Abstract
This article presents a new social framework for understanding the origins of trolling and its expansion from an obscure practice, limited to a handful of boards on Usenet, to a pervasive component of Internet culture. I argue that trolling originated, in the term of sociologists, as a form of boundary maintenance that served to distinguish communities of self-identified online insiders from others beyond the boundaries of their community and to drive outsiders away from their spaces. This framework can help us to better understand the transformations that trolling has undergone in the decades since its inception, as well as the persistence of misogyny and prejudice throughout the history of the practice.
Keywords
This article presents a new social framework for understanding the origins of trolling and its expansion from an obscure practice, limited to a handful of boards on Usenet and other forums, to a pervasive component of Internet culture. Although in recent years the practice of trolling, a term and concept that dates back to two decades, has attracted a growing number of empirical studies, research on trolling continues to draw charges of being atheoretical. This article seeks to add to our theoretical understanding of trolling. I draw upon articles about the practice published decades ago that have been overlooked in more recent accounts, as well as case studies that can help us to better understand the history of trolling, a subject that has thus far received only modest scholarly engagement. I argue that trolling originated, in the term of sociologists, as a form of boundary maintenance that served to distinguish communities of self-identified online insiders from others beyond the boundaries of their communities and to drive outsiders away from their spaces. 1 This framework can help us to better understand the transformations that trolling has undergone in the decades since its inception, as well as the persistence of misogyny and prejudice throughout the history of the practice. The article closes with suggestions for further research.
In 1997, the Internet scholar Michele Tepper published a study on what was then the novel practice of “trolling,” or writing incorrect information online with the hope of being officiously corrected by other Internet users. 2 In her article, she takes the stance of a cultural anthropologist documenting an obscure phenomenon that will likely fade away as time passes and the Internet changes. At the time, few could have suspected how important a phenomenon trolling would become in mainstream culture, both on and off the Internet. Within a few decades, however, Internet users recognized trolling to be a serious cultural problem, in large part because trolling had come by then to encompass a range of hostile and destructive behaviors. 3
Perhaps in any unfolding of events, the online world would have come to absorb the hostile behaviors that we have come to watch out for: flaming, trolling, bullying, and shots fired in the dark. 4 Even when written with good will, digital communication often reads as terse and aggressive, lacking as it does the context cues of face-to-face communication. 5 However, as this article will argue, the feeling of exclusivity that belonged to the early Internet, the sense that it belonged to insiders who knew the secret codes, meant that hostility played a role early on as a gatekeeping measure for cyberspace. For the self-identified Internet insiders who frequented hacker forums and specific online communities, hostility was one of several forms of boundary maintenance that served to identify outsiders to repel from their communities.
Today, researchers disagree about exactly what constitutes trolling. As Christine Cook et al. note in a 2018 study of trolls in gaming, despite its prevalence in cyberspace, trolling as a subject of academic study is a confusing space, with different researchers using different criteria to describe the same phenomenon. This is likely due to the fact that it is such a new field of study: existing studies are few and far between, and nearly all of them have been atheoretical due to a lack of empirical basis upon which to build any theories.
6
Although definitions of trolling tend to agree that trolling is hostile, the precise suite of hostile behaviors that constitutes trolling is a matter of debate. Some definitions of trolling include insults and exclusion tactics, others do not; some count griefing and flaming as forms of trolling, others do not; some count unintentional offenses as trolling, others do not (Cook et al., 2018; Hardaker, 2010: 3323–3340). 7 One ambitious definition suggests that any deception online may be viewed as trolling. 8 In their empirical study, Cook et al. (2018) grouped self-described trolls into three categories: trolls who seek to ruin the pleasure of others, the largest and best-publicized group; trolls who like to stir up drama; and trolls who seek camaraderie through mutual trolling, by far the smallest group (pp. 3323–3328).
Scholars agree that trolling as a practice has changed substantially since its inception. Nonetheless, accounts of the origins of trolling tend to be strongly linked to theories about the behavior’s purpose. The Internet scholar Whitney Phillips traces the “roots” of trolling alternately to the imageboard 4chan and the website Something Awful.
9
Gabriella Coleman, in her book on the 4chan-based hacker collective Anonymous, traces the term and practice simply to “Usenet and many other … mailing lists.”
10
Phillips, drawing on self-reported information from trolls of her own day, describes the purpose of trolling as “the amassment of lulz, an aggressive form of laughter derived from eliciting strong emotional reactions from the chosen target(s).” She joins this drive for lulz via outrageous actions with a historical account in which trolling has always meant acting outrageously: In the ’90s, “to troll” was to disrupt a conversation or entire community by posting incendiary statements or stupid questions onto a discussion board … In these cases, “trolling” was used as a general, condemnatory, post-hoc descriptor of an online encounter. It was—and in many circles remains—something you accused someone else of being.
11
Coleman (2014) similarly predicates her account of trolling as a form of cultural sabotage in the trickster tradition on a history in which trolls have always been agents of chaos: in the beginning, trolling “referred to people who did not contribute positively to discussions, who argued for the sake of arguing, or were simply disruptive jerks (intentionally or not)” (p. 132).
This article likewise finds something of trolling’s destiny in its beginnings, but it lays far more emphasis on the differences between the practice of trolling at its inception and the behaviors that the term later came to define—more precisely, to identify the functions that have remained constant throughout its transformations and functions which helped to motivate its spread throughout the online world. Understanding the role that boundary maintenance played in the early history of the public Internet, and in the early history of trolling in particular, can help us to better understand the functions of well-publicized forms of online aggression. The pranks that constituted early trolling were not acts of undirected aggression; even the campaigns of “mass trolling” that began, in the late 1990s, to inundate whole Usenet forums with spam represented a purposeful game that the regulars of other forums recognized as a purposeful game. However, these mass trolling campaigns also represented an emerging shift in the purpose of the game away from policing small Internet communities to policing the Internet at large.
Alt.folklore.urban
Tepper documents the emergence of trolling in the early 1990s in a specific Usenet newsgroup, alt.folklore.urban (AFU). 12 Usenet was a network of thousands of bulletin-board newsgroups. This newsgroup, which its regulars called AFU, was devoted to discussing urban legends and confirming or disproving their veracity. Today, this work is the purview of the website Snopes.com. It may interest readers to know that the creator of Snopes.com got his start on AFU, where “snopes” was his username. He was, as Tepper (1997) writes, one of “the two most notorious trollers in AFU” (p. 48).
Usenet forums were far from the only sites of malicious behavior on the early Internet, of course. In 1993, the journalist Julian Dibbell discussed sexual harassment in online text-based virtual communities in his famous article “A Rape in Cyberspace.”
13
Spamming, flaming, and other terms for online grievances were familiar parlance by the mid-1990s.
14
Arguably, even some of the rumors that AFU existed to demystify counted as malicious online behavior, since they played on the fears and credulity of Internet novices—the equivalent of efforts today by 4chan users to trick people without technological knowledge into trying to charge their smartphones by placing them in the microwave.
15
One such rumor warned people about an email with the header “Good Times” which, when opened, would delete the victim’s hard drive. In 1994, a satirical post on AFU conflated this rumor with an urban legend that a car with flashing headlights may belong to gang members undergoing an initiation: If you receive a message from AOL titled “VIRUS ALERT!” DO NOT FLASH YOUR LIGHTS AT IT! Your computer will crash! This is a secret initiation ritual that all AOL members have to go through. I know this is true because I read it on the internet. (Dragoon, 1995: 24)
However, the origin of the specific verb troll, and with it the now-obsolete noun troller (today superseded by the noun troll), is worth noting for the clearer, more nuanced view it gives us of how the motivations and self-conception of early trolls informed the motivations and self-conception of other hostile online actors, who carried the term trolling to other platforms and adopted it with pride as a label for their own behavior.
The practice of trolling arose within AFU because, although the newsgroup had a strong, well-known community of regulars, its purpose meant that new people often came in with inane questions or urban legends that the regulars had been seen many times and long since disproved. Regulars replied to particularly egregious newbie posts with even more egregious misinformation. They also began putting up ridiculous posts meant to catch newbies as soon as they arrived (Newbie: “You remember incorrectly. Jamie Lee Curtis was NOT in Star Wars. That was Carrie Fisher.” Troll: “Ridiculous. Carrie Fisher is much too small and slight to carry that heavy hairy suit around all day on the set.”; Tepper, 1997: 40–42). The term trolling was meant to reference the practice of fishing by letting a line drift with bait, as many sources of the time noted—for example, The New Hacker’s Dictionary (1996), which attributes the term’s origin to AFU: troll v.,n. [From the Usenet group alt.folklore.urban] To utter a posting on Usenet designed to attract predictable responses or flames. Derives from the phrase “trolling for newbies” which in turn comes from mainstream “trolling,” a style of fishing in which one trails bait through a likely spot hoping for a bite. The well-constructed troll is a post that induces newbies and flamers to make themselves look even more clueless than they already do, while subtly conveying to the more savvy and experienced that it is a deliberate troll. If you don’t fall for the joke, you get to be in on it.
16
As Tepper explains, trolling on AFU initially required subtle irony: an ability to mimic the officious, pedantic tone of the know-it-all: In trolling’s Usenet incarnation, the hook is baited with misinformation of a specific kind: if it is at first glance incorrect, and at a second or third glance comically incorrect, in a deliberately comic way, it’s probably a good troll … Trolls, like puns, are most despised by those unable to make them.
Also like puns, trolls were most successful if delivered deadpan: “Al Capone, who played the heavy on so many Abbott and Costello movies, died decades ago, so I doubt he’s on The Untouchables.” When snopes trolled a forum discussion about errors in Star Trek (he crossposted to both AFU and the hierarchy of newsgroups under rec.arts.startrek), his complaint—about a small ship casting a shadow on a large one—got the tone just right: “Hello? Are there any technical advisors working on this show? Do they really think that objects cast shadows in a vacuum? I know zip about physics, but even I could spot that one” (Tepper, 1997: 39–43). His mimicry of Internet pedantry is spot on. 17
Trolling was a method for the newsgroup to reinforce its values and guard the community against outsiders. Tepper (1997) argues, it serves the dual purpose of enforcing community standards and of increasing community cohesion by providing a game that all those who know the rules can play against those who do not. It works both as a game and as a method of subcultural boundary demarcation because the playing pieces in this game are not plastic markers or toy money but pieces of information. (p. 40)
The newsgroup prided itself on the ability to get information right; it also prided itself on its own history as a community. People who came in and made noise without knowing that history or doing basic research were therefore punished: The corrector, being outside of the community in which trolling is practiced, believes that he is proving his superiority to the troller by catching the troller’s error, but he is in fact proving his inferior command of the codes of the local subculture in which trolling is practiced. (Tepper, 1997: 41)
Indeed, AFU regulars dropped hints that a troll was a troll using code words that the “trout” being baited would not recognize: This ignorance is highlighted within AFU by the group custom of noting in the message header that the post is a troll; acceptable notations include “troll,” “llort,” and, as an AFU in-joke, any variation on the phrase “Phil Gustafson scooped up from lake, scuba gear and all, dropped on forest fire.” (Tepper, 1997: 41)
The phrase refers to an urban legend that, presumably, AFU members grew tired of seeing reposted by newbies who did not know it had long since been debunked.
Accounts of trolling on Usenet sometimes appeared in introductory guides to the Internet, a genre of printed book that largely faded in the early 2000s. The cautionary tone of these accounts implicitly acknowledged that troll posts often targeted the novices who purchased introductory guides.
18
An official Netscape handbook published in 1997 titled one section, “Look Out For Trolls,” explaining that “A troll is a newsgroup post specifically planted to cause discord.”
19
A UNIX handbook published that same year warned readers to look out, while on Usenet, for deceptive forum posts: This baiting of the gullible is known as “trolling” and is quite a pastime on some groups, such as alt.folklore.urban. Basically, there are subjects that have come up so often that they’re beyond Frequently Asked Questions and into “Good Grief!” status … It’ll be a post of the type that will make all newbies immediately want to write “Geesh, what are you? Stupid?” The group oldies will, of course, obtain great entertainment value from these posts. The more insulting, the better. You’ve been reeled in. How do you tell a troll from someone actually saying something stupid? Often, you can’t unless you’ve been reading the group for awhile.
20
Here, as elsewhere, a major purpose of trolling, and the basis of its power to amuse, was the distinction of those who knew the subtle cues of the insider from those who did not.
Boundary maintenance
Sociologists use the term boundary maintenance to describe how the members of social groups identify and police the boundaries between themselves and purported outsiders. 21 A group may practice boundary maintenance, among other actions, by excluding or attacking outsiders, circulating in-group slang or jargon or cant, adopting styles of clothing and comportment meant to identify insiders (and perhaps repel or intimidate outsiders), or devising taxonomies of in-group and out-group membership. 22 For instance, on underground hacker forums in the 1990s, hackers engaged in boundary maintenance, in part, by classifying outsiders to keep out of their community. According to the taxonomy enshrined on these forums, “real” hackers were not end users, who treated computers as glorified word processors and learned the minimum amount that was necessary to use them; noobs, who were a lower version of end users; crackers, who were criminals with no intrinsic drive for knowledge; script kiddies, a showboating version of crackers; or lamers, who tried to be hackers but lacked the necessary skills. 23 Their method of policing their boundaries was open aggression: if someone on a forum asked a question that seemed to give him away as a lamer, he was flamed and given fake assistance. 24 The language on these forums was also deliberately aggressive, trip-wired with slang shibboleths and laced with obscenities—part of the message being, get out of here if you can’t take the heat. Even the user handles—Dragon Lord, Black Avenger, Death Stalker, Storm Bringer, Hitch Hacker—were meant to intimidate squares 25 (Not coincidentally, the same desire to estrange can be seen in the names of punk bands, which provided hacker forums with many of their aesthetic features and political commitments: Black Flag, the Clash, Conflict, the Dead Kennedys, Dick, DOA, MIA, Minor Threat, the Misfits, the Rejects, the Sex Pistols, the Unwanted, and the Worst). 26 To be clear, I am not suggesting that hacker communities on the early Internet belonged to the same population as trolls; only that boundary maintenance, the identification of in-group members and rejection of out-group members, was a major concern of communities of self-identified insiders or old-timers on the early Internet, of which the behavior of underground hacker communities offers a potent example.
On AFU, trolling functioned explicitly as a form of boundary maintenance, having, as it did, the immediate purpose of identifying outsiders and singling them out for the community’s ridicule. At the end of a trolling session on AFU, the troll might post an acronym, “YHBT. YHL. HAND” (“You Have Been Trolled. You Have Lost. Have A Nice Day”), that, as Tepper notes, “in its very unintelligibility mocks the catch’s outsider status.” (AFU regulars joked that the acronym should be “WAFU, YN,” or “We’re alt.folklore.urban, you’re not”; Tepper, 1997: 42–43). In its function as a form of boundary maintenance, the practice of trolling illustrates the importance, for self-identified Internet insiders, of maintaining an identity separate from that of the world at large.
As a technique for policing the boundaries of a community of self-identified Internet insiders—in the initial instance, that of an Internet newsgroup—trolling was quickly understood to be adaptable for use in other communities, even though the performative stance that it relied on, the stance of the condescending, confident, and thoroughly incorrect self-appointed authority on a given subject, was perhaps especially recognizable in a community dedicated to adjudicating urban legends, Internet rumors, and what came, much later, to be described as “fake news.” The AFU newsgroup’s Frequently Asked Questions (FAQ) discouraged trolling in other newsgroups—“if you must troll, bear in mind that trolling outside of AFU is pretty weak”—but people did it anyway. One AFU regular who wrote in favor of trolling other newsgroups described the members of those groups as “foreigners” and “the filthy masses” (Tepper, 1997: 42–43). By these lights, trolling is what you do to outsiders, to foreigners, and to the masses, regardless of where you may find them online; you use it to punish them for not being insiders and to play a secret game of inclusion with your own community.
Perhaps, those who shared this insider perspective already foresaw what some onlookers missed, which is that trolling would continue to thrive even as most obvious “outsiders” online, namely novices to “Internet Superhighway,” faded as a presence in the online world. Writing in the late 1990s, Tepper (1997) suggested that trolling would soon die out as a practice, since the population of Internet users was growing so quickly that the mainstream should soon overtake the Internet’s old subcultures: as the promise of the Information Superhighway continues to gain in popularity, more people with little previous background in computing are thrown into a preexisting culture with rules they might not understand. As Usenet continues to change, trolling will probably become less and less effective as a community-formation technique. (p. 50)
More than two decades later, we can see that the opposite happened. Trolling became a familiar practice in every corner of the Internet; in the process, it became more vicious, less purposeful, and far less artful: an undirected attack on anyone who belonged (in the infinitely flexible phrase) to the filthy masses. Scholars and journalists have chronicled troll activity on message boards, social media feeds, game servers, Wikipedia pages, memorial websites, and elsewhere. 27 Though Cook et al. (2018) express uncertainty as to whether trolling is an established norm of behavior online, they concede that in the gaming community, at least, players are expected to treat encountering trolls as a normal experience: “It is clear that trolling is considered a negative phenomenon, but it is also an expected phenomenon, a ‘rite of passage’” (pp. 3334–3335).
However, although trolling’s definition has widened together with the range of behaviors that the term purportedly describes, we can note enduring patterns in the choice of trolling’s victims. Susan Herring et al.—who cite an early definition of trolling, written in 1996, which “refers to the generic target of trolling as ‘she’”—argue that forums and other non-mainstream online spaces hold more trolls precisely because those spaces “provide an arena for the enactment of power inequities such as those motivated by sexism, racism, and heterosexism.” 28 Jonathan Bishop (2014) likewise notes that present-day trolling often occurs “in places where there are vulnerable people, who might not be as skilled at Internet use” (pp. 9–10). The variety of behaviors that self-identified trolls engage in has expanded enough to throw the term’s meaning into question, but the targets—women, the vulnerable, and those who are accused of being novices or unskilled or outsiders on platforms where trolling takes place—have remained largely coherent.
One way to understand this transformation from a local to a wider social phenomenon is to consider, for exemplary purposes, the driving forces behind an event that one might call (with anachronism) the Great Lolcat Massacre. 29 In his classic essay, “The Great Cat Massacre,” Darnton traces the symbolic connections that gave meaning to a curious event in 18th-century Paris: a ritual massacre of cats by the apprentices working in a print shop. As Darnton shows, for the participants in this “massacre,” the cats signified, all at once though with the looseness of symbol, witchcraft, the devil, sexual taboo, the master’s household, and the master’s wife. The savage ritual “was meant simultaneously as a trial, a gang rape, a rebellion of the workers against their boss, and a carnivalesque kind of street theater, which the workers later repeated in the form of pantomime.” 30 We can find a curious parallel to this event, I argue, in the 1993 “invasion” of a Usenet forum called rec.pets.cats for the purpose of trolling its members. In tearing apart a community of cat lovers, in describing in gruesome detail the torture and execution of cats, the agents of this “invasion” were mounting a protest against the use of the Internet by those unlike themselves—which also entailed an attack on femininity, propriety, and dull domesticity. In the scale of the Internet’s history, the invasion of rec.pets.cats was a very minor event. Nonetheless, as an early act of mass trolling, whose agents were self-conscious of its status as such, this event offers a view of the processes by which trolling transformed from a niche form of boundary maintenance to a form of harassment endemic to the Internet at large.
The Great Lolcat Massacre
In 1993, the cat had not yet become, in the phrase of pop-culture journalists, the “spirit animal” of the Internet at large. Nonetheless, the choice of cat lovers as targets for this early act of mass trolling was not entirely coincidental. Rec.pets.cats, which dated from 1991, was a place for cat lovers to trade stories, questions, and tips about cats. As Stephanie Brail notes, the regular readers of rec.pets.cats were mostly women. They also happened to be, for the most part, newcomers to the Internet and “light” (i.e. merely occasional) users of computer networks. They visited Usenet to look up information and to chat with other light users, but not for much else.
31
A journalist of the period cited a regular user of rec.pets.cats as saying that many of the people in rec.pets.cats are not what you’d describe as typical computer people. “We have a lot of unsophisticated users in this group.” Many people are steered to the group when they are dealing with the grief of losing a pet, for instance.
32
The term that inhabitants of Usenet used for this kind of event was invasion: the regulars of one Usenet board descended, en masse, onto another board, posting flames and troll messages. In this instance, the hostile force in the invasion was the newsgroup alt.tasteless. In 1990, Usenet administrators had created alt.tasteless as a quarantine board; that is, people who were too disgusting or troublesome for newsgroups like rec.humor were directed to alt.tasteless, where they could exchange dead baby jokes without bothering civilians (Quittner, 1994). Quarantine boards are still in use today on Facebook and other connective-media platforms, where forums sometimes have split-off forums where users post content too tasteless for the residents of the main forum. 33 They may also be in use in gaming; among players of League of Legends, for example, rumors abound that the game’s creator, Riot Games, maintains separate servers to which trolls can be banished, and the game company Blizzard Entertainment has suggested that it has experimented with quarantine servers for trolls. 34 However, as the case study to follow demonstrates, quarantines have limited effectiveness when trolls work together in groups. As Synnott et al. (2017) noted, “studies of group trolling” have been rare in the literature on trolling in comparison with studies of the behavior and motivations of individual trolls; the capacity of trolls in groups to overcome institutional barriers to trolling suggests a reason for conducting more studies of group trolling (p. 72).
Within a few years, alt.tasteless had some 60,000 regular readers, according to the board’s FAQ. Contemporary observers seemed to take for granted that most of those readers were male. Certainly, parts of the board’s culture suggest a male readership. The readers of alt.tasteless loved to post disgusting facts to gross each other out, and some of their favored gross-out “facts” suggest limited knowledge of female anatomy. 35
In 1993, the regulars of alt.tasteless decided to find another newsgroup to invade. Because their mass trolling campaign was an act of boundary maintenance, they had to choose a group of victims who were pointedly unlike themselves. For the regulars of alt.tasteless, part of the joke—the “metonymic insult,” in Darnton’s phrase—was the opportunity to play on the cultural trope of the “cat lady”: out of touch, unthreatening, undesirable, unable to sustain human relationships and so, to replace those relationships, enamored with pets to the point of obsession. A writer for Wired magazine who discussed the event soon afterward understood the gender dynamic at play in their choice, although he communicates this dynamic only implicitly, through choice phrasing: One night last summer, the boys on alt.tasteless were feeling, well … if it were a Usenet group it might be called alt.restless. Maybe they were getting bored with each other. Maybe they craved the sensation of saying something really gross, and getting a Big Reaction. You know, something sisterly, like EEEEEEEEWWWWW GROOOOOSSSSS! You never get that kind of response on alt.tasteless. Someone—no one remembers who—suggested invading another Usenet group. A Usenet panty raid! The suggestion was well received by other a.t.’ers. But whom to raid? After much discussion, a likely target emerged: Rec.pets.cats. (Quittner, 1994)
The “panty raid” took the form of spamming the newsgroup with troll posts, which initially posed as serious queries from cat owners. One poster asked what he should do about his cats when he brought dates over, describing their smell and embarrassing habits in excruciating detail. Another asked how he could kill his girlfriend’s cat without her finding out. When the regulars of rec.pets.cats took the bait, the trolling escalated. Invaders posted queries about cooking cats, advice to kill cats with “multiple .357 copper-jacketed hollowpoints,” advice to nail cats to breadboards, and “articles about topics such as vivisecting the cat and having sex with its innards” (Brail, 1996: 152).
In keeping with the practice, borrowed from AFU, of posting shibboleths in or under a troll post that would alert insiders to the prank being played but leave a “trout” in the dark, the trolls in the cat newsgroup sometimes left clues by giving their (imaginary) cats names that corresponded to slang words in their own newsgroup for sexual acts or genitalia (Quittner, 1994). As with slang terms in ordinary language use, the purpose of these clues was to “form a sociolinguistic barrier within which insiders identify themselves through passwords” that outsiders find unfamiliar. 36 This is a behavior that trolls from our own time share, since, as Phillips has persuasively argued, memes serve as passwords for the users of forums like 4chan, establishing the terms of a shared mental world and its signifiers: “Recognizing a meme, remixing a meme, referencing a meme—these actions establish a set of subculture borders, thus providing a ‘meaningful whole’ to which additional signifiers may cohere” (Phillips, 2012a). Her emphasis lies with the sense of triumph that trolls attain using shared terms from their subculture as weapons against mainstream culture, as when trolls trick mainstream journalists into reciting memes unknowingly. We can add to her framework by saying that passwords, in this usage, extend the work of boundary maintenance, highlighting the difference between those who understand a term and those who do not.
Another unfortunate feature that the trolls in the cat newsgroup shared with present-day trolls was a readiness to escalate past the point of mere ridicule. One woman from the cat newsgroup—who took the post about killing a girlfriend’s cat seriously and wrote to the police—became the subject of “death threats, hate mail and harassing phone calls” (Brail, 1996: 152). Another woman, a software engineer, began receiving death threats: “I got mail from people telling me that they wanted to cut me up with knives, that they were going to tie me up and watch me squeal like a pig,” she told a reporter. “I got one message where the guy had enclosed my work address” (Quittner, 1994). Anecdotes like these are familiar features of news stories about trolling in our own time, which have described trolls sending death and rape threats, doctoring images to portray victims in grisly crime scenes, and publicizing the home addresses of victims. 37
“There are no women on the Internet”
Academic literature on trolling sometimes suggests that misogyny came late to the practice. For example, in her recent works on “gendertrolling,” or trolling that targets and attacks women because of their gender, the media scholar Karla Mantilla describes gendertrolling as “a relatively new kind of virulent, more threatening online phenomenon than the generic trolling described by Phillips.” 38 But in reality, women as a class became a target of trolling almost as soon as the trolling left AFU, and a major reason for this was that online identity was constructed as normatively masculine. 39 Well into the 1990s, cyberspace—to use a term that is losing popularity with the dissolution of the Internet’s boundaries—could seem to be a magic circle. 40 During the salad days of Usenet and the Whole Earth ’Lectronic Link, using the Internet required some technological savvy: users had to navigate using keyboard commands, and they needed to know the phone numbers of the servers that hosted the pages they wished to view. 41 In short, cyberspace required users to develop specialized forms of mastery, and in exchange it offered young men, in particular, a protected space within which to enjoy alternative forms of companionship, authority, status, even masculinity. 42 Some users built their lives around the Internet; some used it as a forum to discuss outside hobbies; some escaped to it as a private world that their family members barely understood. But as the media scholar William Mitchell argues, during this period, users of the Internet tended to share a tacit belief that the virtual world is a world apart and that the virtual and physical realms were separate, with a heavy gate between them. 43
As personal computers grew more user-friendly and the circle of Internet usage widened to include more of the general public, territory conflicts were inevitable. The World Wide Web, which Tim Berners-Lee first promoted on Usenet in 1991, dramatically improved the Internet’s ease of use by giving it a simple point-and-click interface. In the early 1990s, commercial Internet service providers such as America Online, CompuServe, and Prodigy started to sell prepackaged Internet connection services, enabling even self-professed technological dunces to get online. As Internet usage started to enter the mainstream, general-reader books about “cyberspace” began to appear in bookstores, a genre in which a major theme was the writer’s naiveté about technology. 44 Department stores sold tee shirts that read, “Roadkill on the Information Superhighway.” A perceived contrast between insiders and outsiders online was a discursive trope among both self-identified insiders and self-identified outsiders.
This influx of new users seems to have registered among older online communities, in part, by producing a perceived need for boundary maintenance. Contemporary commentators remarked on both the hostility of online discourse and the perception that hostility was meant to constitute a subcultural style that aimed to repel outsiders. The journalist John Seabrook, in his own book on exploring cyberspace, described finding a peculiarly combative tone in much of the chatter online. One of his acquaintances commented, “There is something brutal and predatory about much of what goes on on the Internet. There is a kind of smart-ass style one must either learn to ignore, or capture and exploit for one’s own purposes.” Another suggested that this atmosphere reflected, in part, a reaction against the swarms of newcomers entering the forums: There is an air of preestablished hierarchy there—if you’re new to the Net, or even to a particular group in the Net, you don’t belong a priori. As a woman, I have encountered an additional barrier; the net is heavily male and women who want to play with the big boys either have to be ultra tough-talking—“one” of the boys—or else play off as coy, charming, “little-ol-ME?”-feminine. (Even geeks have fantasy lives, I suppose.) Or use a male-neutral alias with no one the wiser. So part of the boys’ club, I imagine, is the smallness, the selectivity—the geek elite, if you will. For more than a decade these guys had their own secret tin-can-on-a-string way to communicate and socialize, as obscure as ham radio but no pesky FCC requirements and much, much cooler. But then the Internet—their cool secret—started to get press … Imagine these geeks, suddenly afraid that their magic treehouse was about to be boarded by American pop culture. It was worse than having your favorite obscure, underground album suddenly appear on the Billboard charts. (Seabrook, 1997: 103–105)
Ultimately, the “cat massacre” that took place on Usenet in 1993 can usefully be seen as an event that the agents who brought it about purposefully made exemplary and which brought together, as part of that exemplarity, many of the themes that drove early Internet culture: the clash between “the Internet” and the mainstream, the precarious status of women on the early public Internet, and the use of hostility as a form of boundary maintenance. 45 The cultures of trolling that grew and flourished, in the late 90s and early oughts, in the forums of Something Awful and 4chan borrowed more directly from the playbook of alt.tasteless and similar communities than from the more temperate, even artful, trolls of AFU. But since its inception, an important function of trolling has been to police online communities and repel perceived outsiders—through either exposure or outright harassment.
A full understanding of the shifting history of trolling as a practice requires us to consider what has changed about the perceived outsiders of the world of new media, as well as what has, discouragingly, remained constant. Since the early years this article documents, the label trolling has widened to accommodate other communities, platforms, and behaviors. In the field of gaming, for example, trolls are a familiar phenomenon; self-identified trolls often talk abusively, kill people on their own teams, and give bad advice to new players (Cook et al., 2018: 3323). More troublingly, trolls in gaming have inundated women game developers with hostile messages and even rape and death threats, apparently seeking to drive them out of the industry. 46
In fact, few platforms online can claim to have never experienced troll activity; trolls have vandalized Wikipedia pages; flooded Twitter, Facebook, Tumblr, and YouTube feeds with abusive comments; flagged the Kickstarter pages of victims for removal; and organized distributed denial of service (DDoS) attacks against websites, often focusing on targets who were women or members of other minoritized groups. 47 Trolls today may act alone or, in the fashion of alt.rec.tasteless, in mass “invasions”; for example, in 2008, trolls from 4chan’s notorious /b/ board “invaded” the hip-hop news website SOHH.com and flooded it with racist text and images. 48 This is a wide and varied field of activity, and its oft-competing elements highlight an issue that has long challenged scholars: how to define trolling as a practice. One aim of this article has been to establish the concept of boundaries as an important component of that definition since the time of trolling’s origin. Although the role that boundary maintenance plays in trolling today is beyond my scope here, the persistent targeting of women and other minoritized populations as victims of trolling suggests that research in this field may benefit from examining how trolling behaviors continue to assert and, so to speak, militarize the boundaries between in-group and out-group populations on the Internet. 49
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
