Tolerating COVID Misinformation Is Better Than the Alternative

It’s hardly clear that more aggressive content moderation would save lives.

An illustration of a coronavirus particle with an eye at the center
The Atlantic

About the author: Conor Friedersdorf is a staff writer at The Atlantic and the author of the Up for Debate newsletter.

On December 30, 2019, Li Wenliang, an ophthalmologist at Wuhan Central Hospital in Hubei, China, began to warn friends and colleagues about the outbreak of a novel respiratory illness. Four days later, he was summoned to appear before local authorities, who reprimanded him for “making false comments” that “severely disturbed the social order.”

In hindsight, Li was the first person accused of disseminating medical misinformation during the coronavirus pandemic, despite the fact that he was telling the truth. And as the virus spread, many other countries decided that the emergence of a deadly new disease warranted new restrictions on what people say. Human Rights Watch reports that at least 83 governments used the pandemic “to justify violating the exercise of free speech.”

The United States has avoided the worst excesses of this global authoritarian turn. The First Amendment constrains its government from infringing on freedom of speech. And many Americans reliably object to nongovernmental attempts to suppress ideas, favoring the liberal notion that “the remedy for speech that is false is speech that is true,” as Justice Anthony Kennedy once put it.

But like wars, terrorist attacks, and other events that confront us with mass death, pandemics cause some people to doubt the liberal project and to clamor for an alternative that feels safer. So a growing faction in the U.S. feels that, when it comes to “medical misinformation,” liberal remedies for false, unreasoned, or uninformed speech are insufficient to our new pandemic reality—as if being wrong on most subjects is permissible, but being wrong on COVID-19 is too costly to tolerate.

The First Amendment hasn’t kept public officials from calling upon Facebook, YouTube, Twitter, and other tech platforms to restrict false or misleading claims about vaccination and other COVID-related issues. The White House has urged tech companies to censor individuals engaged in protected speech. Senator Amy Klobuchar introduced legislation in hopes of pressuring social-media companies to do more “to prevent the spread of deadly vaccine misinformation.” And the government can apply pressure on private speech in other ways: The Department of Homeland Security, for example, is characterizing misinformation as a terrorism threat. All of these efforts reflect a judgment that, at least on pandemic matters, the liberal approach to dissent has greater costs than benefits.

But that judgment is mistaken. During past crises, even wars, the case for liberal speech norms remained so strong that Americans look back on departures from them with regret. Likewise, I can think of at least four reasons why neither government officials nor corporate bosses should try to protect the public by newly restricting the expression of ideas, even during a pandemic.

1. Open discussion of vaccines enhances trust in them.

Proponents of aggressive measures to restrict misinformation imagine that by preventing others from being exposed to falsehoods about vaccines, they will only increase trust in science. But my early confidence in the Pfizer and Moderna vaccines and my willingness to urge others to get vaccinated was inextricably tied to my confidence that all relevant information, including dissenting perspectives, was making its way into public discourse where countless people could interrogate it, rather than being suppressed by public or private actors with unknown motives. The author Jonathan Rauch calls that process of unconstrained public deliberation “liberal science,” and I suspect that faith in it was especially important for vaccine uptake in the U.S., as Americans report having low trust in government relative to citizens in other wealthy countries.

If you’re vaccinated, think back to before you got the jab and ask yourself: If the Trump administration had announced that federal employees weren’t allowed to spread “misinformation” about any of the vaccines supported by Operation Warp Speed, or if Facebook had suppressed all personal stories of vaccine side effects, would that have caused you to trust the vaccines more or less?

2. Defining misinformation is almost always a subjective judgment.

As a general matter, no person or group or office is capable of assessing what facts or viewpoints constitute misinformation so reliably as to justify censorship based on their conclusions. A concept as malleable as misinformation tends to be interpreted in biased or self-serving ways—and not only by the Chinese government. After Puerto Rico enacted a prohibition on spreading false information about emergencies in 2020, the ACLU filed a challenge on behalf of two journalists who feared the laws would stifle legitimate reporting on the government’s COVID response.

Even if a reliable judge of misinformation did exist in a given jurisdiction, that person likely wouldn’t be the one who decides which ideas are restricted. And even if, for the first time in history, an infallible judge of misinformation was identified and put in charge of restricting ideas, they would still lack popular legitimacy. Many people would disagree and rebel against that judge’s decisions.

Restricting misinformation during this pandemic is especially unlikely to be viable because the scientific consensus continues to evolve, as does the virus itself. So far, the prevailing advice to get vaccinated has served those who took it extremely well, but variants could emerge that pose a greater challenge to existing vaccines. If efficacy drops off, people should be able to discuss that in real time, without having those discussions labeled as misinformation.

The free-speech advocate Nadine Strossen is among those tracking real-life situations in which public- or private-sector authorities have wrongly accused people of spreading misinformation. “The Puerto Rican government charged a prominent clergyman with allegedly disseminating false information on WhatsApp about a rumored executive order to close all businesses,” she reported in Tablet last year. “In fact, only a short time later, the governor did issue such an order.”

Strossen also notes that YouTube took down a video in which Nicole Malliotakis, then a New York State legislator, announced that she was suing then–New York City Mayor Bill de Blasio over his vaccine-passport directive. Malliotakis declared the policy an invasion of privacy. That, Strossen writes, is “a position that the Supreme Court might well end up sharing.” YouTube initially deemed the video a violation of its misinformation policy, though the service ultimately restored the clip. “Whether or not YouTube actually had a good-faith health reason for its initial removal,” Strossen notes, “the vague policy can easily be invoked as a pretext, masking other motives.”

3. Restricting misinformation creates free-speech martyrs.

The backlash against Americans who try to “cancel” others or to shut down conversations rather than engage in them is so widespread that many attempts to deplatform a person inspire others to rally around the target, increasing their fame and reach as well as support for their views.

I suspect that this happens again and again partly for psychological reasons that the political psychologist Karen Stenner describes in The Authoritarian Dynamic. As she explains it, people with authoritarian predispositions want to suppress difference and achieve uniformity, a project that requires “autocratic social arrangements in which individual autonomy yields to group authority.”

But when they try to force everyone to adopt a social consensus for the greater good (for example, “Vaccines are good; everyone must get one, and no one must speak ill of them”) the people who are most averse to uniformity and protective of difference and diversity suddenly rebel.

These rebels “have little concern for the uses of collective authority until other people’s ambitions for its usage suggest that they ought to take an interest in its limits,” Stenner writes. But when autonomy and diversity seem to be in jeopardy, “deemed by those with less stomach for public discord and partisan strife to be too risky for the collective,” they bolster their commitment to tolerance of difference and opposition to social conformity. I recognize that psychology in myself. When liberal free-speech norms are secure, I focus on the substance of debates, such as the overwhelming evidence that getting vaccinated reduces one’s risk of death. But once liberal norms are threatened, as when anti-vaxxers are deplatformed, I’m less interested in the substance of what people are saying than in defending their ability to say it.

4. The concrete harms of medical misinformation are not well established and are speculative in many cases.

The podcaster Joe Rogan has attracted criticism for interviews with vaccine-skeptical guests who made false or misleading claims. Accuracy is worth defending for its own sake, so I wish Rogan would do more to probe dubious claims by guests and to flag and correct statements that prove false. And I don’t dismiss the possibility that his podcast influenced some to forgo jabs or the possibility that some unvaccinated listeners died. The hundreds of medical professionals and scientists who signed an open letter urging Spotify, which carries Rogan’s show, to adopt an anti-misinformation policy are clearly hoping that more assertive content moderation will save lives.

But the evidence for that conclusion is insufficient. As my colleague Daniel Engber has written:

Vaccine refusal, in its broadest sense, has taken a catastrophic toll in the United States … But the claim that pandemic falsehoods aired on Rogan’s show are substantially responsible ignores the sticky facts of our predicament. Surveys now suggest that roughly one in six American adults says they won’t get vaccinated for COVID-19. That’s roughly what the surveys showed over the summer; it’s also roughly what the surveys showed in the summer of 2020, when the pandemic was still young. One in six adults, some 45 million Americans in all, is seemingly immune to any change of context or information. One in six adults—a solid tumor on our public health that doesn’t grow or shrink.

Perhaps America’s pandemic performance would be better in a world without The Joe Rogan Experience. But it could be that the vaccination rate would be unchanged in that alternate reality, because podcasts aren’t an important driver of attitudes toward vaccination, or because vaccine skeptics tend to find voices who question the mainstream regardless. Maybe the shows they’d seek out but for Rogan would be less responsible than he is. Maybe Rogan’s pro-vaccine guests were uniquely able to reach skeptics via his podcast, while Rogan listeners who found vaccine-skeptical guests persuasive were already anti-vax and unlikely to change.

Admittedly, if I were forced to bet on the proposition that COVID-19 misinformation on a popular podcast or cable-news show did some harm, I’d bet yes. But restricting discourse based on mere circumstantial evidence of harm is an authoritarian standard.

That’s not to say that private companies must never engage in censorship in the name of avoiding harm. If someone posted a recipe with poisonous ingredients, or a message directing another person to commit suicide or violence against others, I would not object if Facebook removed it.

But any speech limitation—public or private—needs a clear limiting principle. The Supreme Court set a useful standard in defining illegal incitement: to qualify, statements must be “directed to inciting or producing imminent lawless action” and “likely to incite or produce such action.” Any standards for taking down alleged misinformation online should be similarly demanding. A free society should not attempt to shut down speakers whom millions of people want to hear based on speculative claims that some of their viewpoints cause unquantifiable harm.

It would be nice if everyone agreed about how to fight COVID-19. But normally, Americans understand that this is a huge, populous, multicultural, pluralistic country where diversity and difference are inescapable. In almost every dispute of consequence, tens of millions of us are right and tens of millions of us are wrong. Yet when it comes to COVID-19 vaccines, the fact that a minority of Americans have refused to get a shot is being treated as evidence of a “misinformation crisis” for which media and tech companies are responsible. The presumption is that if not for a dysfunctional information environment, uptake would be universal.

In fact, vaccine skepticism is a phenomenon as old as vaccines, manifesting across centuries of outbreaks that took place in wildly different information-technology environments. More broadly, the unfortunate truth is that all of us are misinformed and incorrect about many important matters, not because of technology-driven misinformation but because of enduring human fallibility.

Americans need only look abroad—for instance, to Li Wenliang’s experience—to verify that abusive actors invoke the need to fight pandemic misinformation specifically to increase their own power and infringe on liberty. While much harm might be avoided if any misinformation a human was about to utter was magically muted, there is no way, in a world without such magic, to suppress misinformation without stymieing difference and debate on every matter of consequence.