“Majority of Americans Want to Scrap First Amendment, Polling Finds” blared the headline. Pundits who present themselves as defenders of free speech — primarily against woke college students and social justice warriors — shared it widely.
A write-up in the Free Beacon warned of “American hostility to the First Amendment.” InfoWars’ Paul Joseph Watson shared the findings and said he was being silenced. Reason’s Robby Soave claimed personal vindication:
The poll comes from a group called Campaign for Free Speech. According to its website, the organization was created because “the United States public has lost its connection with the First Amendment, leaving protected speech more vulnerable than ever before in modern history.” Now they conducted a study that, according to them, shows “just how vulnerable free speech protections are.”
But their poll doesn’t really show that. Some question wording is ambiguous, and some seems designed to get these results. Campaign for Free Speech published the data, indicating good faith intentions. But their study reeks of bias, declaring they proved their predetermined conclusion, even though they didn’t.
Similarly, many who shared the study online likely did so because it claimed to support what they already believe.
Polling With an Agenda
Here’s the first question, the one that inspired all the hand-wringing:
The First Amendment, which provides the right of Americans to have free speech, was enacted more than 200 years ago. Do you agree or disagree with this statement? ‘The First Amendment goes too far in allowing hate speech in modern America and should be updated to reflect the cultural norms of today.’
Fifty-one percent said they agreed.
But why start by noting that the First Amendment “was enacted more than 200 years ago”? It’s not necessary to know that to evaluate the statement.
In social science, that’s called priming. Pollsters tell respondents something right before asking a question to place an idea at the front of their minds. It could be new info, or just a reminder of something they already know, but the goal is the same: get respondents thinking of something as they’re about to answer the question.
Campaign for Free Speech primed respondents with the idea that the First Amendment is old, and then asked a question using language connoting newness. Should the First Amendment “be updated to reflect the cultural norms of today”? Many people, especially younger ones, think old things could use an update.
I don’t know if Campaign for Free Speech did this on purpose in pursuit of a desired result or did it by accident due to bias, but their question wording skewed their data. If they primed respondents to think differently, with something like “First Amendment principles were upheld in multiple court cases this year,” they’d probably get a different result. And if they asked a straightforward, unprimed question, like “should the United States scrap the First Amendment?,” I bet more would say no.
With the priming, 51 percent agreed that the First Amendment should be updated, but 33 percent of the total agreed only “somewhat.” Some of them were probably noncommittal, responding to the priming and a lack of specificity on the changes with something like “an update could be okay, I guess.” If the government actually tried to revise the First Amendment, many of those respondents probably wouldn’t support it, or could easily be persuaded not to.
Twenty-four percent strongly agreed, which is probably a closer measure of the percentage of Americans who really would update the First Amendment. That’s far from nothing, and it’s reasonable to criticize anyone who believes it. But in a democracy, under a quarter of the population is much less threatening than over half. In this case, 1/4 is even less threatening because a Constitutional amendment requires 2/3 of both the House and Senate (or 2/3 of state legislatures, which has never happened), and then 3/4 of the states for ratification.
This is the study’s headline result, and the data does not actually show what Campaign for Free Speech claimed. “Update” does not mean the same thing as “scrap,” and the evidence that 51 percent would update the First Amendment is ambiguous at best.
Hate Speech, Free Speech
The other question appearing in write-ups concerns hate speech. This time, the survey was straightforward, asking “do you think that hate speech should be against the law?”
Forty-eight percent said yes. But when asked what the punishment should be, slightly under half of that group agreed with “nothing more than a ticket or a fine.” That leaves 26 percent of the total who supported “possible jail time,” a number similar to the 24 percent who strongly agreed the First Amendment should be updated.
The survey doesn’t define hate speech, leaving no indication as to what respondents had in mind. Maybe they were thinking of a pundit who doesn’t use a trans person’s preferred pronoun — an action protected by the First Amendment. Or maybe they were thinking of someone running up to a black person and shouting racial slurs in their face, which is harassment and not protected by the First Amendment. Without knowing what respondents think “hate speech” means, we don’t know what sort of changes to the First Amendment they’d support (if such changes were politically possible, which they’re not).
Here’s another ambiguous agree/disagree question:
While I agree in principle with the idea of free speech, there are places where free speech should be restricted. For instance, in universities or on social media where there is the potential to be hurtful or offensive.
Sixty-one percent agreed. Soave and other pundits who see restrictions at universities as the vanguard of an anti-speech movement might highlight this as proof of a growing crisis.
But like the hate speech question, we have no idea what respondents had in mind when they answered. Left-wing college students constantly nitpicking supposed microaggressions, or a professor penalized for denigrating a gay student with slurs? Though the question brings up universities or social media, respondents could be thinking of different situations.
“There are places where free speech should be restricted” is passive voice, leaving open many possible interpretations. Restricted how? And by whom? A boss telling employees that using slurs makes for an uncomfortable work environment? Or a boss firing employees for expressing an opinion widely considered acceptable just a few years ago?
Different specifics lead to dramatically different conclusions. One could, for example, support tight speech restrictions as determined by gender studies professors and want to use that as a stepping stone to revising the First Amendment. Or one could merely support some simple guidelines in university codes of conduct and social media terms of service. Or many other possibilities.
We need to know those specifics to determine whether this is a growing threat to free speech or merely a 21st-century version of the long-standing norm of mutual respect. Campaign for Free Speech’s study cannot provide the answer.
Freedom of the Press
In the most concerning result, 57 percent agreed that “the government should be able to take action against newspapers and TV stations that publish content that is biased, inflammatory, or false.” Those respondents probably include some Republicans thinking about CNN and some Democrats thinking about Fox News. Majorities support this violation of freedom of the press regardless of race, sex, income, or region, suggesting a bipartisan consensus.
However, as with other questions, only 26 percent strongly agreed and 26 percent (not necessarily the same ones) supported “possible jail time.” That indicates restrictionists are closer to a quarter of the population than half.
Many respondents, especially those who agreed somewhat, were probably expressing anti-media sentiment, rather than support for eliminating First Amendment press protections. One indication that’s the case is the survey got different results when asking about “newer alternative media”:
Traditional media companies such as TV stations and newspapers use an editorial process to check facts, even if they are occasionally wrong or slanted. Newer alternative media, such as online podcasts, allow anyone to say anything, regardless of its accuracy. Would you support a government agency reviewing content put out by these alternative media sources?
Thirty-six percent said yes, while 47 percent said no — a solid plurality opposed to government restrictions on this type of speech.
When asked about Facebook, 38 percent said “Facebook should allow all speech” while 49 percent said “Facebook should monitor and restrict offensive speech and views.” Paired with the question about alternative media, this result indicates that 9–13 percent would accept restrictions from tech platforms that they would not accept from the government.
However, like some of the other questions, this is ambiguous, because what constitutes offensive speech is subjective, and Facebook already restricts some of it. Does that 38 percent really think Facebook should allow all speech, or are they just annoyed by a particular restriction? For example, would all of them say Facebook was wrong to take down the livestream of the Christchurch terrorist attack? How about ISIS recruitment posts?
Whatever the answer, thinking a website should revise its terms of service — or improve how it enforces those terms — is quite different from thinking the United States should amend the Constitution.
First Amendment and the Culture of Free Speech
The ambiguities in this study reflect a problem in the larger discourse about free speech: blurring the line between government actions restricting speech and private actions that fail to uphold the “culture of free speech.” Some of the loudest voices warning of a free speech crisis cite actions that go against the culture of free speech — such as college students protesting a speaker or social media users incessantly lobbing criticism — as threats to the First Amendment. However, while they may be worthy of criticism on their own, those actions are speech, protected by the same laws.
Many who blur this important distinction are well-intentioned, seeing a slippery slope from declaring “you can’t say that” to supporting laws that ban it. But some utilize cries of free speech violations to mean “I can say whatever I want and you can’t criticize me in response.”
This problem is apparent in the survey’s final question, which asks, true or false:
The First Amendment allows anyone to say their opinion no matter what, and they are protected by law from any consequences of saying those thoughts or opinions.
Seventy-nine percent said true. It’s false. The First Amendment prevents the government from prosecuting people for speech, but does not protect people from many consequences of their speech, such as harsh criticism or, in some circumstances, losing their job.
At this point, objective researchers would scrap the survey and start over. After all, if more than three quarters don’t know what the First Amendment does, what does it matter if half say they’re open to changing it?
However, biased researchers who aim to advance a preconceived argument would plow ahead and declare they found exactly what they thought they’d find, even though they didn’t.
Meanwhile, pundits and activists who already believe the First Amendment faces a dire threat would share the write-up without reviewing the evidence, or perhaps even clicking on the article.