Monday, August 8

Stathis Kalyvas: “Many political scientists only used Afghanistan to thrive”

His book ‘The Logic of Violence in Civil War’, published in 2010, was a true conceptual revolution on human motivations. In it, Kalyvas (Corfu, 1964) postulated that violence was a process that was more due to human aversion to it rather than the result of fascination with it. This explained the cruelty between people who knew each other well (relatives and neighbors) in civil strife. In the end, it was not chaos and anomie that ruled these wars, but a degenerate manifestation of excess social capital.

– Why has what happened in Afghanistan shocked us so much?

-– There are probably two dimensions, related but independent of each other. On the one hand, the incredible speed at which the government collapsed. Very few, if any, people foresaw it. Most thought that the situation would continue. The process unfolded in a cascade, creating a sense of inevitability that immediately turned into a self-fulfilling prophecy. Indeed, a key characteristic of waterfalls is that they are unpredictable ‘ex ante’. ‘Ex post’, however, everyone has an explanation, everyone saw it coming! On the other hand, the disparity between the Taliban, a medieval-looking rebel group, and the US is shocking. It’s always hard to believe that power asymmetry can be subverted, but sometimes it is.

– He tweeted: “I can’t help but wonder about the implications of the Afghan debacle for counterinsurgency social sciences: dozens of generously funded studies, dozens of cutting-edge articles with Afghan data published in top magazines, for what?” Found an answer?

– As you can imagine, I am critical of many of these studies. They tend to use highly advanced statistical techniques and claim to rigorously identify causal effects. They raise questions such as whether aid shapes attitudes, whether violence produces positive military effects, and so on. But its apparent rigor hides serious problems.

–As which?

-First, since they are based on a scientific apparatus, they project the kind of values ​​that we often find in the hard sciences. However, the social sciences are different. These studies have measurement problems (eg surveys in Afghanistan were notoriously flawed, the security environment introduced all kinds of biases, etc.) and they rely on data that is far from reliable; In other words, these studies are carried out in contexts that require laboratory conditions, but lack them. Second, the results tend to be very conditional: help works sometimes, but sometimes not; some types of help work better than others. Furthermore, the context is always fluid and complex: for example, one study found that, although official aid improved the attitude towards the government, it also attracted Taliban attacks, thus canceling out the pro-government effect. In other words, the results are usually neither executable nor scalable. They are also not very interesting from a theoretical perspective. In fact, many studies produce trivial results. For example, Afghans tended to judge local Taliban less harshly compared to foreign troops because, like most people in all countries, they were nationalists. We knew this without the need for a study!


-– After reading your tweet I looked for the latest academic studies on Afghanistan. The majority (8 out of 10 approx.) Have very optimistic titles about the possibility of building a prosperous and peaceful society. Very few were critical.

-– Much of the research on Afghanistan was explicitly driven by liberal assumptions. I share many of them, but I don’t think they should push the investigation. It is one thing to be a democrat, and another to think that liberalism democracy moves the world. The latter leads to illusion. Many studies criticized what was being done, but in a direction that was not very productive: it is easy and cheap to argue that (much) more money or (many) more soldiers are always needed, but the impact of policies must take into account its viability.

Money distortion

–Why did this happened? Did the academic world also go blind like the others?

-– There are two levels here. One is ideological: liberal democracy so dominates the intellectual sphere of the West that it inevitably generates reactions; we have seen it recently with populism. It also causes major blind spots. The other level is academic professionalism. Rigorous scientific studies lead to publication and professional success. In my field, political science, we have sacrificed expertise in areas and contexts that require long-term immersion, language skills, etc. to the belief that data analysis is the only way to find the truth. Of course, traditional studies were almost impossible to validate, but the analysis of data produced by researchers who normally know very little about one place and study Nigeria one day, Afghanistan the next and Myanmar the next, can also be very problematic. We must not be blinded by sophisticated statistics, even if our work requires a good understanding of them.

-–What role has money played in this phenomenon?

-– This has been a key feature of the studies on Afghanistan: several US agencies (especially the Pentagon and USAID), but also international organizations, released huge amounts of money to attract researchers to the Afghan issue. There is nothing inherently wrong with it: it is an important incentive. But at the same time, it produces distortions. It attracted people who had little interest in learning meaningfully about Afghanistan. In a sense, many researchers “used” Afghanistan to thrive without necessarily producing high-quality research.

-– An American professor responded to his tweet stating that “a political failure does not mean a failure of the Social Sciences.” We live in times when it is difficult to separate science from politics (eg the pandemic). What should scientists do to avoid this confusion?

-– I believe that scientists should do science at the highest possible level, without worrying about whether their research can be immediately applicable. Some of the most important scientific advances have emerged from basic research that at first seemed to have very few practical aspects. My position in political science is that we should strive for the deepest possible level of understanding of an issue, regardless of the political implications. If you come to the Faustian pact to tell politicians that your research can provide them with the solutions they seek, you will have to share the responsibility if that does not happen. You can’t have it all, the money to research policies, but at the same time distance yourself from their results.

The threshold of pacifism

– Do you think that this confusion will have other consequences both in politics and in the evolution of scientific knowledge?

– If the social sciences mix too much with politics, the failures of politics will be reflected negatively in the social sciences, which would be disastrous. This is why universities are so important to research: they can provide an enabling environment for it. However, when universities become dependent on external funding, they push researchers to constantly seek money, often distorting research in the way we have seen in Afghanistan.

-–Western Europe has been in peace for 76 years. It is difficult to find a public opinion on the continent that does not declare itself pacifist. Don’t you think it is distressing to find yourself in a situation in which the solution is the use of force?

– We do not like to fight wars, in part because of the institutionalized past experience of war, occupation and destruction. The US differs from Europe in that sense: it has not experienced widespread destruction since the Civil War or foreign occupation and is more prone to war in terms of public opinion (and generally more accepting of violence; see his attitude towards the death penalty). However, there is a threshold beyond which even pacifists can come to accept the use of force. Of course, what this threshold is is difficult to predict ex ante.

Democratic expansion

-– Where does the notion come from that democratic institutions can be exported to a country without democratic tradition?

-– I think this is the legacy of World War II and the successful experience of Germany and Japan. These examples were widely used to justify the invasion of Iraq in a way that was both naive and manipulative. At the same time, over the past 30 years, we have witnessed an incredible global expansion of democracy. Despite the challenges, this expansion is real and has been a combination of internal factors and external incentives.


– For the debate between institutionalists and culturalists, what happened in Afghanistan should offer us any lesson?

-–I have always been skeptical of dichotomies. Clearly, institutions are not built in a cultural vacuum; at the same time, however, culture is not an obstacle, otherwise nothing would change as culture changes very slowly. I am very concerned that the message emerging from the Afghan debacle is that some countries will never be able to have a democratic regime. Such a point of view would be just as misleading as the opposite, that is, that democracy can be imposed anywhere and at any time. For many years, not long ago, there was a belief that southern European countries such as Spain, Portugal, and Greece were not culturally conducive to democratic institutions. That was a mistake. State building and institutional design are more like an art than a science: although they are obviously more likely to thrive in a conducive social and economic environment, they can also thrive in more difficult spaces if implemented creatively and responsively. to context. Once institutions take root, they can also change cultural practices in ways that few people thought possible beforehand.