Assistant professor of political science Evan Perkoski, studies terrorism, insurgency, and uprisings both violent and non-violent. He studies how armed groups like Al Qaeda and the Islamic State organize and often break apart, how they operate and employ violence, and how they interact and sometimes ally with one another. More recently, he has been studying why states use violence against their own civilians, particularly in the context of popular uprisings. He is the co-author, along with Erica Chenoweth, director of the Nonviolent Action Lab at Harvard’s Carr Center for Human Rights, of a new study, “A Source of Escalation or a Source of Restraint?: Whether and How Civil Society Affects Mass Killings (PDF),” part of a United States Holocaust Memorial Museum research project called The Role of Civilians and Civil Society in Preventing Mass Atrocities. He spoke with UConn Today about the new study.

Mass killing is an issue that has existed for thousands of years. How did this study come about?

The impetus came from the [2010-12] Arab Spring. There were all these uprisings — violent and nonviolent – and we saw lots of variation within them, in terms of which succeeded and failed, and lots of variation in terms of what governments did in response. We saw some governments crack down on and kill thousands of nonviolent protesters, like in Syria. But we saw others maintain restraint, like in Tunisia. So, the idea for this project came partly from international events, but it was the White House that actually got it started. President Obama in 2011 created the Mass Atrocity Prevention Board, since he and his advisors basically recognized that they were blindsided by the Arab Spring, and they had little ability to anticipate where uprisings would happen and which governments would crack down. So the president got this group together, and they asked researchers like Erica Chenoweth and myself to start studying them to see if we can develop accurate models to statistically forecast when mass killings will happen in the context of national uprisings like those in the Arab Spring.

You are using two basic data sources and note they have limitations: the Integrated Network for Societal Conflict Research database on mass killings, which sets criteria for what constitutes a mass killing event; and the Targeted Mass Killings data set from Australian National University, which tracks more than 25 or more fatalities in a given year. How accurate are these resources?

This a perennial challenge when you study international relations and particularly conflict. Yes, these things are visible, but governments also have incentives to make them slightly less visible. We know that the data, the news stories and the information we’re getting from North Korea and China are less reliable than the information we’re getting from the United States and Europe. Despite that, there are constantly efforts by political scientists to collect data and categorize things going on. We’ve had a big push in the last few years to collect better, more micro level data on violence whether it’s mass killings, what rebel groups are doing and how many people are joining rebel organizations. This is following in the footsteps of previous efforts to collect information on things like democracy, income and welfare and other metrics that led the way for what we’re seeing now in the realm of conflict studies. And while there are huge problems with visibility, we do things to make sure we’re getting the most accurate information as possible. I helped to collect one of the data sets we use in this study, the Nonviolent and Violent Conflict Outcomes data. We had a team of undergraduate researchers looking at news sources, academic reports, government reports, and country reports from NGOs like Amnesty International. Because these things are difficult to see, we try to use as many different sources as possible to make sure we’re getting the most accurate picture possible. We know it can’t be perfect, but we try to do the best we can with as much information as possible.

Understanding that there are subtleties and complexities because of the information that you gathered, what are your major findings?

We focused on the strategic interaction between dissidents and states to see when mass killings would happen, because a lot of the existing research focuses solely on the state perspective — Is it a dictator who’s ruling the country? How long has a dictator been in power? Do they rule over the country with a military-based regime or some other power source? We concluded that the choices that dissidents make matter and they shape whether states resort to violence. So if you’re in the Arab Spring do you choose violence or nonviolence? Do you request the United States’ support or not? How broad do you make your campaign and how many people do you get involved? We thought all those things would matter and we found that they do.

After our first study on this topic came out, the U. S. Holocaust Memorial Museum in Washington, D.C. asked us to contribute to a project on the relationship between violence and civil society to learn how different types of non-state organizations, such as community groups and charity organizations, are related to violence. In other words, can we use civilian organizations to prevent violence from happening or from escalating? In that project, we looked at the different metrics of civil society and we found that, in general, civil society doesn’t have a huge effect on its own, but in cases where you see lots of political inequality and it’s institutionalized, you can see civil society being linked to really bad outcomes. So in places where you have political power distributed by wealth, race, ethnic group or by religion, those are contexts where people can mobilize and sometimes do really bad things.

You separated the actions of the optimists and the pessimists within civil society looking for the same outcome, but going about it in different ways. Could you talk about that?

There are kind of two camps in the existing research on the effect of civil society. You have optimists like Secretary of State Mike Pompeo, who has argued for strengthening civil society in Belarus to help contest sham elections and overthrow one of the last dictators in Europe. These people think civil society is great for states. They can control them and reign in their worst impulses. On the other hand, there are pessimists who see lots of cases where civil society has been activated for bad purposes too, such as in Nazi Germany, where it wasn’t just a state that was committing violence against Jews, but you had individual German citizens who were sometimes identifying Jews and participating in Kristallnacht and other terrible events. The same thing happened in Rwanda and many other countries around the world when mass killings have happened. These two camps have diametrically opposed views, and we wanted to see which one is really the most correct when it comes to mass violence. We found that neither one offers a great explanation on its own, but you do see civil society operating differently according to the levels of inequality in the state. The more unequal, the worse the civil society can be, but if you have a representative egalitarian state, then that civil society could be a force for good.

You also cite in the conclusions that for decades, the United States has been encouraging civil society to get involved as a way to spread democracy and solidify it. At the same time, there are more questions raised about the effectiveness of that. Could you discuss what you think that means given this finding?

This is something that the U.S. government and policymakers around the world often do. They latch onto something that inherently seems good and that’s intuitively good, like civil society, and they say, “We’re going to go promote this,” without really giving it the research and the critical analysis that is necessary before you pump a ton of money into these programs. A good analogy for that is democracy. In the early 2000s, the neoconservative foreign policy movement really pushed this idea of democracy as a force for good around the world. It was based on good intentions, but we didn’t really think through the implications of creating democracy in certain states. In Iraq, for instance, this contributed to their civil war. Policymakers basically thought that if we make Iraq a democracy, they’re going to be a much a better country, with better and more peaceful domestic and international politics. But as we saw, when it’s quickly pushed on people who don’t have experience with it and who have significant societal divisions, then it can really lead to some negative outcomes. The same thing is true with civil society. We’re seeing policymakers push it without reservation, but we find that it’s actually pretty complex and that it’s not universally good or bad.

Understanding there is always more research that you can do, what is your hope about how the information from the study can be used?

One lesson from our research is that we need to think carefully about where we support civil society and how we combine it with other policies to make sure we’re doing the most good. Two factors that can make civil society a negative force are the pre-existing levels of political and societal inequality. So we have to think carefully about where we want to strengthen civil society now, and where we want to address some of the lingering political and societal inequalities first. Once these issues are addressed, we can then layer a stronger civil society on top. But policymakers should not think of civil society as a one-size-fits-all solution to foreign problems.

What questions are policymakers asking now that they have the benefit of this research?

People who have read this report having been coming to us with a few particular questions. The first one they’re asking is, how do we build society and support civil society programs when there’s an ongoing crisis? Belarus is a great example. We don’t have the time to go in and work on underlying systemic inequalities while there’s an election being contested and people are in the streets right now. So, our response is that maybe you try to support civil society while also working with the state and with individual civil society leaders to make sure the situation doesn’t escalate. In other words, you can support civil society while being cognizant of, and trying to address, how it could spiral out of control. Other questions we’ve been getting are about the development of civil society. How long does it take to change civil society? Can we do this in a couple of years or is it a five or 10 year project? What do we do in states where there are significant barriers to supporting and developing civil society? And are there means to control some of its worst impulses? For many of these questions we don’t have an answer, and we need to do more research.

Is there a question that you think people should ask you about this study that you don’t get asked?

I would like people to ask whether these findings have implications for everyday life. I think they do. I’m not saying that we should be expecting mass violence or major conflict in the U.S., but one of the broader findings from this study is that it’s better when you have an equal, fair society and when people are active in civil society and with community groups. These are things that states can work on, that individuals can work on. I think working towards this end would have a lot of benefits, especially in light of how polarized American politics has become.

Is there anything else people should know about this study and what you hope to see come out of it?

One thing worth noting is, I think we underestimate how much conflict there is in the world outside of the U.S. It’s very easy to focus on what’s going on at home, especially with so many major events happening here all the time. But most Americans would be surprised at how many mass killings have occurred in the last 10 or 20 years, and in a lot of countries where we don’t really expect there to be such violence. A lot of these are slow-moving, simmering conflicts where civilian deaths have accumulated over a few years and haven’t suddenly spiked like in World War II or in Rwanda. And as a result there are lots of civilians who are very highly repressed, who are facing significant barriers to political participation, to free speech, and to the basic things we take for granted here in the U. S. It’s important to recognize that all this is happening while we’re sometimes overly focused on ourselves.