Twitter's Harassment Problem Is Baked Into Its Design

The first recorded example in Western literature of men telling women to shut up and stay in the house, writes classicist Mary Beard in her 2014 essay, “The Public Voice of Women,” is in the Odyssey. Not-yet-grown Telemachus tells his mother, Penelope, to “go back up into your quarters, and take up your own work, the loom and the distaff ... speech will be the business of men, all men, and of me most of all.”

As Beard noted in her essay, centuries on, the voices of women are still considered illegitimate in the public sphere, including the new spaces of social media. That manifests as verbal harassment, death threats, and doxing online; as complaints about the sound of women’s literal voices on the radio, giving talks, or in podcasts; as sexual harassment in the workplace; as catcalls on the street. All of these can be seen as ways to drive women out of the public sphere, and back to their proper domain of Kinder, Küche, Kirche (children, kitchen, church). On Friday, many Twitter users boycotted the platform, in response to the suspension of the actress Rose McGowan’s account for speaking out about sexual harassment by the film executive Harvey Weinstein. The driving force for the boycott was women outraged that hate speech, including misogynist and racial harassment and threats, routinely go unchecked, and yet McGowan’s account was suspended.

These women did indeed remove themselves from a public sphere. Twitter, with its more than 300 million active monthly users, is a communal space in a new and extraordinary way that’s driven by the specific technological decisions of the site, which carry with them specific affordances. “Affordances,” a term popularized in the world of design and user interaction by Donald Norman, is a way of describing the perceived possibilities of how the user can interact with the product. These affordances shape how users behave.

Much of the power of Twitter comes from retweets, which can carry the words of a user to an audience far beyond their own followers (for comparison, see Instagram, where no such function exists—it makes it much more difficult for a specific image to “go viral” on the site). But retweeting also allows for what social-media researchers such as danah boyd and Alice Marwick refer to as “context collapse”: removing tweets from not only their temporal and geographic context, but also their original social and cultural milieu, which is very different from most public spaces. I described it to a friend once on a New York City subway—“we’re talking in public, in that everyone near us in this subway car can hear what we’re saying, but that’s a very different ‘public’ than hearing ourselves on NPR tomorrow.” While readers may literally know nothing about the poster or the context except for what is said in that one tweet, they can still just hit “reply” and their response will likely be seen by the poster.

While nothing is stopping people from finding out more information before responding, the clearest affordance Twitter has is for these “drive-by” responses (I’ve been mansplained to by many people who I presume haven’t even looked at my bio to see the “engineering professor” there before trying to school me on my research field—per Telemachus, “of me most of all”). This amplification and context collapse, coupled with the ease of replying and of creating bots, makes targeted harassment trivially easy, particularly in an environment where users can both mostly live in their own ideological bubble by following people who share their views, however abhorrent, and who can easily forget that there is a real person behind the 140 characters of text.

So while Twitter may consider itself to be merely reflecting the discourse, these technological affordances ease the way for certain types of hostile behavior. If you think of the experience of the generalized, systemic misogyny and racism of our culture as being bathed in sunlight on a scorching hot day, Twitter might say it’s just a mirror. But it’s actually handing out magnifying glasses that can focuses the already painful ambient sunlight into a killing ray. The targets of this ire, in our society and on Twitter, are disproportionately not just women but people of color. (Imagine how Telemachus would have responded if, rather than his mother, one of the non-Greek household slaves chose to speak up in visiting company.)

One of the most profound social changes of the last few decades is opening up public discourse to a broader range of speakers than ever before, and social media has been a large part of that. The specific affordances of Twitter make it powerful—it can amplify marginalized voices but it can also amplify harassment. Friday’s boycott was intended to be a unified stand against that.

But the point of harassment is to shut women up, either by self-censorship through fear or by driving them away from Twitter, making it simply the newest wrinkle in that long history of exclusion from public spaces and conversations. Many women, especially women of color, therefore found a protest that mandated their silence to be ironic, if not outright misguided: It takes a certain amount of social power to genuinely believe that your absence would be remarked upon and lamented. After 3,000 years of denying the public sphere to all but a small set of voices, some of the new voices are rightly considering their presence to be a sit-in, an occupation, and they are rightly refusing to be driven away. Ultimately, if Twitter wants to be the public sphere, it needs to act like it, by working to create an environment where all voices can be safely heard. Twitter’s social problems are exacerbated by the affordances of technology; they’ll need to bring both ongoing human effort and better design decisions to improve the experiences of marginalized people, and therefore everyone, in their public sphere.