Two Logics of Political Action: The Structure and Social Correlates of Democratic and Coercive Action Likelihood
Based on a survey of 34,300 U.S. adults, this study reveals that support for political action is best understood through two distinct dimensions—democratic participation and coercive violence—rather than a single continuum, showing that while a minority express conditional willingness to use violence for specific causes, this tendency is driven by differentiated factors like vote choice and religion rather than broad economic grievances or geographic context.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Political scientists have long tried to understand why some people turn to violence to settle their differences. A common assumption is that political action exists on a single ladder, where voting and attending a peaceful protest are just the lower rungs, and violence is simply the highest, most extreme rung. Under this view, anyone willing to shout at a government official is just one step away from being willing to hurt someone, and the same feelings that make a person vote should also make them willing to use force. This idea matters because if we believe violence is just the end of a long line of normal political behavior, we might look for the same causes—like anger or economic hardship—to explain both voting and killing. But if violence is actually a different kind of behavior entirely, driven by different reasons, then our understanding of how societies fracture needs to change completely.
Two researchers at RTI International decided to test this idea directly by asking a massive group of Americans how likely they would be to take specific political actions. They surveyed more than 34,000 adults, asking them to imagine fourteen different causes that people care about, ranging from protecting the environment and gun rights to defending religious beliefs and even protecting the country from illegal immigrants. For each cause, the researchers asked the same question: if taking a specific action would protect that cause, how likely is it that you would do it? The list of actions started with the mundane, like sharing information or donating money, moved through voting and attending a protest, and climbed up to the extreme, including harassing officials, destroying property, threatening someone, seriously hurting someone, and killing someone. The goal was not to ask if people approved of violence, but to see if they could imagine themselves doing it if they felt it was necessary to protect something they valued.
The results overturned the idea of a single ladder. When the researchers analyzed the answers, they found that people's responses did not fall along one straight line. Instead, the data split into two distinct but related groups. One group consisted of democratic actions: sharing information, giving time or money, voting, and attending protests. The other group consisted of coercive actions: harassing officials, destroying property, threatening, hurting, and killing. A person could be very likely to vote and protest but very unlikely to ever consider violence. Conversely, a person might be unlikely to do any of the democratic things but still see a chance of using force in a specific situation. The study showed that these two types of behavior are driven by different factors. For instance, feelings of economic unfairness were linked to a higher likelihood of protesting, but they had no connection to a willingness to use violence.
The study also revealed that the way we count support for violence changes the story entirely. If you look for people who say they are likely to use violence across all the different causes, the number is very small, at just under four percent of the population. However, if you ask if anyone has said they might use violence for at least one specific cause, the number jumps to nearly forty-two percent. This means that while very few people are consistently ready to use force, a large minority of people can imagine a scenario where they would. The researchers found that this willingness to imagine violence is not evenly distributed. People who voted for one major party candidate in the 2024 election were significantly more likely to rate violent actions as possible than those who voted for the other candidate, even though both groups were equally likely to say they would vote or protest. This suggests that the political divide is not just about how active people are, but about what kinds of actions they consider acceptable.
Religious identity also played a surprising role. After adjusting for other factors like politics and income, the study found that people who identified as Muslim were more likely to say they might use coercive or violent actions to protect a cause than those from other religious backgrounds. However, the researchers were careful to note that even for this group, the average likelihood was still low on the scale, falling between "no chance at all" and "unlikely." The difference was a matter of degree, not a switch from zero to one. The study also looked at whether people who felt their social status was declining, such as white men in areas where the white population was shrinking, were more likely to support violence. The evidence for this was weak and depended heavily on how the data was sliced, suggesting that the simple story of "status threat" driving violence is not supported by this large dataset.
Perhaps the most striking finding was that where a person lives mattered very little. The researchers checked data on county-level economic decline, changes in population, and shifts in political voting patterns. None of these local conditions helped predict who was willing to use violence once the individual's own characteristics were taken into account. The factors that mattered were personal: who a person voted for, their religious affiliation, and their military service. The study concludes that political violence is not just the extreme end of normal participation. It is a separate dimension of behavior with its own unique causes. Understanding this distinction is crucial because it means that the forces driving people to the polls are not the same forces driving them toward violence, and solving one problem will not automatically solve the other.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.