Why Most Protests Fail ft. Erica Chenoweth | Prof G Conversation

If the pattern obtains in this case, then we would expect a loss in the midterms. Now, I also think none of these models have predictive power, right? If what we saw in 2017 and 2018 tracks in this case, then we would expect in a normal kind of election year and under normal circumstances to see a similar outcome in a 2026 midterm.

I'm in Cambridge today.

I think the most relevant research here is about what is it that makes movements more likely to succeed or fail. And the sort of synthesis I would offer from a huge range of research studies on this, my own and others, is that there are really four things that make movements more likely to succeed than others. The first is very large and diverse participation that builds momentum. The second is the ability to leverage that participation into creating defections within the opponent's pillars of support, whether those are the sort of political institutions, the social and cultural institutions, security forces, the business and economic elite that uphold, you know, a sort of authoritarian status quo, if you will. The third is the ability to shift between methods of protest to methods of non-cooperation, like your campaign, and methods of alternative institutions or building mutual aid networks and things along those lines. And then the fourth is the ability to maintain resilience and discipline even as repression against the movement escalates. And the movements that do those things well tend to succeed more often than the movements that struggle with one or more of those four factors.

I think getting defections is the hardest. In a study that I did with Zoe Marks and Andrew Hawking, we did a sort of computational study that ran three different types of strategies. The first is the mass mobilization strategy: just get as many people in the streets as possible, as quickly as possible, and hope for the best with regard to pillars defecting. The second strategy was what we called a naive pillar strategy, which is, you know, protest in a way that's trying to get those defections, but you don't know in advance which pillars are most likely to defect. You just go to the nearest one and protest and hope for the best. And then the third strategy is what we call an informed pillar strategy. And that's where the activists have some advanced information about which of the pillars are already kind of on the fence and they focus on those first. And that then creates early defections in the movement that can create a cascade of defections. And that third strategy is by far the most likely to succeed in the least amount of time. And the first strategy, the mass mobilization strategy, is to get as many people into the streets as possible as quickly as possible and hope that the pillars defect, is the least likely to succeed. And so I think part of it is just it's actually very challenging both to build a strategy that creates defections and to understand, you know, how to do that in a way that creates the cascade of defections, not just an occasional defection here or there because they happen to, you know, push the right button one day.

Yeah. So I mean, my sense is that the No Kings protests — and by this let's just talk about the national days of protest, right, which is not all that it is, but the protests themselves — I mean, we are seeing growth in numbers over time. We're seeing a diverse range of people from all walks of life coming out and participating in those protests. You know, you could argue it's building momentum in the sense that there are lots of protests that happen between those very large-scale days of action. For example, my team at the Crowd Counting Consortium documented that June 2026 actually had the third most protest events in a month in the entire first and second Trump administrations. And that was largely in reaction to ICE in Minneapolis and the killings that happened there. But that just speaks to the fact that the pace of protest and the pace of protest mobilization is quite high and increasing over time even between these big days of protest. And then when it comes to defections, you know, I think that there are kind of cases where we see that happening. Then there are cases where we don't and where you'd otherwise expect them to happen. And I think that's a pretty natural thing to expect at this stage of a movement, which is to say, you know, the mobilization against, you know, autocratic consolidation in the US has been going on now for the entirety of the second Trump administration, but the average movement takes about two and a half years or three years to sort of run its course in creating the defection cascades that are sort of necessary to bring about a pro-democratic outcome. And so I think it's sort of on pace, if you will, but still with a way to go. And then when it comes to diversity of or sort of a broader range of methods of protest, so it's clear that the large-scale days of action are important in both bringing in new participants to the movement and in encouraging existing and new participants about the power of their collective action. The key here is like channeling it into political power through encouraging people then to meet in their communities to build more community-based organizations, to channel it into electoral power, to potentially channel it into methods of non-cooperation and other strategies of dissent. And then the last thing I'll note is I think it probably is one of the most disciplined movements we've seen in recent years in terms of the ability to not overreact to provocations and to maintain basically non-violent discipline even as repression has been escalating in a way that signals that the movement is both prepared for that repression and that when the repression happens the movement can respond to it in a way that shifts the balance of power rather than sort of succumbing to potential disarray or indiscipline.

It really depends on where it's happening for one thing. So I think the people who have actually done surveys and done demographic counts in those surveys have noted from some of the big cities that they've seen that demographic description that you just gave. But you know there were over three, probably over 3,000 events that happened on Saturday, many of them in places that have never seen a protest in the last generation. Many of them...

Exactly. And I think really there's also kind of a convergence, I would say, in some places where ICE operations have been very intense that has drawn in a lot of younger people and people from different walks of life into the movement. And so I actually think this last No Kings saw the convergence of, you know, it was probably more demographically diverse than the prior in part because of how many streams of opposition and resistance are kind of converging in those big days of action. So clearly like the more people from the more walks of life who take part the more powerful the movement will be. And it's often the case that movements, you know, different formations appeal to different demographics. And the more those formations come together, the more powerful they will be in shoring up the capacities of those different segments of society.

You could be on to something in some cases. I would just note that there's some interesting research out there that shows that, for example, the Women's March of 2017 — the participation in Women's Marches just in that single day had a really powerful predictive effect on the diversity of candidates that ran in the midterm elections in 2018 and the blue wave that came in in 2018 flipping Congress to be a Democratic Congress. And they can show a strong correlation between the numbers of people who participated on that day and the outcome of that election. And the same is true for the 2020 election actually. That large participation in Black Lives Matter protests in the summer of 2020 were also correlated with outcomes in the presidential election results and in shifts in public opinion and in changes that were substantive around, for example, progressive attorneys general or DAs being elevated in local and state elections. So, you know, I think that there are reasons to be skeptical, but also those types of stories aren't necessarily very prominent out there in the world. And I think it can be, you know, really important to elevate the ways that even a single day of protests can have those types of electoral impacts. Same for the Tea Party. The Tea Party protest, there was a paper by economists that showed that it was powerfully correlated with the 2010 midterm election result. Right? So people shouldn't underestimate how important it is and they shouldn't overestimate how important it is compared to the other things that also lead mass movements to succeed.

Yeah. I mean, if the pattern obtains in this case, then we would expect a loss in the midterms. Now, I also think none of these models have predictive power, right? So lots of things can be different, but I think yes, if what we saw in 2017 and 2018 tracks in this case, then we would expect in a normal kind of election year and under normal circumstances to see a similar outcome in a 2026 midterm.