How to Prevent a Political Coup

An Interview with Josef Woldense

"Authoritarian rulers are most often overthrown by the people closest to them. My research is centered around the internal dynamics of authoritarian regimes and how rulers survive threats from their inner circle. In my class, we sometimes play a simple game where a designated autocrat tries to figure out who is with them and who is not. Uncertainly and deception is the norm and even the strongest ruler cannot escape this fact. Now imagine that you are the ruler at the center. Who would you trust and how do you keep your power?"

Josef Woldense, assistant professor in the Department of African American & African Studies and affiliated with the Department of Political Science, researches how dictators maintain power while keeping the regime productive. Specifically, he has been taking a close look at the regime of Ethiopia’s ruler Haile Selassie.

Though authoritarian regimes have dominated much of human political life for thousands of years, little is known about how these regimes operate. By studying mundane details like job appointments and firings, Woldense is mapping how Selassie “shuffled” people within his inner circle to prevent his advisors from forming alliances and getting too much power, tactics that are still used in corporate organizations and governments.  

Can you describe your research and explain why you chose to study it?

My research attempts to map out the internal dynamics of Haile Selassie’s authoritarian regime in Ethiopia, such that we can get a better insight on what's happening and what's driving things inside a regime. That goal is what initially got me going, and since then, it's been a fascinating subject matter. 

Studying authoritarian regimes is really studying organizations more generally. You realize once again just how little we know about them, despite authoritarian regimes being much older than democracies or democratic regimes. 

I'm originally from Eritrea, which at the time was part of Ethiopia, so an opportunity to do research in that region was really an opportunity for me to learn more about my own background. Once you start a doctorate program, there's both your own desire of what drives you, but also the pragmatic questions of, can you get data, an you pull off a project? My research, which focuses on the last emperor of Ethiopia, Haile Selassie, and his regime, is the marriage of those two things. 

Can you explain the historical and political context of your research?

Haile Selassie was the emperor of Ethiopia from 1930 until 1974. He was already a king in the making in the 1920s, and what made him unique among the Ethiopian elite was that he realized the importance of not just engaging Europe militarily, but also diplomatically and directly. So, with the ambition of wanting to “modernize Ethiopia,” very much in the vein of Japan, he was able to get Ethiopia to become part of the League of Nations in 1923. 

Selassie officially became emperor in 1930 and then in '35, Italy invaded. He was in exile in the UK for five years, and then returned in 1940 and from there ruled into 1974. He was in power for quite a long time.

And, according some historians, the Italian invasion of Ethiopia in 1930, lead by Mussolini, was essentially the beginning of World War II, because Ethiopia was part of the League of Nations, and one of the core principles was that none of the members of the League of Nations were to attack another member, and yet that's exactly what Italy was doing. With that, the League of Nations lost its credibility and Germany began to invade other countries, and so on and so forth. 

How did you collect the data on Selassi for this project?

What's special about Haile Selassi in terms of the research that I'm doing is that the data that I'm collecting is very mundane and very detailed. I review the granular data in appointments by reviewing the official government log, where appointments of officials were listed on a monthly basis. I literally transcribed each appointment for I think around six months or so, which was very exciting at first, but then you get to like row 3500 and each entry then becomes more complicated, and you have to cross-check and make sure it's the same person. 

Without understanding the broader context, this data would be almost meaningless. So I was lucky in that there has already been a rich secondary literature, both scholarly work and quite a few autobiographies, written by former officials of Selassie's regime. 

But what was awesome about the combination of these data sets is that you could begin seeing someone's trajectory, literally following along their careers. With that data, I was able to reconstruct the administration at any given point in time, so I know who's working with whom, where they're working, where they’re going, and all these kinds of things. This data allows us to begin to look into parts of the regime that previously we only made conjectures of.

The kind of research that I'm a part of, the community of scholars that are using this granular appointment data, is still quite small. Scholars on China are definitely leading the way, and scholars on Russia and Mexico use this data as well. And the field is growing, absolutely growing. When it comes to African politics more generally, I'd say my work is among the first to really delve into this level of detail when it comes to appointments.

As one of my colleagues, Francisco, said, “essentially what you're trying to do is use publicly available data to get insights on what is otherwise hidden.”

Can you explain the concept of “shuffling” and how it prevents others from gaining power?

First, let’s establish that there is no such thing as an all-powerful ruler, at least not in the conventional way that we like to think of it. We tend to think that when the ruler says jump, then everyone jumps. If the ruler says sit, everyone sits. This person is so powerful that all the consequences, all actions related to the regime, can be attributed to that one ruler. 

You don't even have to talk about authoritarian regimes here, take any organization, take any boss, take even a mom and pop store. Ask yourself, even in that small setting, do you think that the owner of the mom and pop store can control everything that the employees are doing? No. There's all kinds of stuff that's going on that the boss doesn't know about. And that example is an incredibly small setting. 

Now amplify that, more, and more, and more. Now you're talking, you know, thousands of people engaging in a very complex organization. By all means, rulers have a lot of power. Haile Selassie, for instance, had power over appointments and could hire you, could fire you, could move you, he could do all kinds of things to you. But does that mean that he could, for instance, stamp out certain behavior, or opposition, or conspiracies? No. So in that respect, you’re powerful, but you're not all-powerful.

So the biggest challenge for rulers, for any organization, is the formation of cliques within your own organization. These cliques are informal groups that form where people who have been in close proximity with each other, gotten to know each other, maybe shared the same ideology, maybe family ties, begin to cooperate. These cliques essentially represent an alternative center of power, and in a context where power matters, when that is the ultimate currency, well, the stronger and more powerful these cliques become, the weaker the ruler will be in relation to them. 
Shuffling, which we know to be an age-old strategy by rulers, attempts to dissolve these cliques. Think of it as a technology. Shuffling recognizes that the fuel for cliques to form is people being in close proximity to each other and having the opportunity to get to know each other, and disrupts that process. So as people are getting to know each other, but before that relationship matures, what you do is you divorce people from one another by essentially having them move into different parts of the regime. They're still part of the government, but they never get a chance to get too close to each other.

You would also see this in, for instance, police work, where the problem would be that certain officers would be stationed in one location for too long. They started to build relationships with the community, which is good on the one hand, but if suddenly the police officers get too cozy with certain corrupt criminal elements of that community...well, that's a problem. So you would use the strategy here as well, where shuffling systematically prevents, or at least makes it more difficult for, these relationships to form. So that's why shuffling is such a prevalent strategy we see employed. 

The problem with that strategy, however, is that you might go ahead and shuffle people around, but what happens if people don't get a chance to stay in one position for any period of time? Well, they're never able to gain any kind of experience, any kind of expertise. And you want people to be able to do their job. For the function of your state, you need folks who know what they're doing. So this very same strategy that enhances your safety also has the potential to undermine the capacity of your state apparatus.

Essentially, the argument I make is that what Selassie was doing was divorcing people from one another to ensure his own safety, to weaken cliques, yet allowing people to gain a broader sense of expertise, which benefited him in the long-term. 

How can we apply this historical research today?

Certainly, the kinds of dilemmas that authoritarian rulers face are the kinds of dilemmas that you find any boss in an organization faces, primarily because the structures of organizations and firms resemble authoritarian regimes much more so than democracies. This matters in terms of the kinds of dynamics that you see. 

One, for instance, is the incentive of getting the truth out of your subordinates. If you have the power to determine their fate and then ask them, "Is there something wrong?" what is the incentive for them to say "Yes, there is something wrong. I messed up"? None. They have an incentive to say whatever you want to hear. If you want to hear that everything is going well, they're going to come and say, "Yes, everything is going well." 

The problem is that you are the “all-powerful ruler,” but you are unable to get the most basic information precisely because you have the power over people's livelihood. And so that kind of dilemma you'll find in all kinds of different settings.

You’ll also find all the different ways that organization leaders tried to deal with it: maybe instill some kind of trust; maybe give up some of that power; make clear that, you know, there will be no consequences for giving bad news, or maybe even having others do the research for you. 

What did you learn from examining Selassie?

One thing is the extent to which one has to be invested in order to pull off a regime. Being an authoritarian ruler is much more difficult than to be a ruler of a democratic regime because you have to do so much more. You have to always be vigilant of these cliques, and there are all these moving parts that he in some ways was orchestrating.

We still know very little about authoritarian regimes and generally speaking, we're still very ignorant of how these processes unfold, but as more and more research has been done that’s starting to change.

This interview is part of a series that highlights faculty research at the University of Minnesota Twin Cities College of Liberal Arts. See other videos in this series.
Share on: