A specter haunts the Discourse, and it’s the sense that Twitter is bad for you. There’s certainly been some chin-wagging about this—on Twitter itself (in one of its usual ironies) and in spaces like this one, where I’ve argued that the platform’s very design promotes toxic use. But claiming to quit Twitter only to come slinking back is a time-honoured tradition; numerous users embarrassed themselves by proclaiming Elon Musk’s inevitably abortive takeover to be the last straw, only to find the site’s allure impossible to resist.

More alarmingly, powerful and influential people—call them “epistemic elites”—seem to be among the most terminally online, and it’s having an effect on the rest of the world.

As a scholar who's spent much of the past decade arguing that Twitter is “real life” and that harassment on the platform is a major social problem worth paying attention to, I fear a monkey’s-paw finger has curled in the wake of such advocacy. Taking online discourse seriously is important; letting it become your only window on reality is dangerous, and when powerful people do so it threatens democracy itself.

As is so often the case, however, this is more a structural problem than a question of individual virtue or willpower. “Mindfulness” won’t fix what Twitter broke. Mitigating Twitter’s harms is less about badgering people to use the platform less (though that would reduce the collective psychic burden of mass doomscrolling) and more about ensuring that influential individuals—namely academics, journalists, and politicians—stop using the platform in particular ways. Abstinence-only policies don’t work for anything else, and they won’t work for Twitter, but harm reduction is worth seriously considering.

So, getting some of them to step away is advisable. But how might one do this?

For the past several years, academics have been strongly encouraged to be on Twitter. It’s viewed as a great place to build your brand (another noxious inheritance of Web 2.0), reach new audiences, and network.

It’s also become a premier space for science communication through lengthy threads sharing research. This has benefited academics from marginalized backgrounds and those who work in less favored subdisciplines. Some have leveraged Twitter to create audiences for work that would otherwise be hidden behind the padlocks and paywalls of academic publishing.

But this doesn’t always lead to deeper engagement with the work, and the velocity of snappy, viral tweets only deepens the notoriously clubby drama of each profession. The pandemic made this abundantly clear. I’ve spoken privately to numerous epidemiologists and public health experts who lament the effect that Twitter has had on the discourse in their professions, making bitter enemies out of people who might have once conducted esoteric disagreements about the science in more sedate settings.

Diminishing Twitter’s corrupting role here requires a rejiggering of the incentives for academics to ply our trade there to excess. The publishing of academic research has is badly in need of reform. Even as the pandemic has made a mockery of science communication, with countless people bandying about virological studies they only dimly understand, there needs to be more access to such studies, not less. Universities need to throw more of their considerable weight behind open source projects and promote more of their younger graduate students, research assistants, and adjunct faculty through their communications and media teams, reducing the need for alternate platforms.

Such reinvigorated collaboration could only be good for academics and journalists alike, who would benefit enormously from the elevation of early-career researchers who, of necessity, are on the bleeding edge of every vital area of study, from the pandemic to climate change to the internet itself.

Twitter’s popularity among influential journalists increasingly bends everyone’s perception of reality around the day’s toxic Twitter discourse. Journalists seem to assume that the platform’s petty arguments are some kind of real-time snapshot of public opinion. This is how we get the embarrassing and dangerous spectacle of a New York Times columnist arguing that you can’t say the word “woman” anymore and that esoteric discourse among transgender Twitter users somehow led to the US Supreme Court’s overturning of Roe v. Wade—a goal Republicans had been working toward for at least three decades before Twitter launched.

Not every dollop of platform drama says something profound about the human condition or modern society, and not every moment of Twitter backlash to some celebrity brain fart constitutes a “cancellation.” But the old saw that “freedom of speech doesn’t mean freedom from consequences” doesn’t capture the truth here either. It’s worth asking, in a serious way, what consequences are you actually reporting on? What lasting impact is this actually, seriously having, either on the allegedly canceled celeb or the wider population?

In general, these supposedly canceled celebrities and public figures are doing just fine—and, indeed, often profit from the myth of their cancellation—because Twitter eruptions against them are nonstories, and overreporting on them has the effect of creating perverse incentives for public figures to court outrage on the platform. A bunch of socialists with funny memes aren’t really going to hurt these reactionaries. Indeed, pretending to be canceled has become a dominant strategy for the new crop of far-right politicians.

So, just as was the case with scholars, it’s important for journalists to not feel like they need to be on Twitter. Recently, the Times itself changed its social media policies to do exactly what I suggest here and ensure that Twitter doesn’t feel like a job requirement. But supporting this materially is the next step. Steps like overtime for additional unstructured interviews or on-the-ground reporting, bonuses for stories not sourced from social media, and travel budgets that let reporters get stories from the places they’re really happening (rather than someone’s Twitter take about it) would all go a long way.

For large chunks of the past two years, the only way millions of people experienced their social and political worlds was through Twitter. During a crisis—especially a rolling crisis like the Covid-19 pandemic, which has literally become the air we breathe—one is surrounded by people’s wailing despair, by unending jags of catharsis, screamed from every corner of the earth. And for those who've been unable to reconnect to the physical world it can be even more damaging, leaving them caught in an unending spiral of doomsaying from Twitter personalities who’ve built their brands around apocalypticism, and who have now glommed onto monkeypox as the next big thing to terrorize their followers with.

Consider how Twitter helped build the punditry careers of people like Eric Feigl-Ding or Jennifer Sey, polar opposites in their views on pandemic policy but each expressing a form of Twitter-driven obsessiveness with such questions that led to ignoring literally every other bit of context around the complex problems the pandemic presented us all with. In a conversation with Sey, WIRED’s Gilad Edelman identified Twitter as the “killer app” for building Sey’s brand; she agreed.

Or consider the tortured case of Dan Price, the former Gravity Payments CEO who has been accused of rape, domestic abuse, and sexual harassment by several women. (He denies these allegations.) New York Times reporter Karen Weise, who interviewed dozens of people who established Price’s pattern of toxic behaviour at Gravity, observed in her report that Twitter in particular enabled him to overcome the allegations of domestic violence by his ex-wife and reinvent himself as a high priest of ethical capitalism, called on by political campaigns, The Daily Show, Aspen Ideas, and more: “Tweet by tweet, his online persona grew back. The bad news faded into the background. It was the opposite of being canceled. Just as social media can ruin someone, so too can it—through time, persistence and audacity—bury a troubled past.”

This isn’t to say that Twitter can’t be a potential source for hiring (goodness knows I’d be a raging hypocrite if I said so), but rather that more attention should be paid to the source of a prospective pundit’s popularity. The distorting vortex of Twitter means that it can be risky to hire pundits from the platform; you’ll only be giving that sort of mentality a bigger megaphone. To give but one example, the allure of Twitter stardom has also ensured that scientists who became associated with either advocating strict pandemic policies or lax ones were left with little room to maneuver when the facts on the ground changed: Their fans demanded consistency.

If the popularity stems simply from riding the platform’s unending waves of drama and perverse incentives, scepticism should be applied. Otherwise, all that media gatekeepers are doing is contributing to Twitter’s toxic reality-show atmosphere.

Meanwhile, the Biden Campaign in 2020 famously lived by its dictum that “Twitter isn’t real life,” which, unsurprisingly, went over badly on Twitter. Many users saw the slogan as an excuse to ignore the progressive activists who thrived on the platform and formed part of the Democratic Party’s base. The line is also simply untrue: Twitter is continuous with real life, and what happens there is impactful.

Here, Twitter itself needs to be more proactive about banning unscrupulous politicians who try to stoke hatred through the platform. Twitter prides itself on being a destination for so many politicians around the world, but it might want to reconsider its own financial incentives after the implosion of Elon Musk’s takeover bid. First and foremost, keeping extremist politicians on the platform contributes to the efflorescence of toxicity that makes the site radioactive to potential buyers. No one wanted to literally own the question of whether Donald Trump should’ve been banned from Twitter. So, just lean into it. Ban most politicians. Take that question and its controversy off the table. It might just make the site more saleable.

This is, of course, trickier than it sounds. Who counts as a politician? Those already in power or those vying for it? What of people who want to be politicians but come from outside the existing clubs of power, and for whom Twitter might be a leveler? Why not ban journalists? (Frankly, not the worst idea at this point.) My response is that successful campaigns—like, say, Alexandria Ocasio-Cortez’s—owe their success as much to classic door-to-door campaigning as to Twitter, if not more. Twitter’s viral incubation may help promote a speech or a policy, but it’s one tool, not the whole box, and there’s always room for deepening these other forms of engagement. At any rate, banning politicians doesn’t stop ordinary people from talking on the platform about their favorite candidates. It just protects them from the too-online obsessiveness that results from our leaders spending too much time there.

Politicians themselves, meanwhile, must assign Twitter its proper place in our constellation of online spaces. It is real life, but it is not a scientific poll. It represents a particular community of users whose importance has been inflated by the presence of tastemakers, media influencers, politicians, and academics.

There is no easy hack for having good judgment, but it’s worth gut-checking every time you find yourself het up over a Twitter drama by asking a not terminally-online colleague or loved one about it. If your rant about Bean Dad draws a blank stare, that’s a sign that you’re engaged in the political equivalent of talking about your World of Warcraft Moonkin Druid build in mixed company and should, perhaps, back off. Larger political forces, like Black Lives Matter or the Hong Kong pro-democracy movement are, by consequence of their very embeddedness in the physical world, in a different class altogether and must be taken more seriously. And that’s a useful heuristic for politicians and their aides to consider: Is today’s trending topic the result of on-the-ground organizing in the physical world, or is it a wholly online discourse?

Twitter, often hyped as a “public square” by everyone from scholars and decent people to men like Elon Musk, is nothing of the kind.

It is a hyper-individualist domain masquerading as a collective one, and it subtly promotes a similar attitude towards one’s own responsibility. Brand-building and expressions of catharsis are individual enterprises that demand the unwilling participation of others. When you take to Twitter to lament the forthcoming end of the world, whether from Covid, monkeypox, climate change, ascendant neo-fascists in the West, or the neo-imperialism of China or Russia, you’re expressing something that many people feel—and with good reason. But you are also, primarily, venting. You’re doing so to feel better, because it feels good to let this out, and to experience the picosecond’s worth of validation you receive from others agreeing with you. That, in the end, is all the platform really is.

What Twitter doesn’t induce is any sense of responsibility to others. To scream about the end of the world may help you feel better even as it adds to a cacophony that is ruining someone else’s day.

Epistemic elites need to walk away from the malign influence of this behavior. For all our sakes. Or at least just step back enough to assign Twitter its proper place in our lives, as one window among many onto the larger world. Abstaining is not reasonable for many, but harm reduction is. Replacing the impulse to scroll with something else is a small but vital first step, as is checking your instincts about popular opinion with your less online friends.

To act as if Twitter is a microcosm of humanity is to threaten us all, regardless of whether we’re on the platform or not, with its toxicity. And, as someone who once likened the internet to the agoras of old, I think we do need to seriously reconsider our desire to cast every virtual space as a “public square” and ask serious questions about what we actually want and need from such spaces.

To borrow from the platform’s vernacular, it’s time we all touched grass.