google has also accommodated the Chinese administration by censoring several sites (by failing to report them in search results) blocked by the government.[6]  But similar incidents have occurred here in the U.S. In 2008, for example, POPLINE, the world’s largest database on reproductive planning, yielded to pressure from U.S. AID and changed its settings to ignore “abortion” as a search term.  (The dean of the Johns Hopkins School of Public Health, which produces POPLINE, reversed the decision once he got wind of it.) In 2003, the alternative journalism website YellowTimes.org was removed from the web by it web-hosting company for refusing to take down footage of U.S. soldiers killed by Iraqi troops.  In March 2008, the New York Times reported that, at the request of the U.S. Treasury Department, a British host blocked a Spanish travel agent’s web site advertising travel to Cuba.  According to the Times, the Treasury Department’s Internet blacklist has 6,400 names on it. And of course content-control software is readily available and used by schools, employers, libraries and other institutions.

Rarely does such censorship attract media coverage, much less serve as a rallying call for anti-regulation activists.  Yet the ease and invisibility with which politically-based restrictions on web speech have been imposed should give any lover of free expression pause.  The more-than-a-little-ironic contrast between, on the one hand, the current flurry of panels and scholarship considering whether violent, abusive and misogynist web expression deserves constitutional protection[7] and, on the other hand, the deafening silence around the censoring of political dissent on the net, speaks volumes about what sorts of speech are seen as dispensable and what sorts aren’t—and about the freedom we do or don’t have on the web.

---

[1]. Other familiar tensions that resurface in this context involve the First Amendment – which I will only touch on here, but which has always served as a limit on sexual harassment claims – and the related topics of hate speech and pornography.

[2]. Of course, just as in the workplace context, the regulation of behavior on the Internet would have to be balanced against free speech concerns. For a brief comment on that, see Part III.

[3]. I discuss the issues surrounding political dissent and Internet freedom in Part IV.

[4]. See supra Part I.

[5]. Another danger of web regulation worth considering is the danger of repression of marginalized sexual expression, such as gay erotica. This danger was famously raised during the pornography debates, in an amicus brief written by the Feminists Against Censorship Task Force (FACT).

[6]. Google just recently said that it will stop cooperating with the censorship, after suffering an attack on its system that many assume was perpetrated by the Chinese government. See Edward Wong, Hackers Said to Breach Gmail Accounts in China, N.Y. Times, Jan. 18, 2010.

[7]. At this year’s Association of American Law Schools annual conference (held January 2010, in New Orleans), for example, a panel of big-name scholars discussed whether even professional censure of such behavior is appropriate.