A Rorschach Test for the State of American Free Speech

If someone says that they feel more of a need to keep their views to themself than used to be the case, what does that tell you? Does it mean that society has grown more censorious? That free speech is on the outs? It’s easy to interpret it that way, especially if we’re on the lookout for such evidence. But these are precisely the situations when we should be most wary so as to avoid overinterpreting data to advance a narrative.

In a recent New York Times essay, columnist David French, a long-time First Amendment litigator, noted that “the law of free speech is only getting more robust. Americans have more concrete rights to speak free of government censorship than they have at any prior period in American history.” But, he added, “according to a survey from the Cato Institute, a libertarian think tank, a strong majority of Americans self-censor. They’re afraid to exercise their rights.”

That survey’s central takeaway is that, “Nearly two-thirds—62%—of Americans say the political climate these days prevents them from saying things they believe because others might find them offensive,” a number that “has risen several points since 2017...” More conservatives feel this way, according to the survey, than liberals, though both groups have enough opinions they’re uncomfortable expressing to tip their numbers into the majority. What the survey leaves out, and this is crucial to what we should make of its findings, is what actual, specific beliefs it is they feel unable to express.

Thus I was skeptical of drawing conclusions from the survey when it came out, to much attention, and I remain so today. Without knowing the content of the beliefs, the survey is little more than a Rorschach test, appearing to confirm what the reader already believes, without actually giving them reason to believe so.

That’s not how the survey’s author frames it, though. This, from the concluding paragraph, makes clear that we are intended to see a rise in people feeling that some of their beliefs will get them socially shamed as obviously troubling.

This large number from across demographic groups suggests withheld opinions may not simply be radical or fringe perspectives in the process of being socially marginalized. Instead many of these opinions may be shared by a large number of people. Opinions so widely shared are likely shaping how people think about salient policy issues and ultimately impacting how they vote. But if people feel they cannot discuss these important policy matters, such views will not have an opportunity to be scrutinized, understood, or reformed.

Taken at this high level of abstraction, yes, of course it’s not good if important policy conversations within a democratic society are one-sided because reasonable views have been unfairly excluded from the public sphere. But policy discussions don’t happen in the abstract. They’re about the actual beliefs of actual voters engaged in actual discussion. And the survey tells us nothing about those.

We are told that 77% of strong conservatives feel that the “political climate prevents [them] from saying things [they] believe because others might find them offensive.” Again, without knowing which “things” they’re worried about, we shouldn’t rush to view this overwhelming majority as a sign of trouble. If the Trump years and the GOP’s far-right culture war turn have shown us anything, it’s that plenty of American conservatives hold incredibly repugnant views about immigrants, racial minorities, and LGBT+ people. Given that it wasn’t until the mid-1990s that a majority of Americans supported interracial marriage, it’s not unreasonable to think that maybe a pretty big chunk of the “things” those 77% of strong conservatives are afraid to discuss fall within the category of views where this number would be a sign of social progress.

We also need to be careful about taking growing worries as a sign that there’s actually something underlying to be worried about. Not too long ago, many on the American right were quite upset about the “War on Christmas,” the myth that progressives had made it so you couldn’t say “Merry Christmas” anymore. On the campaign trail, Donald Trump promised far-right crowds that once in office, “we are going to say ‘Merry Christmas’ again.” Except, obviously, America never stopped saying it.

Along these lines, David French notes that, “Americans have read story after story (from across the political spectrum) of activists, corporations and colleges targeting individuals for speech that is squarely within the mainstream of either progressive or conservative thought.” But reading an increasing number of stories about something is not the same as the thing those stories are about actually becoming more common. Journalists are as susceptible as anyone else to moral panics, and the gotta-get-clicks nature of the only journalism economy incentivizes writers to return, again and again, to topics that create outrage and so get eyeballs on websites. This is why Americans almost always believe crime rates are going up, and are almost always wrong about that. People getting fired for saying things their boss doesn’t like, or losing their jobs because of outside-of-work expression isn’t new. Try being a gay man in 1950s corporate America.

What’s at least as likely going on is not that the range of socially acceptable views is narrowing, but that it’s shifting—as the range of socially acceptable views has for as long as there’s been societies to do accepting and rejecting. Yes, there are Americans who feel that a larger number of their beliefs will get them in trouble today than a decade or more ago. But it’s also true that a huge number of Americans, particularly in historically marginalized communities, feel considerably more free to express themselves, and find social acceptance in doing so, than ever before. If you associate primarily with the former group, chances are you’re convinced America is becoming more censorious. If you associate primarily with the latter group, chances are you’re convinced it’s become less so.

It’s also not clear that an unwillingness to express some views means those views aren’t getting “scrutinized, understood, or reformed,” as the survey’s author worries. Social shaming—making some views risky to state in public—in fact is a way that views get scrutinized and reformed. America is less racist today than it was 60 years ago in large part because shouting racist slurs brings consequences now in a way it didn’t in 1963. Hardcore, overt KKK-style racism has been scrutinized and America’s racist culture has seen considerable—though far from complete—reform, and that’s not because we gave up on socially marginalizing racists and instead decided to have dispassionate debates with them. The free exchange of ideas is important—tremendously so—but no one actually believes all ideas should be on the table of civil discourse or be free from social punishment.

Finally, we absolutely must avoid the trap of thinking that it’s wrong to direct opprobrium at views that happen to be commonly held. After all, hate and bigotry can be quite mainstream. French worries that “a person can be cast out of polite society for saying something completely conventional,” but “conventional” is not the same as “moral” or “acceptable.” In fact, even a cursory tour of history shows that the opposite is frequently the case. Moral relativism is wrong, after all.

Arguably nothing matters more to a liberal society than free speech, and arguably nothing destroys the possibility of a liberal society faster than restricting speech. But America remains, thanks to the power of the First Amendment, almost uniquely safe from government restrictions on expression. But when it comes to social pressures, and given the importance of expression to liberalism, it’s critical that we base our conclusions—and our picture of the state of things—on data that actually tells us what we are motivated to believe it tells us. To do otherwise is a disservice to liberalism’s defense.