Why is racism online so common? Why does it persist? If, as Gabriella Coleman said during her keynote address at the 2013 Association of Internet Researchers meeting in Denver, Anonymous is depicted by the popular press as a “Hate Machine,” this is only because it is so clear that the amount of hate on the Internet is so vast that it must exceed the abilities of humans to craft it individually. If, as Steven Marche claims, “the Internet has reached peak hate,” this year, this is a function of the Internet’s having really arrived as an irreversibly central part of daily life. Hopes that the Internet would become less racist as it became more mainstream have been definitively dashed. 
But to return to the question—why is the Internet so racist and sexist? Where scholars fear to tread, the Internet goes without fear, thus the knotty question of Internet racism’s source and longevity has been answered by collective intelligence. The Greater Internet Fuckwad theory: or GIFT, a popular piece of vernacular Internet criticism that first appeared on Penny Arcade in 2004, posits that anonymity provides a cover for online racism, that platforms such as video channels and games provide an audience, and that “normal” people are unable to resist indulging in performing racism when given these affordances. This is a very common argument, both within and without academia, and as such deserves serious analysis. This theory makes online racism out to be an effect of the Internet’s abilities to provide both cover and alibi, asserting that it is inevitable, natural, and “normal,” given these conditions to produce racist discourse. This theory assumes that there exists within every person a flood of otherwise governable desire to shout racist, sexism, homophobic words: that the Internet produces Tourette’s syndrome. However, this gives a bad name to Tourette’s; people with this condition utter offensive language that they do not, in fact, feel or believe, making this a fundamentally innocent use of language. In contrast, the notion that the Internet’s anonymity (really, pseudonymity in most cases) and audience encourage an act that “normal” people have never had access to before—a way to hide the behavior and an excuse for performing it—endows the Internet with its own agency in the production of racist discourse. The Greater Internet Fuckwad Theory makes us all out to be the fuckwads. Identifying racist behavior on the Internet as emanating from “normal people” foregrounds the act’s technicity. It’s not the actor, it’s the network. 
The current move towards acknowledging the autonomy of media devices, specifically digital media devices, as animated by their own forms of communication, liveness, agency, can be found in the work of Wolfgang Ernst and other European media archaeologists. For example, in the Software Studies: A Lexicon , Olga Goriunova and Alexei Shulgin write that “a glitch is a short- term deviation from a correct value and as such the term can also describe hardware malfunctions.” Racism is regarded within Internet culture as spam, noise, and trash: as a digital artifact, in the purely technical sense: when we see big blocky pixels in our VR worlds, feature phones, or throttled “streaming” videos that stutter or refuse to stream, we are forcibly reminded of the network’s limits—it’s material. Online racism is that limit’s horizon, paradoxically regarded as an obstacle to communication rather than inevitably a part of it. Viewing online racism as an artifact—of an earlier, less civil time that is both reviled and viewed with intense nostalgia, of a network that inevitably reminds us of its activity—lets us move beyond perceiving of it as an inevitable but inconvenient fact. Racism such as the example above, received by a female gamer who shared it with Fat, Ugly, or Slutty, does more than reduce the value of gaming networks as digital commodities.[i] We know we have encountered a glitch, as Goriunova and Shulgin write, when “something obviously goes wrong.” Glitches represent the user’s loss of control over the machine; they are often exploited as part of avant-garde art practice because they forcibly remind viewers of the material base of digital events. When we see arguments that racism is the inevitable product of the Internet, we are in the presence of vernacular digital archaeology: the network has become animate again, but in a way very different from nineties cyberpunk narratives that imagined intelligent autonomous networks that wanted to take over the world, a la The Matrix. This network doesn’t want to enslave humans in pods and use them for energy, instead it wants to channel that energy in a network to produce a regime of swears. Networks produce profane and nonsensical discourse just as they produce other types of unintentional and unwanted behaviors, like slow loading, dropped signals, and damaged files: something has gone wrong, but it is always doing so and always will. Errors aren’t alien to the system, they are part of the system. 
I don’t agree with the Greater Internet Fuckwad theory, however popular it may be. It claims that the “intentions, feelings, or opinions of users” don’t really come into play in everyday racism, that this kind of racism isn’t really racism because it is a network effect as well as a human effect—the product of human computer interaction, indeed all human computer interaction. In other words, everyday online racism it is a “glitch” or malfunction of a network designed to broadcast a signal, a signal that is highjacked or polluted by the pirate racist. Perhaps this is why everyday racism is so common in 4chan; ironic, anarchic, and otherwise non-mainstream critique is part of the overt ideology of these sites, and racism becomes conflated with other, less sinister political projects, like protecting free speech and sharing code. What if, in the spirit of media archaeology, we understood online racism not as a glitch but as part of the signal? What if we paid attention to racist comments with the same intensity that we do the rest of the content? Why is Internet racism the bad penny of the internet economy? Online racism is unruly, nonsensical, but it is not an interruption from the “content,” indeed, it is less an exception than an inevitability. On this count, I do agree with GIFT. New media theory has been obsessed with the new: why must racism be more interesting when it takes on new forms? It is as old as the network itself.
Rather than viewing Internet racism and sexism as “off topic,” an example of filter failure, what could we gain by seeing it as an effect of the Internet on a technical level? Viewed in this way Internet racism is neither a barometer of sincerely held opinions, nor the act of evil and lawless users who want to destroy communities. It is a discursive act in itself, not an obstruction of that act.Content that includes people of color often becomes part of a technosocial assemblage that produces racism and sexism. For example, user comments on electronic and new wave music videos from the eighties rarely veer from the ecstatic, while comments on hip hop videos are far more polarized. While there are plenty of sites and communities in games, mobile media, and social networks that are devoted to overt and organized forms of racism and sexism—white supremacists and the like, studied beautifully by Jessie Daniels and Manuel Castells—I am more interested in “normal” racism, the kind that we all accept as part of the bargain we’ve made with the Internet in exchange for free and less-moderated content than we can get in “old” media. As many have told me, if you boycotted every place on the Internet where you’ve seen a racist comment, you wouldn’t go anywhere. In self-defense many people have stopped reading comment sections, or at least claim to, in the same vein that many disavow television watching.If the Internet in any way resembles the public sphere that so many scholars have compared it to, it is in the comments sections of local newsites, popular blogs like the Huffington Post, and other sources of news and information that our social networks so often link us to. Hence, the inescapability of Internet racism, for these are exactly the sites where it can most reliably be found. Our desire to learn what others think about things we’re obsessed with make us the audience members that provide the conditions for online racism. Miley Cyrus’ twerking performance, the trial of Trayvon Martin’s murderer, and the U.S.’s use chemical weapons in our foreign wars are all racialized events that we want and need to talk about online. However, our experiences when we try bear predictable fruit.These are “real time” media, the kind that Geert Lovink claims characterizes the digital realm, the kind that venerable digital media scholar Henry Jenkins has dubbed “participatory,” and it is a precursor for everyday or real-time racism. The digital humanities, a field that has been organized around the paradigm of the archive, has shied away from analysis of “real time,” leaving this to social scientists. In contrast to the often obsessively (and poorly) evidenced arguments that appear on Stormfront.org and other overtly racist media, it is characterized by its incoherence, the digital equivalent to being ching-chonged by a carful of speeding kids or having the word “bitch” or “slut” whispered or shouted at you on the street. Online racism’s resemblance to embodied racism stops there, however. The difference is in context—online racism’s audience isn’t people of color, or women, or gay people. It’s “the public.” Trolling, or performing obnoxious, controversial, or emotion-inducing speech in order to garner attention, is called this because it’s not about targeting an individual and is therefore perceived as benign. As Kishonna Gray’s excellent study on Xbox racism found, the abusive gamers “view their behaviour as annoying – not racist”. One abusive player, when confronted about using the n-word, explained: “It’s just a stupid word. I just say it to fuck with people.” Let’s Not Call It Trolling To call Internet racism “trolling” is to minimize it as an act that is not really about racism, but fundamentally about something far more benign; its not really about anything. Many claim that trolls want attention, or are simply displaying healthy irreverence (what other kind is there?). Trolls indulge in their desire to skewer anti-racist sincerity as naïve, or even worse, boring. It turns racism from black to white, bleaching it clean of its overt meanings. 
And the Internet Fuckwad theory is complicit in this. It makes it clear that everyday racism is an effect of digital networks—the more pervasive and integrated into everyday life these networks become, and the more pseudonymity is available, the more racism that equation will produce, regardless of the intentions, feelings, or opinions of users. It is technologically deterministic. Going by this notion, there is really nothing that we can do about online racism except to change the network and its behavior by requiring authenticated real names participation in all digital environments and improving filtering such that all comments this like are flagged and removed, thus removing the audience. However, none of us would want an Internet like that. If GIFT is correct, the above strategy should work, but online racism is not simple. “Trolling” is a euphemism used to dismiss and minimize hate speech because it is indeed a glitch, a sign that something has gone wrong, and yet a sign of exactly the opposite: “trolling” is viewed as a sign that a network is robust, and it is being used and used hard. Users themselves understand it as an algorithm, as a system or assemblage, formulated to explain one of the most everyday aspects of online life, ranging from innocuous to triggering in its effects. Parsing the logics of this algorithm (for what is algorithm but ideology in executable form?) can help us understand the assemblage of disavowal of racism and its effects and how it has come to inform life after new media. The network is shouting at us, and these shouts aren’t noise: dismissing them as such exposes filter failure on a different magnitude.