The first time Priyanka Singh suffered the wrath of Twitter’s brutish underbelly, she cried herself to sleep. A 27-year-old from Delhi working in education publishing, Singh is also an atheist and an outspoken critic of India’s casteism and rape culture — a choice she pays dearly for in her Twitter mentions. Her abuse on the platform began in February and in the ensuing months has become routine: rape threats, body-shaming, egg-avatared strangers wishing her and her family harm.
The harassment took a toll. “I became aggressive online and defensive at home,” Singh told BuzzFeed News. Her relationship with her then-boyfriend deteriorated as she obsessed over the vitriol and invective directed at her on Twitter. It consumed her time, her energy. At work, Singh was preoccupied and easily agitated. When a troll alerted her employer to some of her tweets, her colleagues and managers urged her to delete her account, suggesting that Twitter “wasn’t real life.” Clearly, fighting back wasn’t helping.
“It only adds to the humiliation when you pour your heart out and you get an automated message saying, 'We don't consider this offensive enough.’”
Meanwhile, on Twitter, Singh was a “cunt." Or a "feminazi" or a "whore and jihadi." When her Twitter tormentors discovered Singh had been diagnosed with bipolar disorder, they attacked her for that, too, tweeting, “Your brain is fucked up.” Eventually Singh began cataloging her abuse. She diligently took screenshots of the harassing tweets and reported them to Twitter. Each time (save for once in August when Twitter suspended a lone tormenter’s account), the company’s response was the same: “We’ve investigated the account and reported tweets for violent threats and abusive behavior, and have found that it’s currently not violating the Twitter rules.” Singh was devastated. “It only adds to the humiliation when you pour your heart out and you get an automated message saying, 'We don't consider this offensive enough,’” she said.
So, like countless victims of harassment, Singh stopped trying. She made the tedious task of troll-blocking part of her daily routine, and tried to shrug off the abuse.
And on Twitter, the trolls kept on trolling.
Over a decade into Twitter’s existence, harassment of the sort Singh suffered is a well-documented and common practice on the platform. But Twitter’s protocols for reporting abuse and addressing it remain largely opaque. Until 2014, simply reporting abuse was a cumbersome process, requiring users to fill out an arduous, but thorough nine-part questionnaire. Today, the process is somewhat more streamlined. Users can now report abuse they see on behalf of others, and in April of this year the service finally allowed users to report multiple tweets at once. But despite these tweaks, countless targets of abuse have found that their reports — no matter how vile their content — don’t meet Twitter’s standards for harassment. Users receive no justification besides a “thank you” and a prompt to contact law enforcement if they feel they are in danger.
Take Kelly Ellis. Ellis, a software engineer at Medium with a verified account and over 11,000 followers, was mercilessly tormented by a single Twitter user for weeks this past summer. By the time her antagonist, @fredcarson9151, concluded his 70-tweet barrage of abuse, Ellis had listened to her accuser publicly express a desire for her to be raped. She’d also been berated as a “psychotic man hating ‘feminist’” and told that “men don’t need her opinion” on matters of sexual assault.
Ellis reported the abuse and Twitter investigated it. But the conclusion of the company's investigation was perplexing. Ellis said Twitter told her that the behavior she reported did not violate its rules, which explicitly state that one may not “threaten other people on the basis of race, ethnicity, national origin, sexual orientation, gender, gender identity, religious affiliation, age, disability, or disease.” BuzzFeed News documented Ellis's experience in a story widely shared across social media. Two hours after its publication, @fredcarson9151 was banned from Twitter.
Singh and Ellis’s stories are common; despite vigilantly reporting abuse to Twitter, countless victims of harassment allege they see little in the way of an effective response from the company. Attacks recur, mentions remain toxic. To better understand how Twitter handles reports of harassment, BuzzFeed News invited readers to complete an online survey about abuse. More than 2,700 users responded. The results suggest that an overwhelming majority of reported abuse requests ended with Twitter taking no visible action toward the offending account.
Before we dig into the results, a few important caveats: Our survey was distributed primarily through BuzzFeed’s social channels, largely on Twitter and via a link in a BuzzFeed News post. As such, the 2,702 respondents are likely not reflective of Twitter’s users overall. Respondents who participated in this survey were largely English-language speakers. By virtue of the subject matter, participants were probably more likely to have experienced abuse on Twitter than the average user of the service. (That said, respondents were given an option to state that they had not experienced abuse on Twitter.) BuzzFeed News also conducted follow-up interviews with a number of respondents.
Also worth noting: We've seen evidence that Twitter occasionally takes action against tweets reported as abusive without revealing that it has done so. But this leaves targets of abuse who reached out to the company in the dark about whether their appeals for help were heard. In some instances, Twitter has publicly revealed that it deleted tweets after a high-profile outcry. Just yesterday, Binyamin Appelbaum, a New York Times correspondent with more than 40,000 followers, retweeted some particularly horrific anti-Semitic remarks that had been directed at him and copied Twitter's CEO, Jack Dorsey. Afterward, the company deleted the tweets and Dorsey notified Applebaum of the deletions publicly on Twitter.
All this said, the responses to our survey do offer a human window into Twitter’s underbelly of abuse, providing not only harrowing examples of harassment, but data on how it is handled — or not — by the social network.
According to survey respondents, Twitter’s most common response to an abuse report is inaction. Roughly two-thirds of people BuzzFeed News surveyed who’d received an abusive tweet said they logged it via Twitter’s violation reporting page; nearly 80% of our respondents said they reported an abusive tweet on behalf of somebody else. But very few reported ever receiving a response from the company.
46% of respondents told BuzzFeed News that the last time they reported an abusive tweet to Twitter, the company took no action on their request; their only recourse was to personally block the offending account. Another 29% who reported abusive tweets said they never heard anything back at all. And 18% of those who reported an abusive tweet said they were told that the tweet did not violate Twitter’s rules, which explicitly forbid violent threats, harassment, and hateful conduct. In just 56 instances (2.6% of the time) respondents said Twitter deleted the offending account, and in 22 instances (1% of the time) respondents said Twitter issued a warning to the user who’d sent the tweet. Of the 2,115 people who responded to this particular survey question, just five individuals reported being contacted by a Twitter representative to discuss the abuse they reported.
"Safety is our top priority"
On Sept. 15, Twitter declined BuzzFeed News' request for an executive interview on the subject of harassment. In response to a Sept. 19 letter detailing the findings of our survey, Twitter's head of communications, Kristin Binns, provided the following comment:
"Safety is our top priority — we're building better tools and processes every day. We can't comment on a third-party survey, and its anonymous nature makes it impossible to verify data or corroborate response. While we know there’s still much to be done, we’re making progress toward giving people more control over their Twitter experience and to better combat abuse."
Of our 2,702 respondents, 1,478 (55%) said they had been the target of an abusive tweet or Twitter direct message. Of those who answered yes, 18% said they had been harassed in just the past week, while 26% said they had been harassed at some point within the past month, but not the past week.
Examples of harassment from those surveyed frequently appeared to violate Twitter’s rules, which prohibit tweets involving violent threats, harassment, and hateful conduct. Twitter’s rules explicitly state that one may not “threaten other people on the basis of race, ethnicity, national origin, sexual orientation, gender, gender identity, religious affiliation, age, disability, or disease.” Some examples — which were submitted anonymously to BuzzFeed News via our survey — include:
- “Someone spread a video around of an unnamed girl taking part in perverse activities with an animal, and captioned it with my underage friend's name. It was taken down the next day, but the damage was done.”
- “Someone took a google cached but long deleted photo of me from my fb, photoshopped it on to a naked body, and posted it alongside my real name, openly calling me a dyke (I'm not out to anyone except a few very close friends). That's one instance in a series of nearly 500 abusive tweets from 50 different troll accounts (likely created by same person or few people, based on how they were posting).”
- “I get a lot of messages threatening rape, men trying to find my physical location and disseminating it so I can be raped, threats of murder, people wishing miscarriages on me, to be sent into war zones to be raped and murdered.”
- “Pictures of my family, my address, my employer, threats to 'rape this bitch with a cactus,' threats to 'take a crowbar to that pretty throat,' insinuations that I am only academically successful in a male-dominated subject because I am attractive.”
- “A map to my home was circulated. There's nothing left of my life for them to dox.”
- “After a witty retort (that was polite) and me wishing him a good day he said 'what would make his day better would be to do this' and inserted a graphic video of a jihadist chainsawing a kneeling man's head off.”
- “A picture of a person pointing a gun at me; telling me they'd bury me out back; calling me a cunt and telling me to stop talking out of my clit; posting my full name (I am anonymous on Twitter... or was).”
- “My meant to be private pictures (nudes) posted for months non stop, my drivers license posted, my address posted, my work place address posted, my parents info posted, threats to kill me and my kids, Photoshopped pics of me with gunshot wounds through my head and chest, inciting people to swat me, my naked pics tweeted at two workplace Twitter accounts, it goes on and on. Pedophile tweets about my kids.”
As the above examples demonstrate, respondents of the BuzzFeed News survey reported experiencing a broad spectrum of targeted abuse on Twitter. Over 67% of respondents described the tweets they received as misogynistic in nature. Nearly 30% reported being targeted with homophobic slurs. One-quarter said they’d been subjected to racist epithets, one-quarter reported death threats, and one-quarter reported tweets encouraging them to kill themselves.. Nearly 20% of respondents reported being threatened with rape; another 20% said they’d received tweets threatening to publish their private information, photos, or videos.
While the examples vary in frequency and severity, the trauma of abuse in combination with Twitter’s response as described above has created a growing feeling of futility among BuzzFeed News survey respondents. “[Twitter] can be presented with multiple examples of a user violating [terms of service] and STILL say the account broke no rules,” one wrote. “It's disheartening. You feel utterly vulnerable. I've never been so afraid to upset the wrong person.”
“They care more about copyrights than they do about people.”
“Twitter claims to be balancing concerns for free speech with comfort of users, but the fear of harassment … or even worse, doxxing routinely causes me to self-censor,” one wrote. Another respondent agreed: “In essence, Twitter's protection of hate speech in the guise of free speech infringes on my own free speech.” The self-censoring effect can create a culture of fear for frequent harassment targets like women and people of color. “While I personally haven't been harassed, I HEAVILY censor what I say on Twitter because I've seen how abusive people are towards women there,” one respondent wrote.
Other respondents suggested that Twitter’s harassment problem is systemic and worsening. “Having been on Twitter since it started, I have seen a drastic change in the past year, especially in terms of harassment. It has become 'normal,' almost the purpose of the platform itself,” one wrote. “There seems to be absolutely no way to take action against it. Reporting used to have some impact, lately I don't even get responses to reports,” said another.
Particularly frustrating for users is Twitter’s penchant for quickly taking down content that violates copyright for its media partners. This summer, victims of Twitter abuse criticized the network for suspending accounts posting Olympics-related content without permission while not responding to their claims of user abuse. “[Twitter] took down a video I posted about the Olympics about 12 hours after I published it. But it has never taken down a harmful tweet or an account reported by me,” one respondent said about the apparent double standard. “They can delete accounts sharing broadcast content in minutes, but abusers are rarely if ever sanctioned, and almost never banned,” a second surveyed user wrote. Others surveyed were more blunt: “How much money do I need to have to protect myself and my friends? Olympics money? NBC money? Taylor Swift money?” And: “They care more about copyrights than they do about people.”
Though the hundreds of written responses collected by BuzzFeed News vary, a majority of those surveyed expressed concerns about Twitter’s internal commitment to abuse and to the platform’s average, non-celebrity users. For many surveyed, years of inaction have led to a lack of trust on behalf of victims of harassment and an expectation that the problem, now systemic, will never be resolved. “The only thing I get from reporting people is the knowledge that I took an action to stop it,” one user told BuzzFeed News. “I don't expect anything back from Twitter anymore, they don't care.” Another respondent suggested that “[harassment is] just clearly something that Twitter has grown to tolerate, despite what they might say, which is wrong and it only fuels hate.” While a third summed up their experience by saying, "I constantly feel disrespected by Twitter, as they ignore threats made against me, but have the resources to support me."
“I love Twitter and won't leave, but many others just can't carry on with the abuse they get.”
Despite the vitriol, some respondents still expressed affection for the platform. “I love and use Twitter every day, since 2010, but I definitely have never thought that Twitter has taken harassment seriously,” one respondent wrote. Others are worried that the lack of improvement could lead to the eventual silencing of some of Twitter’s most vital members. “It's out of control and delegitimizing Twitter as a channel. I don't want to lose the diversity of voices, but can't imagine some people staying on the platform given how they are treated. It has also allowed racism and misogyny somehow become openly tolerated.”
Repeat abuse victims are left with a bitter choice: stay and be repeatedly targeted or leave the network and all the connections behind. As Ellis noted, the decision to leave is not only a victory for Twitter’s most vile communities, but also effectively silences the voices of some of its marginalized users. And in cases where Twitter is vital to the the jobs and well-being of others, it is a painful alienation. “It sucks having to choose between feeling safe & avoiding vitriol w the ugliness of rape, vs keeping social connections that matter to me,” Ellis tweeted after Twitter rejected her abuse reports on the grounds that her harasser did not violate the company’s rules.
For Twitter, which has long struggled with user growth issues, its failure to handle abuse reports could drive away some of its most passionate power users. “It's ruining the platform,” one survey respondent noted. “I love Twitter and won't leave but many others just can't carry on with the abuse they get.”
You'll find full responses to our survey below.