Skip to content

Gender, Censorship, and the Internet: Who Decides What’s “Not Possible”?

On May 1st, the New York Times ran on its front page the second in a series of articles entitled, “Cyberwar: Facing Filters and Firewalls.” According to the article, the Iranian government exercises especially stringent censorship standards: “Search for ‘women’ in Persian and you’re told, ‘Dear Subscriber, access to this site is not possible.’” This is an extreme example of restriction of access to information about sex and gender issues. Recent protests by women in Iran in February and Afghanistan in April make visible the effect of censoring the very word ‘woman’ on the Internet; that is, that it becomes acceptable to restrict women’s basic human rights. It is easy to see how intolerance and fear prompt the exercise of censorship, which in turn perpetuates intolerance by denying people the ability to educate themselves against prejudice. But more subtle forms of censorship on the Internet arguably have an equally restrictive effect on our ability to conceive of genders and sexualities other than the ones with which we are already personally and culturally comfortable.

A series of questions guide this paper and attempt to use censorship in order to think about gender, and gender to think about censorship. How does censorship of gender- and sexuality-related content on the Internet create and regulate gender and heterosexuality themselves? If, as Judith Butler writes, gender is a regulating mechanism, is having a gender like having a constantly functioning internal and external censor? Are we striving for an ideal state of being gendered or not gendered? If so, what could that state be like, and is the Internet helping us to get there?

Expanding the concept of censorship beyond its habitually-used definition helps to address all these questions. The OED defines the modern “censor” this way: “An official in some countries whose duty it is to inspect all books, journals, dramatic pieces, etc., before publication, to secure that they shall contain nothing immoral, heretical, or offensive to the government.” The psychoanalytic definition is: “A mental power or force which represses certain elements in the unconscious and prevents them from emerging into the conscious mind.” This second definition has its roots in Freud, but when a similar phenomenon happens in a political context, it is called a chilling effect. Because of vague laws that broadly restrict certain kinds of speech, a person might self-censor in fear of being prosecuted. The Communications Decency Act of 1996 prohibited the transmission of “indecent” or “patently offensive” speech to minors, but the Supreme Court struck down the CDA as being in violation of the First Amendment (Deibert 228). Censorship, then, can be external or internal. It can be overt or more insidious; J.M. Coetzee said that censorship “looks forward to the day when writers will censor themselves and the censor himself can retire.” When censorship runs most smoothly, citizens feel as though they are being watched even when they aren’t and adjust their behavior according to the norms established by the censoring body*.

Censorship can also be passive as opposed to active in that an absence of laws or policy can stifle speech as much as overt censorship. Policies of affirmative action attempt to fight historical implicit censorship of non-whites’ participation in professional fields, but many other contexts could, and do, benefit from new attempts at compensation for a lack of minority voices. For much of the relatively short history of the study of literature in Western thought, the voices of people of one race and one gender had a virtual monopoly on the field. In A Room of One’s Own, Virginia Woolf famously argued that a woman raised in an almost purely male discourse will not only fail to find literature in a woman’s voice, but also will hardly know how to use the language to express her own. If we expand the concept of censorship to encompass the silencing of female voices in this extremely far-reaching way, then “censorship” becomes much like gender in Butler’s terms when she uses Foucault to talk about gender. Both gender and censorship become “disciplinary discourse[s]” that not only regulate and use individuals’ behavior but also “actively constitute” the individual in that he or she can only speak or exist with the given discourse (Undoing Gender 50). The creation and maintenance of norms in the service of preserving the privileged status of elites—both gender and censorship strive towards this goal.

Two norms that are constantly negotiated in the U.S. are the designation of what is safe for children and what is moral (Deibert 226). These are vexing categories because their legislation always poses some degree of threat to the First Amendment. These threats can be described as disputes over genre, since censorship concerns itself with the maintenance of categories. To be censored, an act of speech must be obscene as opposed to acceptable, immoral instead of moral, porn instead of art. The censor maintains these divisions by enforcing and creating taboos, which vary by country and time. A current governmental debate in Indonesia over anti-porn laws threatens not only to restrict the expression of artists and activists but also to challenge the norms of indigenous culture. Feminist scholar Gadis Arriva called a proposed 2006 law banning many representations of nudity “something very alien to us” because bare breasts are the norm among some Indonesian cultures (Jacobs 36). The mutability of what is considered obscene, immoral, or pornographic suggests that these labels are “only” culturally constructed—but the culturally constructed is always a profoundly convincing fiction.

Perhaps the strongest cultural categories of all are “male” and “female.” A host of characteristics distinguish one in contrast to the other, but when examined closely, the binary can never live up to the ideal that defines it. No woman is free of “masculine” characteristics and vice versa. So what is gender? Judith Butler describes it this way: “Gender is the apparatus by which the production and normalization of masculine and feminine take place along with the interstitial forms of hormonal, chromosomal, psychic, and performative that gender assumes” (Undoing Gender 42). Far from a static definition, gender is a machine or a technology that constantly creates and regulates the categories—masculine and feminine—that are normally thought simply to comprise it. Gender, like censorship, is concerned with the maintenance of categories: male vs. female, straight vs. gay, natural vs. unnatural. As I said above, the censor maintains its divisions by the creation and enforcement of taboos, which vary by location and time. Gender does the same: what is “masculine” in the U.S. in 2009 looks little like what was “masculine” in France in the eighteenth century, for example.

What are the consequences, then, of saying that both gender and censorship are mechanisms of control? Both are internal and external forces that constantly shape how we think about what we read, see, hear, say, and do. One way of considering their relationship is to view censorship as being in the service of gender. People in the U.S. would not consider “women” a “dirty word” because the mechanism of gender is functioning at a level where “women” may be freely discussed in public. In Iran, on the other hand, gender may be working in the culture in such a way that women are in fact considered a dirty subject. If Iran’s censors place the word “women” into the category of obscenity, the real-life results for women range from lower-paying jobs to violence and all forms of sexism in between. The more women can be discussed in public forums, the less people will feel allowed or compelled to abuse them.

To put it another way, the less the mechanism of gender is functioning, the less censorship will seem necessary, and the more normal it will seem to treat women and men equally well. On the other hand, when gender is a pervasive force in a society, the distance between the genders seems unbridgeable. Butler says that using “a restrictive discourse on gender” that accepts male and female as a binary has the effect of “naturaliz[ing] the hegemonic instance and forecloses the thinkability of its disruption” (Undoing Gender 43). Understanding gender simply as male vs. female necessarily means understanding men as dominant over women, legitimizing a power structure that privileges some and marginalizes others. Similarly, if we allow censorship to convince us that some speech is acceptable while other speech is obscene, we privilege one culturally constructed category over another. Butler suggests “undoing gender” as a way to open up new possibilities and overcome a hierarchical power structure. Undoing the censorship of “obscene” representations of sexuality and gender could create the same effect.

This censorship of open discourse about non-normal gender and sexual practices promotes intolerance and restricts a more open way of thinking about other people. The Internet presents a new and uniquely revealing way of tracking how the technology of gender uses censorship. Every country that gains access to the Internet has to grapple with reconciling its own cultural gender mechanism with the chaotic exchange of ideas online. This process of determining the “correct” state censorship response is profoundly complicated by the international nature of the Internet. The New York Times describes a case where political censorship and sexuality intersected in Turkey in March 2007. The Turkish government censored from all its citizens in response to a series of videos alleging that Mustafa Kemal Ataturk, the founder of modern Turkey, was gay; to insult this leader is a federal crime. Google, which owns YouTube, struggled to square the site’s terms of service, which prohibit hate speech while permitting political speech, with the laws of the Turkish government. Their solution was to block the videos, but only from servers in Turkey. The government seemed appeased until last June, when they requested that the offending material be taken down worldwide in deference to Turks everywhere. Google refused, and the entire YouTube site remains censored in Turkey.

Insofar as it is an instance of gender in action, this incident hinges on the word “insult” and the controversy over international standards of decency. Other issues besides gender are certainly in play here—after all, most U.S. citizens would probably not be able to identify Ataturk, while he is a revered figure in Turkey. Yet the status of “gay” in Iran as an insult serious enough to warrant the blocking of a popular Internet site from the entire country is an example of how cultures stigmatize sexualities. Censorship and gender aid sexism and heterosexism at their most basic levels: at the formation of categories, where “normal” sexual expression is defined in opposition to “anomalous” sexual expression. So, in a country with less clearly defined gender categories, “gay” may not be a slur that requires censorship by the state.

Controversy over what is obscene occurs not only between but also within countries. Censors are the agents in a society who possess legislative and discursive power in the first place; historically, these people are wealthy white males. The key concerns and values of these censors will likely be different that those of other groups—thus the issue of privacy, for example, might be different for a woman who works in the sphere of the home than for a man in a business environment (Ramilo 4). Clearly, if the power to censor were to lie in the hands of those whose identity expressions have been most often censored—“gays, women, heretics, traitors, and troublemakers” (Jacobs 38)—then the definition of “obscene” would look quite unlike what it does here and now.

To talk about sexuality and gender (as with “gays and women,” above) simultaneously in terms of censorship is not to conflate them because, in Butler’s terms, “the implicit regulation of gender takes place through the explicit regulation of sexuality” (Undoing Gender 53). To censor queer sexual representation on the Internet is to insist on a normative definition of male, for example, which includes a clause stating that males are attracted to females. A “mainstreaming of sexual politics” on the Internet by groups like the UK’s Internet Watch Foundation could lead to intolerance of all of kinds of sexual practices that do not coincide with normative gender politics (Jacobs 39). Normative sexuality upholds normative gender roles, while new Internet-based sexual practices have the potential to subvert them in ways that give people new freedoms of personal expression. The practice of nudity-chatting—long-distance sexual communication using webcams—is currently being censored in China under obscenity legislation (Jacobs 36). In the U.S., texting nude pictures of oneself, or “sexting,” has recently been in the courts and in the news as a controversy between standards of decency and First Amendment rights. “Decency” is relative, as are sexuality and gender themselves—to “mainstream” them by restricting the legality of their tools of expression is to deny some people their freedom of speech.

In the early days of the Internet, people who believed that online spaces were so novel and chaotic that the state could not interfere with their governance called themselves cyberlibertarians and were disparaged as cyberutopians (Deibert 106). As the conflicts over Internet decency standards demonstrate, the Internet is no censorship-free utopia. The truth is that it cannot nor should it be. Children do need legal protection from predatory pornographers, and parents have the right to use filters to censor what their children see online.  The misogyny of much porn is certainly problematic for feminists.  Porn produced from hidden cameras in bathrooms is unethical. As the history of the First Amendment has shown, there is a delicate balance between the right to privacy and the right to free speech: between “freedom from” and “freedom to.” In order to challenge categories and the status quo, however, the safer option—and, paradoxically, perhaps the more courageous one—is to err on the side of more speech, not less. To do so is safer because the protection of free speech is protection against authoritarian states in which no one can remain safe from the despotic power**. At the same time, ensuring that people whose views one thinks are dangerous have the right speak does always seem like the safest option.

In terms of gender, “the status quo” challenged by free speech has many meanings. In the U.S. and many other countries, the status quo means a pay gap between women and men, beauty standards that are unrealistic at best, homophobic environments, and certain definitions of “masculinity” that reward aggression and of “femininity” that do not reward personal assertion. In countries like Iran and Afghanistan, it means severe legal restrictions on civic freedoms and basic human rights for women. Freedom to express dissenting opinions and alternative actions means, for women in particular, freedom from oppression by male hegemony.

The Internet is no post-gender, post-censorship utopia, but it does create new spaces in which to “practice freedom.” In fact, the Internet offers so many possibilities for non-normal gender and sexual expression that entire books, such as Katrien Jacobs’ 2007 study Netporn: DIY Web Culture and Sexual Politics, have recently been written about these activities. There is no utopia, online or off, but there is relative freedom from categories that you never chose for yourself, but were imposed upon you from “above”: gay, straight, man, woman, female, male. Butler argues for the deconstruction of the category of “women” because to assume that all women have something in common is to ignore the varied “cultural, social, and political” factors that shape people’s lives (Gender Trouble 14). Secondly, the “something” that is common in and unique to women has never been identified and, most likely, simply does not exist. To escape the categories of “women” or of “straight,” etc., by breaking them apart has the important effect of destroying or stripping away the labels that go along with them: immoral, wrong, obscene, pornographic.

A recent Internet scandal, known by its Twitter tag “#amazonfail,” strikes at the heart of the intersection between gender and censorship because it raises two questions: What are the consequences of limited searches? And, why do we desire to assign the labels designating sex and gender content in the first place? The public relations debacle, documented here by Roisin, took place over Easter weekend. Originally, the author Mark R. Probst discovered that his book The Filly had been removed from the sales rankings on After he submitted a complaint, the site informed Probst that his novel, a gay-themed romance, had been designated as “adult content” and eliminated from searches in an attempt to make the site more family-friendly. The subsequent barrage of Twitter and blog postings exposed more wide-reaching effects, which amounted to Amazon’s apparent censorship of content that presented homosexuality and feminist theory openly or in a positive light. Books like Queer Theory: An Introduction and Greek Homosexuality did not appear in searches, while the first result in a search for “homosexuality” was a homophobic guide for parents ( Heteronormative representations of nudity such as Playboy: The Complete Centerfolds seemed to have been considered appropriate for all ages ( In a public statement, Amazon admitted to human error but not to human intent, calling the incident an “embarrassing, ham-fisted cataloging error,” a “problem,” and an “accident.” Many Internet users remain unconvinced and consider “Amazonfail” a serious breach of trust.

The censorship of certain terms in search engines may be the most dangerous breed of online censorship. Describing China’s Internet filters, the New York Times writes, “As with George Orwell’s ‘Newspeak,’ the language in ‘1984’ that got smaller each year, governments can block particular words or phrases without the user realizing that their Internet searches are being censored.” The alarming consequences in a case like Amazonfail are that the “restrictive discourse” of gender to which Butler refers may become infinitely more restrictive, depending only upon the degree of fear and prejudice of the ruling censors. Words, after all, are the building blocks of all language, but specifically of free speech.

While Amazonfail may be an all-too-good example of the censorship of search engines, perhaps the more pertinent issue it raises is why the labels that allowed the censorship to happen exist in the first place—both the “gay” category and the “adult” value judgment assigned to it. In her project, Roisin deems it a moot point whether people at Amazon had the homophobic intent to censor these products, and she asks an excellent question: “What, exactly, is the point of gendered tags to begin with?” She questions the difference between “adult, erotic, and sexuality” in terms of categorizing books and concludes that an “error” like this was only to be expected because Amazon is “run by people, and people live and breathe gender and sexuality dichotomies.” Amazonfail cannot be explained away by a glitch (as the subsequent Twitter tag #glitchmyass suggests). The error was enabled by gender and sexuality designations that are necessarily divisive because they call attention to categories that are non-normative and outside the mainstream. To announce them is to invite them to be censored. The ubiquity of gender and sexuality labels makes them invisible, hidden in plain sight; it is largely for this reason that the censorship of non-normative representations may be so easy to perpetrate and so easy to ignore, as Amazonfail makes manifest.

At an even more basic level, censorship is in effect as soon as a label is assigned to any product or anyone. Assigning one identity or category precludes the expression of others because categories of gender themselves only exist in contradistinction to their opposites (Gender Trouble 9). If one is male, one cannot be female. If a book is “gay,” that is, if a book has gay themes, it cannot be “appropriate.” What follows from this categorization is that to be “gay” is to be “obscene,” or “not normal.” The widespread outrage at Amazonfail is a testament to the fact that in the U.S. at this time, the function of gender does not generally output “obscene” when the input is “gay.” However, part of the issue between Turkey and YouTube may be that in Turkey, this is currently the case. Moreover, as long as “gay” exists as a category, it must always be in opposition to “straight,” which, as evidenced by the word itself, inherently implies the moral. “Gay” signifies “immoral” as long as it remains its own genre.

What does it mean to input “woman” and receive, “not possible”? To restrict a discourse, to censor free speech, is to allow the mechanism of gender to grow so large and powerful that the normative definitions of male and female arrive at their logical conclusions: man becomes the only subject, and woman as an object is reduced not simply to a private but to a secret sphere of existence. To represent her is impossible. Censorship may not be merely a tool of gender, but its essence.

In utopia, categories of gender and the censorship they demand would not be necessary for us to make sense of the world—but utopia means “no place.” A “post-gender,” “post-censorship” utopia reaches beyond language, making it unrepresentable and thus “uncreatable.” The disassociation of “gay” and “immoral” may be impossible, but that doesn’t mean that Internet users are not trying. It takes only an optimistic outlook, not a utopic agenda, to err on the side of “freedom to,” instead of “freedom from,” and expect that a constantly expanding discourse will lead eventually to the best possible world. Adult, immoral, wrong, obscene, pornographic—if more exposés like Amazonfail make the designations, and “censorship” itself, the dirty words, rather than the representations they designate…that kind of intervention might massively expand the limits of what seems possible.

*A macro-application of Foucault’s reading of the Panopticon; Discipline and Punish.
**Fictional representations of these anti-utopias built on censorship include Margaret Atwood’s The Handmaid’s Tale and Jorge Luis Borges’ “Utopia de un hombre que está cansado” (“Utopia of a Tired Man”).

Works Cited

Butler, Judith. “Gender Regulations.” Undoing Gender. New York: Routledge, 2004.
— “Subjects of Sex/Gender/Desire.” Gender Trouble: Feminism and the Subversion of Identity. New York: Routledge, 1990.

Diebert, Ronald, et al., eds. “Internet Filtering in the United States and Canada.” Access Denied: The Process and Politics of Global Internet Filtering. Cambridge, MA: The MIT Press, 2008.

Jacobs, Katrien. Netporn: DIY Web Culture and Sexual Politics. New York: Rowman & Littlefield Publishers, Inc. 2007.

Markoff, John. “With New Software, Iranians and Others Outwit Net Censors.” New York Times 1 May 2009.

Ramilo, Chat Garcia, et al. “Who is Ruling the Internet?: Gender Sensitive Research into Internet Censorship as a Central Area of Internet Governance.” Creative Commons Attribution- Noncommercial- No Derivative Works 3.0 Germany License.

Rosen, Jeffrey. “Google’s Gatekeepers.” New York Times Magazine. 30 November 2008.

Zittrain, Jonathan, and Palfrey, John. “Reluctant Gatekeepers: Corporate Ethics on a Filtered Internet.” Diebert et al. 104-122.