This New York Times opinion piece lauds Alex Garland’s recent film Civil War as exactly the “warning” America needs in this election year. This article, conversely, worries that Civil War’s “‘us vs. them’ warfare will become MAGA fantasy fuel.” So which is it? Neither, it turns out, according to some quick impressionistic research we did.

In early April 2024 we randomly assigned about 800 people into two groups. One group saw the Civil War film trailer on their mobile phones. The other group saw a neutral “placebo” piece of media. Then we asked everyone six questions about political violence, democracy, and civil war, in order to understand how four audiences with different values profiles reacted to the fictional world presented in Civil War. (We include the question battery below.)

Here’s what we found. Viewing the film’s trailer tends to normalize political violence for all audiences. In other words, across all audiences, watching the film’s trailer makes people more likely to disagree with the statement, “Violence should never be used against political opponents.” And it makes audiences who hold communitarian and social order values, including people who tend to vote reliably progressive, more likely to agree with the statement, “Because things in the United States have gotten so far off track, Americans may have to resort to violence in order to make things better.” The only audience for whom the film’s trailer may have served as somewhat of a “warning” or a prod to deepen their engagement with democratic forms of participation was the audience we refer to as People Power, who already tend to be pretty engaged with traditional forms of political action.

The upshot of this kind of research is that nowadays there’s no need to guess about how audience engagement with a fictional world affects their attitudes or beliefs about something. This does not need to be a matter of opinion. We can test it empirically. We can know whether a fictional world dramatizing some feared thing might increase the likelihood of that thing happening, by diminishing taboos, normalizing behaviors, and/or encouraging copy-cats; or decrease the likelihood of its happening, by allowing audiences to fully live into the horror, to identify with a heroic resistance, and/or to be otherwise inoculated. And we can use the testing results to inform creative practice, marketing campaigns, or the resources we offer to support audiences in processing the content they engage with.

Obviously we need to keep telling important, difficult stories like the one Civil War tells. The question isn’t whether we tell them, but how. How do we responsibly present such powerful imagery to the public? In an election year where polarization is already at an all time high, and political violence is on the rise, there may be a lot riding on that question. Given the stakes, certainly we should want to do more and deeper research than this tiny impressionistic study allows. After all, we only looked at audience response to the film’s trailer, not the whole film. That’s why Harmony Labs, together with Democracy 2076, is embarking on a much larger study to understand how government and the future have been depicted in entertainment content over the last few years, and how these depictions actually affect different audiences. Can engagement with entertainment content increase audience demand for problem solving through democracy versus through action by autocratic leaders, for example? We hope to find out by the end of the year.

Join our mailing list to stay in the loop, or reach out to start a conversation about participating in this study.