The dinner was cordial. But soon after Stephanie Guthrie and graphic artist Gregory Alan Elliott met in 2012 to chat about his designing a poster for Guthrie’s grassroots feminist organization, their acquaintance took, as Guthrie later put it, “a bad turn.” The whole thing would eventually bottom out in a Toronto courtroom, with Guthrie and another Toronto activist, Heather Reilly, accusing Elliott of criminal harassment via Twitter, and Elliott emerging as a #GamerGate poster boy, standing up for free speech in the face of #FascistFeminists.
The details of that “bad turn” will sound familiar to a lot of women who spend time online. After the poster deal fizzled, Elliott kept tweeting at Guthrie with invitations to meet again: offering her a ride to the airport, suggesting he drop by with a bottle of booze, asking if she’d like to drive up north and stay with him at a mansion overnight. In retrospect, Guthrie testified later, he’d seemed creepy in person, and when she wheeled back through his Twitter timeline (weird sex stuff, breast cancer conspiracies), she was similarly squicked out by his online presence. Finally, when the pair came to Twitter blows over a proto–#GamerGate microcontroversy, Guthrie blocked Elliott and helped rally other local feminist activists—many of whom had also sniffed negative vibes wafting from his account—to attempt to shoo Elliott away from their online spaces.* Instead, Elliott dug in. Over the course of several months, he called Guthrie and company “cunts” and “bitches.” He also trashed Guthrie’s “shit music,” lewdly ridiculed Reilly’s “fat ass,” and didn’t stop mentioning the women on Twitter, obliquely referencing their whereabouts, or chiming in on their #WiToPoli hashtag when they asked him to leave them alone. You know: Twitter stuff.
While this type of social media persecution isn’t unusual, online harassment is so rarely investigated by cops and prosecuted in court that the Gregory Alan Elliott affair represents a landmark case. On Friday, Ontario judge Brent Knazan released an 85-page decision that illustrates just how bananas it is to translate the criminal code to a social media property. It took Knazan three hours to read the whole thing out loud in front of a packed courtroom. At the end, he declared Elliott not guilty of criminal harassment. But first things first: “I begin,” he said, “with what Twitter is.”
In Canada, criminal harassment occurs when you repeatedly follow someone from place to place, repeatedly communicate with them “either directly or indirectly,” stake out their house or office, or engage in “threatening conduct” against someone in a way that causes them to “fear for their safety.” The elements of a harassment charge vary by state and country, but Canada’s standard mirrors the one at play in many U.S. jurisdictions. It’s a criminal offense that hinges on a meeting of the minds between perpetrator, victim, and judge. Prosecutors must prove that the perpetrator purposefully (or in some cases, just recklessly) harassed the victim. The victim must feel so harassed that she fears for her safety. And the judge must deem her fear reasonable given all the circumstances of the case.
That meeting of the minds didn’t happen here. Though Knazan agreed that Guthrie and Reilly felt they had been harassed, and acknowledged that Elliott knew or should have known that he was harassing them, he did not agree that the women feared for their physical or psychological safety as a result of that harassment. For one thing, they were being very saucy on Twitter. Guthrie’s “retweeting of forceful, insulting, unconfirmed, and ultimately inaccurate attacks” against Elliott hardly gave the impression that she cowered in his presence, Knazan wrote. Neither did Reilly, when she retweeted a comment telling Elliott to “disappear up your own arse you supercilious fart golem.” As Knazan put it: “This is the register of language sometimes used by both those arguing for an open Twitter and those wanting to prevent further attacks on women.” You say “fat ass,” I say “fart golem,” and as far as the court’s concerned, it’s not harassment when everybody’s spitting equally crude insults.
The more Guthrie, Reilly, and their supporters fought back against Elliott, the more Elliott’s own tweets about Guthrie and Reilly could be perceived as his mounting a legitimate defense. The idea that Guthrie “could control people’s non-threatening, non-sexual use of her handle and hashtags that she used” is “at least arguably incompatible with Twitter,” Knazan wrote, particularly given that she “was leading or at the least playing a major role in calling him out for allegedly harassing women online.” Elliott couldn’t be pinned for harassing these women because he was engaging in a legitimate debate about whether or not he was committing harassment.
These, then, are the guidelines provided by the Ontario Court of Justice on how a victim of online harassment ought to behave. She should announce to the aggressor that she feels harassed. And then she should shut up.
Reading the decision, I kept wondering how the case would’ve played out differently if a smart teen—or, really, anyone who’d tooled around Twitter for two weeks—took the judge’s place. Gamergate target Zoe Quinn once compared explaining online harassment to cops to “trying to push cooked pasta through the eye of a needle.” The judge in this case gleaned his understanding of Twitter’s function and norms through statements from four people: a police constable he says was “qualified as an expert in Twitter,” testimony from Guthrie and Reilly, and comments made by Elliott himself, who declined to testify on his own behalf but “expressed his views about Twitter in some tweets that he sent.” The parties did not always agree, and when a disagreement arose—like when Guthrie and the cop differed in their explanation of which Twitter users can read a tweet that starts with a period—Knazan couldn’t just resolve the disagreement by, like, Googling the correct answer. The disputes just hung in the air, mysterious Internet-y computer thingies the court could never hope to crack.
Also vexing: The court often assessed the wrong tweets, and missed the right ones. Using software from Sysomos—a service cops claimed was capable of resurrecting deleted tweets—police searched for tweets in which Elliott cited the women’s handles, or they cited his. They also grabbed tweets from Elliott that mentioned hashtags created or frequented by Guthrie and Reilly. It was an imperfect system. Tweets and replies that provided context were often omitted from evidence. One contentious Elliott tweet never surfaced in the courtroom, for reasons that were not explained. And in one instance, Knazan wrongfully attributed a tweet to Elliott that he never wrote. That vile tweet—addressed to Guthrie and a couple of other users, it advised them all to “Enjoy your AIDS … faggots”—was actually sent by an Elliott imposter who’d created a parody account to crudely mock him. The fake account copied Elliott’s profile photo, and secured a handle just one letter off from Elliott’s own, and Knazan fell for it. On the strength of one fake tweet, he described Elliott as “homophobic” in his ruling four times.
Knazan did come to understand one important thing about Twitter. The service’s standard recommendation for dealing with harassers—just block them—is an “ineffectual step” that “does not work,” the judge acknowledged. When you block another user on Twitter, you lock them out from viewing your timeline while they’re logged into the service, and you prevent their tweets from popping up in your mentions. But seasoned harassers can slither around those barricades, getting your attention by provoking your friends, tweeting to a hashtag you use for work, or just shouting so loudly that their tweets begin to affect your reputation, whether you saw them or not. On this point, Knazan’s decision shrugs. “Once someone creates a hashtag, anyone can use it,” he decided. “Everyone has to be able to use it freely; anything less will limit the operation of Twitter in a way that is not consistent with freedom of expression.”
Of course, women who find their local feminist activist hashtag derailed by a serial harasser are no longer able to use it so freely. As a consequence of the failures of Twitter and law enforcement to remedy the online harassment problem, some women and other marginalized groups have turned to ad-hoc—some might call them vigilante—solutions. They tip each other off to harassers lurking in their midst, trading intel to protect each other from sketchy professional and social situations. Or they hit back against harassers on Twitter, calling them out for bad behavior in an attempt to change the culture, protect their own reputations, and gain some sense of control over an impossible situation. “Exposing and criticizing harassers can serve as a powerful deterrent,” University of Maryland law professor Danielle Keats Citron wrote in her 2014 book Hate Crimes and Cyberspace. “Talking back allows victims to tell their side of the story. In explaining what happened, they might combat the assumption that they warranted the harassment. Counterspeech could also help victims gauge threats made against them” and “gain confidence about taking matters into their own hands.”
Now, Knazan is telling victims that these trusted tactics may disqualify them from legal victory. Women who lash out at harassers online aren’t necessarily less scared than those who don’t. In many cases, they’re distressed enough to rally friends and followers to come to their aid. But the same Twitter norms that make it quick and easy to lob abuse at a target also goad that very target into hitting back in a tone that’s cool, confident, and maybe a little bit nasty—and not nearly quivering enough to build a criminal conviction.
When it comes to online harassment, this decision seems to say, you can’t have both Twitter justice and real justice. You have to pick. In most cases, though, you’re still likely to receive neither. Lodging a harassment complaint with the cops remains, more often than not, a surefire way to waste an afternoon. And lodging a complaint of harassment on Twitter is the easiest way to receive more harassment on Twitter.
Over the past several years, Twitter has made some efforts to remedy its harassment problem. It’s admitted it “sucks” at dealing with abuse, added a handy “report abuse” button, introduced a feature that helps users report potentially criminal tweets to law enforcement, and banned some of its most notorious trolls.
And yet abuse thrives on the network. Now that Elliott’s been acquitted, a hive of supporters with anime avatars and anonymized handles have stepped up to continue the harassment campaign. Guthrie has been inundated with threats of rape and torture. The hashtag, #WiToPoLi, that Guthrie carved out for women in Toronto politics, has been trashed by trolls, who have spammed the hashtag with cheers for Elliott, anti-feminist memes, and porn. What can be done about this? This decision raises the possibility that the answer is absolutely nothing, because Twitter was designed to work this way.
Consider what happened when Reilly complained to Twitter about Elliott’s behavior back in September 2012. Twitter reps declined to take action against Elliott’s account, instead encouraging her to resolve the issue through the legal system. But now, the court is looking back to laissez-faire Twitter to set the standard for acceptable online behavior. Going forward, this ruling will influence how other courts gauge what constitutes normative discourse on the platform. If, in the eyes of the law, the harassment of Stephanie Guthrie is par for the course, then we can look forward to many more years of the status quo—women getting harassed, and everybody saying that it’s somebody else’s problem.
*Correction, Jan. 26, 2015: This story originally misidentified a circa-2012 Twitter altercation as a “#GamerGate microcontroversy.” The #GamerGate hashtag was not in circulation until 2014.