Violent Versus Nonviolent Revolutions: Which Way Wins?

During her training as a political scientist, Erica Chenoweth was taught to assume that the most effective tool for achieving political goals is violence. After all, no evil dictator is going to give up his autocratic power without a fight, and throughout history, there have been numerous examples of tyrannical governments viciously crushing their opposition.

This weekend, Harvard psychologist Steven Pinker brought Chenoweth to Arizona State University as part of a workshop on the “Origins of Violence.” The speakers included an impressive array of scholars from around the world, including distinguished anthropologists Richard Wrangham and Rob Boyd, neuropsychologist Adrian Raine, and political scientist John Mueller. Although Erica Chenoweth is substantially junior to those eminent gray-haired fellows, however, she stole the show with her talk on civil resistance. Chenoweth presented not only an argument about why nonviolent revolutionary movements are more likely to succeed as violent revolutions, but also an impressive body of evidence to back up her claims. And she laid out several additional findings to elucidate why nonviolence trumps nonviolence as a tactic.

Chenoweth and her colleague Maria Stephan painstakingly collected data on 323 violent and nonviolent political campaigns since 1900. To qualify for the analysis, the movement had to be substantial in size, involving at least 1000 people active in the movement. They counted a campaign as successful if the goal had been achieved within one year of the peak of the event (as when Corazon Aquino and the People Power Revolution peacefully ousted dictator Ferdinand Marcos from the Philippines in 1986).

When Chenoweth started out, she was fairly certain that the violent political campaigns would be more likely to accomplish their goals. But she was wrong.

Nonviolent campaigns have a 53% success rate and only about a 20% rate of complete failure. Things are reversed for violent campaigns, which were only successful 23% of the time, and complete failures about 60% of the time. Violent campaigns succeeded partially in about 10% of cases, again comparing unfavorably to nonviolent campaigns, which resulted in partial successes over 20% of the time.

Why the difference? As Chenoweth and Stephan lay out in their book Why Civil Resistance Works, there are several interlinked answers. First, nonviolent campaigns typically attract more participants, including women, elderly folks, and others who do not want to take on the risks or the moral burdens of running around with guns and explosives, but are willing to pass on information about government atrocities, and to engage in boycotts, strikes, or nonviolent protests. Second, when a tyrannical government acts to suppress a nonviolent movement, it is more likely to backfire. Government security forces don’t want to fire on unarmed civilians, especially when the crowd might include their mothers, daughters, friends, and neighbors. And if unarmed civilians are attacked, other citizens are likely to mobilize, and the government loses support from the international community and from the other pillars of its own society, such as the local media and the financial sector.

And Chenoweth had more good news: When a government is overthrown nonviolently, the new government is more likely to be democratic, and less likely to itself be overthrown, as compared to those that won using guns and bombs.

All of this raises questions about the wisdom of government policies that involve sending arms to revolutionaries, who often replace the current violent and tyrannical government with another one (eliciting longstanding hatred for the governments that helped the current dictators take hold).

Douglas Kenrick is author of The rational animal: How evolution made us smarter than we think. (with Vlad Griskevicius), and of Sex, Murder, and the Meaning of Life:A psychologist investigates how evolution, cognition, and complexity are revolutionizing our view of human nature.