In a new study published in Political Psychology, researchers found evidence that some conservative voters in the United States view inflammatory pro-gun rhetoric as a sign that a political candidate is unwilling to compromise with Democrats—and that this unwillingness to compromise can make the candidate seem more trustworthy and appealing to those voters. This pattern reflects a broader trend in American politics, where signaling opposition to the other side can be used as a strategy to build trust with a party’s base.
The study was prompted by an effort to understand the strategic use of inflammatory rhetoric in U.S. politics. The researchers were especially interested in why certain candidates, such as Mehmet Oz during his 2022 Senate campaign in Pennsylvania, chose to produce provocative political messages that seemed likely to alienate moderate voters.
Although such messaging might appear politically risky, the researchers proposed that these communications serve a specific signaling function: they demonstrate a candidate’s unwillingness to work with the opposing party by intentionally offending them. This approach draws on signaling theory, a concept from economics and evolutionary biology. In situations where one side wants to demonstrate a trait that cannot be directly observed (like loyalty or unwillingness to compromise) actions that carry a cost can serve as a signal.
“We started thinking about this work in earnest when we were trying to understand the mysteries of the Mehmet Oz campaign for Senate a few years ago,” said study author Nicholas Buttrick of the University of Wisconsin-Madison. “Why would a media pro seem so tin-eared in his politicking? What did he understand that we were missing? In modeling his behavior out, we think that what he (and other politicians who take similar approaches to their communication) was doing was sending a signal about his fitness as a candidate for the Republican party. It’s an unusual sort of signal, we think – one that actually does things with words.”
To examine this, the researchers conducted three preregistered experiments with more than 1,700 American conservatives. In each experiment, participants were shown hypothetical social media posts from fictional Republican candidates. These posts were designed to either include inflammatory pro-gun rhetoric or more civil, conciliatory messages, often framed as “thoughts and prayers” following a mass shooting.
In the first experiment, participants read vignettes that varied along two dimensions: the rhetorical style of the candidate’s tweet and the candidate’s personal background. Some candidates were described as having typical Republican credentials, while others had less traditional, more liberal-leaning histories. Participants rated each candidate on their perceived trustworthiness, likelihood of compromising with Democrats, and how likely they would be to vote for the candidate.
The researchers found that candidates who used inflammatory rhetoric were viewed as less willing to compromise with Democrats, especially when participants believed that liberals would react negatively to the post. This perception, in turn, made some candidates appear more trustworthy to conservative voters who disliked Democrats.
Interestingly, the rhetorical style of the message had a greater influence on perceptions than the candidate’s background. This suggests that what a candidate says—especially how offensive it is to the other side—can override doubts based on their prior political history.
The second experiment built on these results by examining how voters’ feelings toward Democrats shaped their reactions to inflammatory messages. Participants again read tweets from fictional Republican candidates, but this time the researchers focused on measuring the participants’ affective polarization—specifically, how negatively they felt about Democratic politicians.
The findings showed that conservatives who held more negative views of Democrats were more likely to find inflammatory candidates trustworthy and were more inclined to vote for them. On the other hand, conservatives with more neutral or positive views of Democrats preferred the candidates who used less inflammatory, more civil language. This pattern suggests that the effectiveness of inflammatory rhetoric depends not just on the message itself, but also on the audience’s existing attitudes toward the political opposition.
The third experiment aimed to separate the effect of the content of the message from the tone. Both messages in this experiment expressed strong opposition to gun control, but one was phrased in inflammatory language while the other was civil.
Even though the substance was the same, participants perceived the candidate using inflammatory language as less willing to compromise. However, they also rated that candidate as less trustworthy and were less likely to say they would vote for him. This indicates that the language itself—apart from the political stance—can signal unwillingness to work with the opposing side, but that this doesn’t always lead to greater support.
Across all three studies, the researchers found consistent evidence that inflammatory rhetoric functions as a signal of political intransigence. It sends a message that the candidate is unwilling to compromise with the other party, and for voters who strongly dislike that party, this can be seen as a positive trait. However, this effect is not universal among conservatives. Those who feel more moderately or positively about Democrats tend to react negatively to inflammatory candidates.
“By using language that is designed to offend Democrats, Oz is demonstrating not only that he’s uninterested in crossing the policy aisle, he is taking steps to ensure that he literally cannot,” Buttrick told PsyPost. “By making himself politically-radioactive to anyone but his political base, he is guaranteeing that nobody will be willing to work with him, and therefore he will quite simply be unable to compromise with the other side. If you’re a partisan in a polarized political environment, burning bridges with outgroups is seen as a good thing since it means that your candidate can’t betray your group, since it’s the only base they have left.”
“We demonstrate that this reading of Oz is plausible in a set of preregistered survey experiments where we present conservatives with a set of carefully-designed political tweets. We find that for participants who dislike Democrats, candidates who make inflammatory pro-gun statements are viewed as more trustworthy, to the extent that they are seen as saying things that would offend Democrats. However, not all Republicans are quite so polarized, and for those who feel neutral-to-positive about Democrats, the offensive tweeting seems to actually backfire, reducing evaluations of the tweeter. As we point out in the paper, while Oz’s strategy may have made sense in the primary, he did handily end up losing his bid for Senate.”
As with all research, there are some limitations. The studies were conducted online using hypothetical scenarios, which may not fully capture how voters respond to real-world political campaigns. Participants were also not representative of the general population, as they were drawn from an online pool of self-identified conservatives.
The experiments also focused specifically on inflammatory pro-gun rhetoric. While the researchers believe that similar patterns might apply to other types of inflammatory political speech, further research is needed to determine whether the findings generalize beyond this issue area. There may also be cultural or institutional factors that influence how these dynamics play out in different political contexts or across party lines.
Future studies could explore how inflammatory messages influence actual voting behavior, campaign donations, or social media engagement. Researchers might also examine how these strategies differ between primary and general elections, where the audience and incentives may vary. Another important direction would be to study how political elites and media figures use inflammatory rhetoric to shape public opinion and build their personal brands.
The study, “Love him for the enemies he has made: Signaling by inflammatory pro-gun rhetoric,” was authored by Sosuke Okada and Nicholas Buttrick.
