The Problem With America’s Protest Feedback Loop
Mass demonstrations are becoming more frequent but less effective.
![A megaphone radiating sound waves](https://cdn.theatlantic.com/thumbor/2J7asGyFANbgJJfrgPLRzZC_twE=/0x0:1300x731/960x540/media/img/mt/2024/05/megaphone/original.gif)
The country is stuck in a protest feedback loop. In recent months, students opposed to the Israel-Gaza war have occupied lawns and buildings at college campuses across the country. Emulating climate activists who have stopped traffic on crucial roadways, pro-Palestine demonstrators have blocked access to major airports. For months, the protests intensified as university, U.S., and Israeli policies seemed unmoved. Frustrated by their inefficacy, the protesters redoubled their efforts and escalated their tactics.
The lack of immediate outcomes from the Gaza protests is not at all unusual. In a new working paper at the National Bureau of Economic Research, Amory Gethin of the Paris School of Economics and Vincent Pons of Harvard Business School analyzed the effect of 14 social movements in the United States from 2017 to 2022. They varied in size: About 12,000 people marched against a potential war with Iran in January 2020; 4.2 million turned out for the first Women’s March. Pons told me that these large social movements succeeded in raising the general public’s awareness of their issues, something that he and Gethin measured through Google Trends and data from X.
Yet in nearly every case that the researchers examined in detail—including the Women’s March and the pro–gun control March for Our Lives, which brought out more than 3 million demonstrators—they could find no evidence that protesters changed minds or affected electoral behavior.
As the marginal cost of reaching hundreds of thousands, even millions, of potential protesters drops to zero, organizers have mastered the art of gaining attention through public demonstrations. Mass actions no longer require organized groups with members who pay dues, professional staffers who plan targeted actions, and designated leaders who can negotiate with public officials. They just need someone who can make a good Instagram graphic. But notwithstanding the clear benefits of social media for protest participants, the lure of racking up views on TikTok or X and getting on the homepage of major news sites can overwhelm other strategic goals. Protests are crowding out the array of other organizing tools that social movements need in order to be successful—and that has consequences for our entire political system.
The contours of mass protest have evolved over time. Researchers have found that since roughly 2010—perhaps not coincidentally, when smartphone adoption spiked—political protests have become more frequent around the world, particularly in middle- and high-income countries. The “size and frequency of recent protests,” one analysis claims, “eclipse historical examples of eras of mass protest, such as the late-1960s, late-1980s, and early-1990s.”
Movements learn. Over the years, social movements have internalized the strategic superiority of nonviolence: More people are willing to join a peaceful march than are willing to join one that includes violent confrontations. The UC Berkeley professor Omar Wasow’s research bolsters the argument for strategic adoption of nonviolence by looking at Black-led protests from 1960 to 1972. Wasow found that violent protests increased Republican support in the electorate and may have even tipped the 1968 presidential election toward Richard Nixon and against Hubert Humphrey, the lead author of the Civil Rights Act.
Much of the academic literature on mass protest focuses on movements, in countries around the globe, seeking to topple a government or win independence. According to the Harvard political scientist Erica Chenoweth, violent insurgencies against state power have declined, while nonviolent movements have become more common. (Chenoweth defines violent resistance as including not just “bombings, shootings [and] kidnappings” but also “physical sabotage such as the destruction of infrastructure, and other types of physical harm of people and property.”)
Yet seeking change through peaceful persuasion has also become less effective. Since 2010, Chenoweth wrote in a 2020 essay in the Journal of Democracy, fewer than a third of nonviolent campaigns, and just 8 percent of violent ones, have been successful—down from about two-thirds of nonviolent insurgencies and one-quarter of violent ones in the 1990s.
Mass struggles have come to rely too much on street protests, Chenoweth observes, and to neglect the “quiet, behind-the-scenes planning and organizing that enable movements to mobilize in force over the long term, and to coordinate and sequence tactics in a way that builds participation, leverage, and power.” Past research by the sociologist Kenneth Andrews on the Mississippi civil-rights movement and the War on Poverty found that counties with “strong movement infrastructures” yielded greater funding for anti-poverty programs; activists in these areas had better access to decision-making bodies and more influence over how social programs worked. “Movements were most influential,” Andrews explained, “when they built local organizations that allowed for an oscillation between mass-based tactics and routine negotiation with agency officials.”
Even under the most favorable circumstances, public protest will never be perfectly orderly. As the prominent sociologist Charles Tilly once wrote, a social movement is not unitary. It’s a “cluster of performances,” a “loosely-choreographed dance,” or even a “jam session with changing players”—all of which, he says, “have well-defined structures and histories, but not one of them is ipso facto a group, or even the actions of a single group.”
Many critics of modern protests are fixated on a picturesque, Tocquevillian vision of democracy—an imaginary world where interest groups always argue respectfully and compromise amiably. This vision isn’t aspirational; it’s fundamentally at odds with how human beings normally behave. Real-life democracy is a marketplace of ideas and emotions and arguments bouncing off one another, scrabbling for purchase in the hearts of voters, the minds of the cultural elite, and the press clippings skimmed by harried politicians.
The Gethin and Pons study about the inefficacy of modern American mass movements identified one glaring exception: the protests over George Floyd’s murder. In the summer of 2020, nearly 2 million people participated in more than 5,000 separate racial-justice protests in the United States. Gethin and Pons found that after the protests, Americans expressed “more liberal answers on racial issues.” They also appeared more likely to vote in the upcoming presidential election and less likely to vote for then-President Donald Trump. This finding about the effectiveness of the 2020 anti-racism protests on the American public is supported by other research.
Policy change did occur in the aftermath of these protests. The Brennan Center for Justice found that, in the year following Floyd’s death, half of American states enacted legislation regarding use-of-force standards, police-misconduct policies, or both.
The Black Lives Matter protests during that period were different in part because they defied the caricature of protesters as radical college students with nothing but time. According to a study led by the Johns Hopkins economist Nick Papageorge, on factors such as gender and race, the demographics of the protests were actually more representative of the American public than the 2020 presidential electorate was.
What’s more striking is that a full third of protest participants identified as Republicans. Underscoring the ideological diversity of the movement, 30 percent of summer 2020 protesters in the researchers’ survey sample had attended BLM rallies as well as demonstrations seeking less stringent pandemic precautions—even though the two causes were widely characterized as coming from opposite sides of the political spectrum.
Another reason the BLM protests succeeded is that they were overwhelmingly peaceful—despite some high-profile outbreaks of violence in cities such as Minneapolis, Seattle, and Portland, Oregon. According to research by Chenoweth and the political scientist Jeremy Pressman, more than 96 percent of the 2020 racial-justice protests resulted in no property damage or police injury, while nearly 98 percent resulted in zero reports of injuries among participants, bystanders, and police.
The Floyd protests did not materialize out of nowhere. The intellectual foundation had been laid by years of previous protests that created some organizational infrastructure and steadily increased the public’s support for the BLM movement until it surged upward in June 2020. Perhaps the other movements in the Gethin and Pons sample will prepare the way for future actions when the circumstances are ripe.
Still, many movements seeking to capitalize on public attention find themselves trampled underneath its power. Media attention flocks to the most radical and provocative elements and emboldens the voices on the fringes. Movement leaders have lost their ability to promote an overall message. Not surprisingly, despite the full slate of potential reforms that could have gained traction after Floyd’s murder, the slogan that everyone remembers is “Defund the police”—a policy demand that represented just a minority of voters’ views even as the majority of Americans were calling for far-reaching reforms of police departments. Who can credibly claim to speak for the campus protesters who oppose the war in Gaza?
Even though nobody knows who the leaders are, some of the protesters’ positions do seem to resonate off campus: Morning Consult polling from late last month suggests that 60 percent of Americans support a cease-fire, 58 percent support humanitarian aid to Palestinians, and fewer than half of voters support military aid to Israel.
Still, other stances taken by protesters—such as pushing universities to divest from companies with ties to Israel or, in some cases, calling for an end to Israeli statehood—have scant support among the general public. And the college protests themselves are widely frowned upon: In another poll from May 2, when asked whether college administrators had responded too harshly to college protesters, just 16 percent of respondents said administrators had responded too harshly; 33 percent thought they weren’t harsh enough.
While even entirely nonviolent protests cannot count on public support, escalatory actions such as trespassing, vandalism, and property destruction undermine and distract from broadly shared goals. People in left-leaning movements know full well that some of their own supporters are undermining message discipline and strategic imperatives. Groups critical of Israel have tried to organize boycotts of a handful of companies that, in their view, have been complicit in harming Palestinians. But among sympathizers on social media, perhaps the most prominent boycott target has been Starbucks, which is not on the list.
Yet even as the burden is on protest organizers to articulate clear, feasible policy and persuade their fellow citizens to go along, everyone should be concerned if protesters whose demands have substantial support fail time and again to register gains in Washington. Civil unrest is inherently delegitimizing to a government. Protests are in part a rejection of traditional methods of registering opinion. Their increasing regularity indicates that people believe voting and calling their representatives are insufficient. In fact, many people who participated in the 2020 protests—both the Floyd ones and the anti-lockdown ones—did not end up voting in the presidential election that year.
In remarks about the campus demonstrations last week, President Joe Biden offered a tepid defense of nonviolent protest, saying, “Peaceful protest is in the best tradition of how Americans respond to consequential issues.” Later on, he added that “dissent must never lead to disorder.”
But the disorder that Biden warned against is not just a matter of college students getting graduation canceled this year; it’s also a matter of some Americans deciding over time that voting may not be worthwhile. Polls suggest that the public is deeply dissatisfied with how the U.S. political system is working. A feedback loop in which demonstrations proliferate to little effect, while radicalized protesters become ever more disillusioned with democracy, is a dangerous one. If you’re worried about the disorder on college campuses now, imagine if Americans lose faith in the power of democratic voice altogether.