After Twitter permanently suspended Donald Trump’s account, earlier this month, the reactions were quick, ubiquitous, and mostly predictable. Many of the takes seemed canned, the way an obituary of a terminally ill celebrity is often pre-written. On the Trump-apologist right, the suspension was denounced as Orwellian tyranny, deep-state collusion, or worse. (Glenn Beck, during a segment on Tucker Carlson’s Fox News show, compared the Trump ban and other Big Tech crackdowns to “the Germans with the Jews behind the wall. They would put them in the ghetto. Well, this is the digital ghetto.”) Among Trump’s opponents, reactions were more mixed. There was a good amount of gloating—the only thing easier than kicking a man when he’s down is dunking on an account after it’s locked—but the Schadenfreude was tempered with caution. Jameel Jaffer, the director of the Knight First Amendment Institute at Columbia, neatly summarized the tension (in a tweet, naturally): “It’s coherent—and in my view absolutely appropriate—to believe both that (i) the social media companies were right to suspend Trump’s accounts last week; and (ii) the companies’ immense power over public discourse is a problem for democracy.” In another tweet, he added, “The First Amendment question is easy. All the other questions are hard.”
Let’s take the easy question first. Nothing in the Constitution prohibits a private company from enforcing its own policies; if anything, the First Amendment protects a company’s right to do so. Now the harder questions. Does censoring a head of state set a dangerous precedent? Yes, it does, but so does allowing a head of state to use a platform’s enormous power, over the course of several years, to dehumanize women, inflame racist paranoia, flirt with nuclear war, and incite armed sedition, often in flagrant violation of the company’s rules. Is it worrisome that Jack Dorsey, a weirdly laconic billionaire with a castaway beard who has never been elected to any public office, is able to make unilateral, unaccountable decisions that may help determine whether our country survives or self-immolates? Yes, it is. But, given that Dorsey and a handful of other techno-oligarchs have this ability, they might as well be pressured (or shamed, or regulated) into using it wisely.
The suppression of speech we despise can lead down a slippery slope toward the suppression of speech we cherish; indeed, it almost always does. We should worry about this, but we should also worry about another slippery slope: the one we are already on. Twitter and the other major social networks spent their first decade of existence branding themselves as “the free-speech wing of the free-speech party,” using this as a catchall excuse to absolve themselves of any real responsibility for moderating their platforms. They seemed to assume, blithely and conveniently, that the marketplace of ideas would take care of itself. This isn’t what happened. Instead, with shocking speed, social media decimated professional media, abraded our civic life, coaxed us into unhealthy relationships with our phones and with one another, harvested and monetized our personal data, warped our brains and our politics, and made us brittle and twitchy and frail, all while a few entrepreneurs and investors continued to profit from our addiction and confusion. Social media was hardly the only malign force in the world, but it certainly didn’t seem to be helping. Just a few years into this unprecedented global experiment, several formerly stable liberal democracies found themselves on the precipice of authoritarianism. Britain left the European Union, Brazil and the Philippines came to be ruled by thugs who routinely threatened to kill their political opponents, and India, once a beacon of religious pluralism, descended into Islamophobic mob violence. It seemed as if there were no more ways for the nightmare to grow more dire, and yet it always did. Soon enough, millions of Americans were radicalized, lost in an epistemic fun house of pernicious drivel, and one day a few hundred of them formed a mob and assailed the Capitol, planting bombs and smearing shit through the halls, leaving at least six people dead. For years, social-media tycoons have been allowed to avoid accountability by relying on airy abstractions—we want to change the world; we believe in people; we support free speech. It’s long past time, at the very least, to weigh the benefits of these abstractions against a frank accounting of social media’s measurable, tangible harm.
If the Constitution provides a way to remove a sitting President from office under extenuating circumstances, then there must be a way to remove a sitting President from the Internet. Twitter was right to ban Trump—I think the ban should have come years ago, when Trump started repeatedly flouting the platform’s rules—but my confidence in this opinion shouldn’t be mistaken for a glib assumption that an action of this magnitude can come without downside risk. The hard questions are hard precisely because there are no good answers, only bad ones and worse ones. “No problem that landed on my desk, foreign or domestic, had a clean, 100 percent solution,” Barack Obama writes in “A Promised Land,” a book whose phlegmatic tone is almost shocking against the backdrop of the present chaos. One of many differences between Obama and Jack Dorsey is that the problems Obama faced during his Presidency—an American sailor held hostage by Somali pirates, the Deepwater Horizon oil spill, the war in Syria—were not problems of his own making. If it weren’t for Jack Dorsey and a few of his buddies, though, Twitter wouldn’t exist. They created it from nothing, invented its deeply flawed mechanics and its perverse incentive structures, spent years encouraging as many people as possible to devote as much time and attention to it as possible, and then, essentially, washed their hands of it and walked away. The horrific optics of January 6th were enough to shock Twitter and other platforms into action. But any ban, no matter how prominent, is still a relatively superficial intervention, because it doesn’t change the platform’s underlying architecture. Jack Dorsey often muses publicly about how he might improve “conversational health” on his platform: by diminishing or eliminating the importance of such metrics as retweets and follower counts; by introducing significant friction to make disinformation less likely to go viral; by rebuilding his company’s algorithms from the ground up. A few of these ideas have been implemented, in part, but most of them, so far, have been little more than talk.
The Trump problem hardly caught Twitter by surprise. In 2019, Jack Dorsey did a round of podcast interviews and press appearances, hoping to boost “conversational health”—and, surely, Twitter’s stock price—with yet more public conversation. The podcast host Joe Rogan asked Dorsey whether he’d considered getting rid of Donald Trump, one of the most influential and least healthy conversationalists on the platform. Dorsey demurred, arguing that the words of a President are inherently newsworthy. “We should see how our leaders think and how they act,” he said. “That informs voting, that informs the conversation.” In the end, Twitter banned Trump, ostensibly, for two tweets posted on January 8th. The first, in which he referred to the seventy-five million Americans who had voted for him as “patriots,” was hardly one of the most incendiary things he’d ever posted. (It wouldn’t even make the top fifty.) The next tweet read, in its entirety, “To all of those who have asked, I will not be going to the Inauguration on January 20th.” This was, ironically, one of the tiny minority of Trump’s tweets that really was unambiguously newsworthy. Twitter argued that “President Trump’s statement that he will not be attending the Inauguration is being received by a number of his supporters as further confirmation that the election was not legitimate”; to my eyes, on the contrary, it looked like the closest Trump will ever come to a concession. If you take Twitter’s reasoning at face value, then the most generous way to interpret the ban is that the company made the right decision for the wrong reasons. Perhaps the real reasons for the ban were simpler—that Trump is now a lame duck who can no longer punish Twitter with the levers of the federal government; that the siege of the Capitol was simply one bad press cycle too many; that the company is worried about violence in the near future, and is trying to avoid ending up with even more blood on its hands. If Twitter is being coy about its real motivations, or if the thinking leading to this monumental decision was really as muddled as the official explanation suggests, then there is little cause to think that its future decisions will be much more coherent.
“I doubt I would be here if it weren’t for social media, to be honest with you,” Donald Trump said in 2017. He may have been wrong; after all, he uttered those words on Fox Business, a TV network that will surely continue to have him on as a guest long after he leaves the White House, and even if he loses every one of his social-media accounts. Perhaps Trump could have become President without social media. There were plenty of other factors militating in his favor—a racist backlash to the first Black president, the abandonment of the working class by both parties, and on and on. Still: Trump wanted to be President in 1988, and in 2000, and he couldn’t get close. In 2012, just as social media was starting to eclipse traditional media, Trump was a big enough factor in the Republican race that Mitt Romney went to the Trump Hotel in Las Vegas to publicly accept his endorsement. Only in 2016, when the ascent of social media was all but complete, did Trump’s dream become a reality. Maybe this was just a coincidence. There is, tragically, no way to run the experiment in reverse.
