In 2016, the Toronto Star‘s long-serving film critic, Peter Howell, suffered an internet beat-down to remember. In an attempt to praise the film Moonlight, he called out director Barry Jenkins’s resistance to any kind of “coatswitching”—cleaning up street vernacular, to better gain acceptance with mainstream society. There was just one problem: The term isn’t “coat switch.” It’s “code switch.” Twitter flayed Howell, then boiled his bones for soup. The hashtag #coatswitch almost immediately became a byword for racial cluelessness. Thus did a well-meaning attempt to support a black filmmaker end up making its author look like a fool.
This might seem like a goofy accident, readily avoided by anyone else keen on the mores of social media. But a surprising new study sponsored by the Knight Foundation, out today, reveals that the #coatswitch incident isn’t the exception when it comes to how the media deals with minority communities on Twitter—it’s the rule. And when media outlets do try to cover minority groups, they actually drive up disdain on the platform. Not only is hate-tweeting articles a real thing, it may be the dominant mode for how minorities talk about media coverage about their communities.
The study was authored by Deen Freelon, a media studies professor at the University of North Carolina, Chapel Hill (and son of architect Phil Freelon), and three co-authors: Lori Lopez, University of Wisconsin, Madison; Meredith D. Clark, University of Virginia; and Sarah J. Jackson, Northeastern University. Freelon led the quantitative part of the study. The ingenuity lay in the setup. To figure out the boundaries of three virtual Twitter communities—Asian Americans, African Americans, and feminists—his team pored over the hashtags used by community leaders and scraped news coverage mentioning both those groups and also containing a hashtag. Freelon then turned those terms into a cloud of affiliated hashtags that provided a rough map of the community.
The next step was to gather the actual tweets those communities were writing. To do that, Freelon simply bought them from Twitter: Some 45 million tweets containing the hashtags he’d gathered. The penultimate step was to winnow those tweets down into those mentioning news stories while also offering some kind of analysis. In the final step, his research assistants read 1,000 instances of media critique from each group—Asian-Americans, African Americans, and feminists—and gauged the balance of positive to negative commentary. “It turns out, the most shared outlets were also the most negatively perceived,” says Freelon. “It’s the scale that’s really surprising.” Put another way: Twitter really hated the mainstream media—and they hated TV news and CNN most of all:
As part of the project, the product-design firm Postlight—founded by coder and journalist Paul Ford with technologist Rich Ziade—created a website for sharing all the results on Twitter. Gina Trapani was the project’s lead engineer. Though she wasn’t a study author, she had to immerse herself in the data. As she points out, “The media uses Twitter to write stories and invent heat, but there’s a disconnect in how the community talks and how the media talks.” That’s a familiar point for anyone who has ever hated a media story, but perhaps for the first time, that disconnect—or media cluelessness, if you’d like—can be quantified and charted. Take this example, of the disconnect between Asian-American Twitter’s use of the hashtag #freepeterliang—a reference to Peter Liang, an NYPD cop whose gun discharged in a dark stairwell and killed Akai Gurley, leading to Liang’s trial for manslaughter:
You can see a spike, representing a grassroots mobilization effort, which was nonetheless almost wholly absent from media coverage.
Freelon’s co-authors point to some fascinating dynamic in play behind all this mistrust. The communities themselves have become wary of any news coverage; frequently, they won’t speak to the media, because they presume they’ll be misrepresented. The obvious solution then would be to have reporters with more concern for the authenticity of their reporting—or who themselves are members of the communities they’re covering.
But the more unsettling conclusion may simply be that the dynamics are haywire all around. Parachute in for a story on #blacklivesmatter, and you’re not likely to garner much trust for future stories. “The community doesn’t view itself objectively,” Freelon points out. “When you have supposedly objective reporters, there’s going to be friction.” Put another way, tell stories that minorities love, and you might be sacrificing your objectivity.
Another tension lies in the media platforms themselves. It’s surely no coincidence that the most loathed media outlets in the chart above are TV news channels. By the nature of how limited the media is—by airtime, most of all—the possibility of nuance is almost nil. By contrast, one of the only media outlets to gain positive sentiment was Buzzfeed. It so happens that Buzzfeed, because it lives on the web and spreads through social media, has no limits to the stories it covers; it, therefore, has the wherewithal to cover minority communities without having to “explain” them to outsiders.
To be clear, only about 5% of media tweets in the study actually had any sentiment in them at all; most were mere retweets without comment—that is, the sharing of news that presumably was useful in some way or another. The media may, in fact, have been doing its job, even if Twitter only riled people up to comment when they hated something.
And yet it’s hard not to think that this overwhelming negative sentiment when people do say something spells trouble for media. TV news still has the greatest reach of all. If it’s broken, what replaces it? In 2018, given everything that’s happened in the last three years, it’s hard to believe that social media would serve us better. Maybe #coatswitching is a sign of tentative progress, no matter how embarrassing.