The Exchange: Erica Chenoweth and David Scheffer on When to Get Violent

Political scientist Erica Chenoweth, like many of her peers, had always viewed violent movements as strategically more successful than civil resistance at changing a regime or expelling foreign occupiers. When she started scrutinizing the data, however, she discovered exactly the opposite: Nonviolent movements were, in fact, two times more effective than their weaponized counterparts. Nevertheless, political violence is still rampant — a gory reality with which David Scheffer, as the U.N. secretary-general’s special expert on assistance to the Khmer Rouge trials, is all too familiar. From 1997 to 2001, he served as the first U.S. ambassador at large for war crimes and led the U.S. delegation in talks to establish the International Criminal Court. Scheffer and Chenoweth (Global Thinkers in 2011 and 2013, respectively) recently connected in Foreign Policy’s recording studio in Washington, D.C., to debate the efficacy of armed struggle and what justice in Syria might look like.

Erica Chenoweth: My finding that civil resistance tends to be, on average, a more effective method of political change than violent resistance still holds up. And that might surprise a lot of people who are intrigued by headlines about violent insurgencies around the world. But when we actually count side by side the effectiveness rates of such campaigns, one of the most striking things is that now nonviolent resistance is actually about three times more effective than its violent counterparts. That said, the absolute rates of success have gone down a little bit, but the absolute rates of success for violent insurgencies have dropped dramatically. That is under 10 percent in the past five years. Also, I’ve sort of modified the formulation, by watching different campaigns unfold, for what makes them work. Certainly participation rates are the most important indicator of a successful mobilization. But I’ve become even more sold on the necessity of factors like whether the campaign is able to maintain nonviolent discipline even in the face of repression, whether the campaign is able to shift between techniques like protest to techniques like stay-aways or boycotts or strikes, and whether the campaign is able to elicit loyalty shifts among elites. One of the key assumptions going into understanding why civil resistance works is that no opponent is monolithic — even the most brutal, authoritarian leader relies on people to stay in power.

David Scheffer: I don’t have the empirical background, but I can tell you from experience that civilians actually have the advantage within society with nonviolent resistance. They have the disadvantage if they turn to military resistance. Why? Because the government typically has the military forces, so the government clearly has the advantage in terms of soldiers and in terms of the assets of a military regime. And so, it’s extremely worthwhile for civilians and for groups within society that are under repression and seeking to find a way to resist, particularly, a regime that is imposing a tremendous range of human right violations upon that population, to find a nonviolent means of resisting. And that’s where I think so much of Erica’s research benefits the discussion, because it shows us that, empirically, it works. Just consider the linkages that a nonviolent resistance movement can have transnationally, building support for that nonviolent resistance movement. We saw a little bit of this in the Arab Spring, although the end results were not as hopeful as everyone had hoped they would be. But nonetheless, there was a tremendous amount of social media and of other forms of communication and linkages with the international community, and quite a warm response to a lot of the nonviolent resistance, which we saw spur the Arab Spring forward.

DS: When violence is imposed upon a civilian population through atrocity crimes, there is a break point when the civilians whose lives are at risk have to turn from nonviolent resistance to how they can persuade receptive governments and the United Nations to respond, so as to actually save lives. So when do these civilians transition from a nonviolent mindset to armed resistance and insurgency? When is it most valuable to do that? All of this, in turn, raises a very interesting discussion about the responsibility to protect, a principle which has been around since 2005 and which is mostly about nonviolent measures to protect civilian populations. If all else fails, there is the invitation of the Security Council to step in and use military force to, in fact, protect that civilian population that is resisting, though we don’t have that formula down well in execution at all.

EC: I’ve grappled with these issues a lot with myself. My sense is that if it is too inhospitable for a civilian population to organize and effectively wage nonviolent struggle, I don’t know how well they’re going to do with violent struggle either. And, in fact, the turn to violence might unleash a level of repression that the nonviolent struggle had kept at bay, simply because it was too risky for the opponent to order such mass atrocities directly against that population. It’s kind of conventional wisdom now that if you want to avoid mass killings, avoid a war. There are very few cases where there are mass killings on the order of more than 1,000 people in a single episode that are directed against nonviolent mass campaigns. So why is it that nonviolent mass movements would be so much less vulnerable to this? The answer is because usually it’s an authoritarian regime we’re talking about. If leaders are facing a popular uprising, they know that if they give an overt order to their subordinates to put it down, they risk them saying no; and that’s the endgame for them. So many times they won’t do that; they’ll see if those actors will take the initiative on their own. And often they won’t, because they want some cover from above. If they do give the order, what we find in the data is that about half the time the security forces do disobey. It really is kind of a survival decision to not give the order to kill people involved in a mass demonstration. Of course it happens from time to time, but it’s much rarer than the types of killings against civilians that occur in the context of armed struggle. So I’m a little dubious that there are points at which the international community should be reflecting back to these populations, “You’ll be at less risk if you actually start moving to armed struggle now.” Violence usually makes the situation much worse; it also kind of creates perverse incentives to accelerate a humanitarian crisis and all kinds of strange dynamics about when we deal with things like responsibility to protect. How often has the international community really gone in to protect civilians while they’re in the context of a nonviolent struggle? If you start going in to protect civilians during unarmed struggle, then it creates all kinds of difficulties with discrimination and who deserves it and other things.

DS: If you look at international law or at the international treaties of the last 70 years, there is a clear recognition of the right of self-determination and a protection, more or less, for national liberation movements within a whole body of international law. What it seems to do is to give license to use violent resistance for this principle of self-determination — to insulate national liberation movements from international humanitarian law and the laws of war, essentially. If you look at nonviolent expressions of self-determination, and we’ve seen it in Europe — Scotland and Catalonia, et cetera — then that’s a totally different picture. But if you’re still advancing violence in the name of self-determination, whether it be internally or otherwise, international law is in a state of confusion about that point. 

EC: One of the things this brings to mind is the fact that we have some 1,000 years of international law, both codified and customary, relating to the laws of war. But when is it OK to use it? What are the boundaries? We don’t have a ton of criteria about when it is OK to wage mass nonviolent struggle. I’ve been thinking about this lately, with regard to some of the liberation movements or mass movements that have maybe not had very democratic goals. This raises questions about what are legitimate uses of consent and the withdrawal of consent by the population.

DS: What deserves more empirical research in the academic community is how one calculates the cost of these conflicts. With the migrant crisis now flooding through Europe, what is the ultimate cost of all of this versus what might be the cost to take more effective, preventive measures at some point during this five-year saga in Syria? It would be interesting, perhaps as negotiating leverage, to somehow compile the cost of what has been unleashed and then actually just present a bill to President Assad and say, “This is your bill. We’re going to take the assets that have been seized, and we’re going to apply it to the bill that has been generated out of this conflict.” In thinking about prevention of atrocity crimes, it would be very helpful in the future to have sort of a building databank that would enable policymakers to say, “We can let this fester for another month, for one or two more years. If we do so, the best empirical calculation is that the cost of cleaning all of this up — of rebuilding that society, of resettling the refugees back into that society — will probably be X. If we act now, we might be able to avoid a substantial amount of that.”

EC: I always think that it’s useful for us to be able to calculate, to the degree we can, what the various costs are — primarily and most importantly, human life. But also, that would be very tricky, just from an empirical perspective. First of all, it’s always going to be based on counterfactual analysis: educated guesses about what might have been with other scenarios where there were lots of moving parts and lots of contingencies. Then the second issue is that states often intervene in places where they already know they can make a difference or where there aren’t a lot of friends protecting the country anymore. This isn’t always the case, and there’s variation, of course. But if you think about the Libya versus Syria example, the thing that many Syrians couldn’t understand is why NATO would protect civilians in Libya and not in Syria. They were saying, “Syria is such an important country. You have to go and do something.” And the paradox is, exactly, it’s too important a country.

This conversation has been condensed for publication. Listen to the full discussion by subscribing to FP’s Global Thinkers podcast on iTunes or by visiting here.