Recent research published in Proceedings of the National Academy of Sciences When a political leader suddenly expresses support for censorship, it suggests that his or her supporters are more likely to adopt the same view. The findings provide evidence that statements from prominent politicians can easily sway public opinion regarding fundamental democratic rights such as freedom of speech. This highlights how political language can influence voters to abandon long-held values in favor of restricting the rights of opponents.
President Donald Trump has frequently portrayed himself as a strong defender of free speech. Following the assassination of conservative activist Charlie Kirk, Trump and his administration suddenly shifted their message. They began to argue that speech should be restricted under certain conditions, suggesting that overly negative media coverage should not be protected by the Constitution.
Matthew E. K. Hall, David A. Potenziani Memorial University Professor of Constitutional Law and director of the Rooney Institute for Democracy at the University of Notre Dame, explained the motivation behind the study. He pointed to the unusual nature of this sudden shift in political messaging.
“After the assassination of Charlie Kirk, Donald Trump and various members of his administration dramatically changed their rhetoric regarding free speech,” Hall said. “For years, Trump has portrayed himself as a champion of free speech. Suddenly, he and his allies repeatedly claim that various forms of expression are not protected by the First Amendment.”
This change has provided scientists with a unique opportunity to test competing theories about public opinion. One theory holds that ordinary citizens have a deep and abiding commitment to the core values of democracy. The opposite theory argues that people tend to blindly follow the instructions of influential political elites such as presidents and members of Congress.
“Americans have traditionally expressed strong support for free speech, and the Republican Party has become particularly supportive of this right in recent years,” Hall said. “So we wanted to see how the public was reacting to the sudden change in President Trump’s rhetoric.”
Previous research provides evidence that the American public overwhelmingly supports free speech in the abstract. However, this support often weakens when asked about protecting the speech of groups that people strongly dislike. The authors wanted to know what happens when major politicians openly challenge the rights of the broader public. They sought to understand whether loyal voters would change their views on censorship to align with the new positions of their preferred leaders.
Mainstream political leaders openly calling for government censorship is incredibly rare in contemporary American politics. The Trump administration’s recent policy change marks the first time in recent history that a modern president has explicitly advocated for restrictions on fundamental freedoms. The researchers designed a study to see if this new rhetoric would change attitudes in a highly polarized political environment.
To explore these questions, researchers designed an online experiment involving 1,305 adult participants in the United States. The participant group was matched to the general population in terms of age, gender, education, and ethnicity. The sample was evenly divided and included 653 people who voted for Donald Trump in 2024 and 652 people who did not vote. The group of non-Trump voters included those who voted for Kamala Harris, those who voted for third-party candidates, and those who did not vote at all.
Participants were randomly assigned to one of three experimental groups. The first group read quotes against free speech based on actual statements by Trump, Federal Communications Commission Chairman Brendan Carr, and Attorney General Pam Bondi. The second group read aloud pro-free speech quotes that the researchers created by completely reversing the anti-speech quotes. The third group served as a control and did not read any quotation marks.
Within the quote reading group, the scientists introduced another layer of randomization. Half of the participants viewed quotes from real sources, such as Donald Trump and Pam Bondi. The other half had exactly the same quotes, but from popular politicians. This design allowed the authors to test whether the identity of the speaker changed participants’ responses to the message.
After reading the assigned material, participants answered a series of questions designed to gauge support for government censorship. Researchers asked whether governments should be allowed to censor media sources, websites, and talk shows. Specifically, we asked about media outlets that spend more time attacking participants’ own political parties than the opposition. In political science, an individual’s political group is known as the “inside party,” and the opposing group is known as the “outside party.”
The survey also measured support for censoring everyday individuals. Participants pointed to whether the government should be allowed to censor authors in local libraries and speakers in the community. These questions specifically targeted individuals who were highly critical of the participants’ party.
The data showed that exposure to statements that violate free speech increased support for censorship among Trump voters. When people who voted for President Trump read quotes calling for restrictions on speech, they were far more likely than a control group to support government censorship of outside media and individuals. Exposure to quotes supporting free speech did not significantly change Trump voters’ attitudes.
The free-speech quotes provoked a very different reaction among non-Trump voters. For these people, reading restrictive quotes creates a backlash. Non-Trump voters significantly lowered their support for censorship of non-party media and individuals after reading speech control statements.
“We found that exposure to President Trump’s new anti-speech rhetoric increased Trump voters’ support for government censorship of non-party media and individuals, but spurred a backlash among non-Trump voters, increasing their support for protecting non-party speech,” Hall told SciPost. “Our findings suggest that Republicans are more enthusiastic about supporting Trump than they are about free speech, whereas Democrats are actually more supportive of free speech when they hear that Trump opposes it.”
Interestingly, the researchers found that explicitly specifying the source did not change the results. This anti-speech rhetoric, whether the quote was from Trump or politicians in general, increased support for censorship among Trump voters and decreased support among non-Trump voters. This suggests that the style, content, and familiarity of political language convey implicit partisan signals.
“I was surprised that Republicans had the same reaction whether we told them this rhetoric was coming from Trump and his allies or not,” Hall said. “This finding suggests that most Republicans were already aware of this rhetoric or were able to recognize the tone coming from their political leaders.”
The researchers also conducted additional exploratory analyzes to measure broader attitudes. They considered whether the citation influenced skepticism about the First Amendment, which guarantees free speech. They found that exposure to pro-speech rhetoric actually increased First Amendment skepticism among non-Trump voters. Furthermore, this anti-speech quote reduced warm feelings toward Donald Trump among non-Trump voters, especially when the quote was directly attributed to Trump.
The authors note that there are several limitations to how these findings should be interpreted. The most notable unknown factor is the durability of these changing positions. It’s entirely possible that the increased support for censorship among Trump supporters is a temporary reaction to recent news. Over time, these voters may naturally revert to their basic support for constitutional rights.
Future research should track public opinion over time to see whether these changes persist. Scientists suggest that the longevity of these attitudes is likely to depend on whether political elites continue to normalize anti-free speech rhetoric. If politicians routinely call for censorship, the shift in public opinion could become permanent. Even short-term shifts in public sentiment can pose a challenge to democratic norms, highlighting the enormous influence politicians wield over their supporters.
The study, “From Free Speech Advocates to Critics: The Trump Administration’s New Rhetoric Drives Support for Censorship Among Trump Voters,” was authored by Matthew E. K. Hall, B. Tyler Lee, and Brittany C. Solomon.
