Theory Talk #24: Robert Bates

Friday, December 19, 2008

Robert Bates on the Politics of Coffee, Small-N Studies, and the Crumbling Definition of the State


Social sciences in general and International Relations in particular is preoccupied with quantitative studies. As a result, individual case studies, or, in methodologically correct language, Small-N populations, have been somewhat discredited. In this Talk, Robert H. Bates, essentially an economic historian, explains amongst others how (single) case studies are just as valid as large-N studies, how coffee can illustrate the workings of international markets, and how the definition of the State should be reconsidered.



What is, according to you, the biggest challenge or principal debate in current IR? What is your position or answer to this challenge or in this debate?


For me, the most important challenge is to recognize what ‘states’, the principal actors in IR, actually are. More people are slowly finding out that states are not stable, but rather equilibriums. To put it very plainly: what we refer to as a state is simply a place where people have agreed to put their weapons down. However, little by little, people from all over (that is, security studies, development studies and International Relations) are realizing that this Weberian definition does not always apply.

Rather, perhaps due to the extension of the concept of ‘security’ since the end of the Cold War, this definition has come under scrutiny. During the Cold War, security referred principally to the possibility of interstate conflict. If there was no risk of interstate conflict, security was not seen as threatened. Since the end of the Cold War, we’ve partly shifted our attention to intrastate conflict elsewhere as a possible source of insecurity, such terrorism in Afghanistan, which could spill over.

Currently, our definition of the state as principally ‘a place where people have agreed to lay down their weapons’ is being challenged by a great number of cases. In Russia and neighboring countries, armed groups are widespread; Italy has its known problems with the mafia; in Serbia, pirates in recent years have hijacked many ships passing the Danube, and many armed groups still exist in the rest of the Balkans. All these groups use non-state violence to manipulate interests. Without even mentioning Latin America and Africa, one has already enough reasons to doubt if the Weberian definition of the state as a monopoly on violence actually exists outside of our normative minds.

How did you arrive at where you currently are in IR?

I am actually reluctant to mention specific people or books that have specifically shaped my thought, also because I have always had the feeling that the most important issues are as of yet not addressed. I think it has always been my own curiosity that motivated me most. One of the experiences that has influenced strongly my thinking on international relations, is the work I did on coffee. I never expected so many politics to be involved, and with such effects, in the market for a product as coffee. Another important experience was the fieldwork I did in rural Zambia. Being there in rural communities, I found out that these people know what they are doing, and that policy makers with ideas from abroad, without local knowledge, do not.

If I were to name some scholars whose work impresses me, I’d surely mention the work of Charles Tilly, who offers a very interesting perspective on the development of states, taking form of government essentially as the result of contention between different factions over the means of coercion and capital. Also, the things David Friedman wrote on the size and shape of nations, interests me a lot. These works interest me especially since I am currently trying to understand the economy of the military and the relationship of the military to the state.

What would a student need to become a specialist in IR or understand the world in a global way?

There are a few things here that I would like to point out. The first one is that in the States, there are these dominant habits of mind that have dumbed our generations down. One of such habits is the relationship the social sciences have had to mathematics. The best mathematicians have tried to stay away from the social sciences while the most exciting work could have been done on this frontier. On the other hand, the dominant way of formal modeling and quantification has led to a straitjacket in social sciences that has doctrinated whole generations of scholars to focus on very limited issues.

One has to look at different fields. I can receive a quantitatively brilliant economics student for a PhD, but I will always ask him: ‘did you read up on law, anthropology, sociology?’ One has to have a broad basis to understand the world. The world is not just math, law, economics or psychology – unfortunately for scholars who love clean models, the world consists of all that and much more. So any explanation should incorporate, albeit just implicitly, knowledge on that complex reality.

Another essential aspect of training is fieldwork. I do not accept PhD students who finish their degrees on Africa without actually having been there. That has to do with my last point: one cannot pretend to generate new knowledge about a complex reality by reading up on it.

In 1998, you co-edited a book called Analytic Narratives. The book expounds a methodology for drawing generalizations from specific historical examples. Could you briefly reiterate the argument made in the book, and indicate if you find it still holds now?

The actual reason for the book was that me and a few other ‘historians’ found we were having the same problem: we were all convinced that you can take a (that is: one) historical example and legitimately extrapolate it to a generalized model. The French Revolution for instance was the case in point for one contributor to the volume. Sure, the French Revolution will never repeat itself, so what’s the use trying to generalize, one could ask. But it is our point of view that one can learn something from history, and we tried to defend a method that proves as rigorous as possible.

The idea is as follows. One has a historical small-N case study – that means, studying a population of one or two cases instead of lets say, 100. It is common wisdom that the bigger the population, the more trustworthy the conclusions about tendencies. However, if one generates a model, that is, infer an abstraction from a small-N case study, the procedure remains basically the same and is thus equally valid: in the end, the validity of the model or theory always depends on its testing. And here, academic rigor comes into play: we have a case, we created a model, and now we have to proceed to formulate testable implications of this model, of the formula ‘if … , then …’. If one can generate testable implications – and test them, even better – whether it be from a big-N or a small-N population, one has in principle a valid scientific study.

A year earlier, you published Open-Economy Politics: The Political Economy of the World Coffee Trade, a book in which you challenge the domestic-international divide and study the workings of international market-governance institutions. First of all: why coffee? And, secondly, how ‘open’ is the world economy at this point?


I actually stumbled on the subject of coffee by accident. I was doing a lot of research on Africa in that period and when the government proposed me to do some research on the subject, I was actually happy to ‘get out of Africa’ for a while. And actually, the whole investigation was a big adventure: the International Coffee Organization, which was the principal object of study, was (because it ceased to exist) located in Colombia – and political tensions rose high while I was there, up to the point that I had to leave because of violence. But apart from that, it was actually very interesting to see how much this institution had in coordinating policy by big coffee producing countries. Every decision made by the organization on policy in situations of market problems such as overproduction had far-reaching and distinct consequences for, for example, Brazil and local producers in Africa. Every decision has different effects on who wins and loses. So the International Coffee Organization constituted a real ‘coffee government’. In my book I try to explain why and how it rose to power, and why it ceased to exist. After having dedicated about a decade to coffee, I was ready to go back again to Africa – now with a fuller understanding of how international dynamics influence local outcomes.

Do countries with an open economy behave differently international from countries which are not ‘open’?

This question is often asked and American policy makers, be it aid-workers or foreign policy specialists, do believe that open economies behave in a certain, perhaps ‘better’, way. The answer is, however, more complicated. An ‘open’ economy refers typically to an economy that is not governed by strong state regulation or intervention, in order to let the market do its work. But the dichotomy ‘open vs closed’ does not reflect actual politics: policy choices are constantly being made, even in so-called open economies – it’s a lot of work to keep an economy ‘open’. By that I mean that there’s always policy choices involved, so labeling a country’s economy ‘open’ or ‘closed’ reflects ideology by the observer. One can only ask: open to which interests?

Your most recent book, When Things Fell Apart: State Failure in Late-Century Africa argues from a game-theoretic perspective that African states did not manage to tax their societies and thus engaged in the path-dependent history that led to current exploitation. How does your explanation go?

I tried to ask in this book: when does a country work? That is, which are stable situations? States make money through violence – either by extortion or by policing. They characterize themselves by their monopoly on violence, meaning that they can use that violence either to engage in taxation by policing or for exploitation using extortion. Societies can either engage in productive activity or not. A functional country is an equilibrium, where the government uses violence to police, and where society engages in production. In Africa, we saw the choice being made (perhaps not consciously) of government exploitation because taxation didn’t work well in a context of global price stagnation. That, in turn, destroys the perspective of society to produce, leading to a relatively stable yet little desirable situation.

We’ve witnessed a relative stability since the 90’s, with very little structural violence and stable prices. But the effects of the current crisis could be devastating, at least in terms of political stability: if global prices drop, society will not be able to survive and rebel against their governments – and since governments do not depend on parliaments or on taxation of society, they will have little incentives to either improve the situation or not to harm society.

You’ve spent a lot of time in Africa, and are considered one of the worlds leading specialists on the subject. What’s the biggest misconception existing about African politics (or political economy)?

The biggest misconception which reigns on Africa is the issue of agency. Many westerners see Africa as a passive continent in trouble that will not move without the active implementation of western best practices. And with that in mind, they engage in analysis and developmental aid. Aid based on that way of thinking can be highly detrimental. I’ll give you an example. Foreign AIDS programs in Africa have the tendency to transfer existing resources away from local hospitals and doctors to internationally established and implemented programs. If in one place a hospital is running with local experts but 20 miles away an internationally funded hospital opens, the local hospital can close and doctors will be pushed out of the market. And in situations like that, engaging in aid is trying to solve one problem while causing another. We feel that we are the solution, they are beggars. That’s simply not true. I, for one, owe a lot to Africa – a whole career and loads of happiness – having added little to the over-all balance.

Robert H. Bates is Eaton Professor of the Science of Government at Harvard University. His research focuses on the political economy of development, particularly in Africa, and on violence and state failure. Bates has conducted field work in Zambia, Sudan, Uganda, Kenya, Columbia and Brazil. Bates currently serves as a researcher and resource person with the Africa Economic Research Consortium, Nairobi; as a member of the Political Instability Task Force of the United States Government; and as Professeur associé, School of Economics, University of Toulouse, where he has taught since 2000. Among his most recent books are Analytic Narratives with Avner Greif, Margaret Levi, Jean-Laurent Rosenthal, and Barry Weingast. Princeton: Princeton University Press, 1999, Prosperity and Violence, W.W. Norton, 2001, and When things fell apart (2008).


Related links


  • Read Bates’ Institutions and Development (2006, Journal of African Economies) here (pdf)
  • Read the Introduction of Bates' 2008 book When Things Fell Apart here (pdf)
  • Read Bates’ Ethnicity, Capital Formation, and Conflict here (pdf)

Print version of this Talk (pdf)