Winner of the 2016 Lewis Fry Richardson Award, Paul Collier: Clarity and Compassion in the Study of Civil War

1 Introduction

The Lewis Fry Richardson Lifetime Achievement Award, which was awarded for the sixth time at the annual meeting of the Network of European Peace Scientists on 21 June in Milan 2016, is presented on a tri-annual basis to a scholar “who has made exemplary scholarly contributions to the scientific study of militarized conflict” and who has spent most of his/her academic life in Europe. This year the award was given to Paul Collier at the University of Oxford.

Paul Collier is a leading scholar in the field of development economics. Here, he is recognized for his very important contributions to the study of civil war. His best-known article on this topic and most frequently cited article is “Greed and grievance in civil war,” written with Anke Hoeffler and published in 2004 after having been circulated in draft form since the late 1990s. As of mid-July 2016, this article has been cited over 1000 times in the Web of Science and had over 5000 citations in Google Scholar. His second-most cited article, also on conflict and co-authored with Anke Hoeffler, “On economic causes of civil war” (1998) had been cited 528 times.

The jury highlighted the key role of Collier’s work in reviving academic research on civil war in the late 1990s, the broader impact of the World Bank group led by Collier in spurring advances in the field as well as stimulating important data collection efforts, as well as his central role in popularizing the insights of academic research to a broader audience.

In this note we briefly outline some of the main highlights of Collier’s contributions to the study of armed conflict and their legacies. Of course, Collier has worked on many areas outside the study of violent conflict, including key contributions to development in Africa (Collier and Gunning 1999), natural resources (Collier 2010), migration and social trust (Collier 2013). Indeed, much like Richardson himself, who was a meteorologist (see Hunt 1998), Collier came to conflict research as an outsider, and his background in development research has helped bring a fresh and novel perspective on conflict research.

2 The past is a different country: back to the 1990s

Civil war is now so central in conflict research that it is hard to imagine a time when this was not the case (see Blattman and Miguel 2010). Nevertheless, despite the fact that both Richardson (1960a) himself and later the Correlates of War project (Small and Singer 1982) had collected extensive data on civil war, there was only limited comparative academic research on civil war during the Cold War. Up until the early 1990s, the primary focus in conflict research was very much interstate conflict, and often the relations between superpowers. Topics such as arms races, arms control, and interstate crises figured prominently in journal articles and professional conference papers. One immediate impact of the end of the Cold War was a renewed interest in the role of domestic politics in international behavior, in particular explanations for the so-called democratic peace or the findings that democracies do not appear to go to war on one another, and the possible role of domestic factors in the fall of communist rule and the end of the Cold War (see, e.g. Gaddis 1992/93).

In retrospect it is easy to see why the end of the Cold War stimulated increased attention to civil war. Although interstate conflict had not been frequent since 1945, the perceived risk of conflict between the superpowers was seen as being much lower after the end of the Cold War (see Mueller 1990). Civil war had been by far the dominant form of conflict in the international system in the post WWII-period, and although interstate wars tend to be more severe and cause larger death tolls, civil wars are more frequent and tend to be more persistent. The dissolution of the Soviet Union and Yugoslavia coincided with an increase in civil war outbreaks, leading to a widespread anticipation of a coming anarchy after the end of the Cold War (for an example of this pessimism see Kaplan 1994, and see Mueller 1994 for an entertaining review of the doomsayers with a much more optimistic assessment). While only 0.2% of the articles in the Journal of Conflict Resolution and the Journal of Peace Research had the term “civil war” in the title over the period 1986–1995, the share grew to 5% over the following decade, and is as high as 9.9% for the period 2006–2015.

Increased attention to civil war after the Cold War, however, did not immediately translate into a coherent theory or framework for systematic inquiry. Much of traditional security studies became near obsessed with a rather simplistic analogy of anarchy between ethnic groups in conflict and the security dilemma (see Posen 1993), inspired in part by the dissolution of the former Yugoslavia and conflict between successor states based in part along ethnic lines. Yugoslavia is in many ways an atypical conflict where the federal government to some extent disappeared, while most ethnic conflicts bear a stronger resemblance to hierarchy, where a dominant group controlling the state faces challenges from typically much weaker rebel groups representing excluded ethnic groups. The focus on the risk of dyadic conflict in the quantitative international relations literature, focusing on power ratios and features defining interaction opportunities (see e.g. Bremer 1992) also did not lend itself easily to testable propositions on civil war, given the absence of any clear data on non-state actor and potential rebels.

It is no exaggeration to say that Collier with collaborators almost single-handedly revived interest in scientific study of civil war. As far as we are aware, the first study by Collier on conflict is a 1994 article on the economic aspects of demobilization and transitions from conflict to peace, which reflected a growing interest in the economic consequences of civil war. Later, Collier became interested in the causes of civil war outbreaks, and the possible role of economic factors. Popular discourse on observed civil wars – with Syria as a recent example – tends to stress grievances, or the political factors that are believed to have motivated mobilization against the government. Collier and his collaborators noted that this reason rested on implausible theoretic underpinnings as well as potentially problematic empirical analyzes.

In an early and often overlooked article, Tullock (1971) pointed out that revolution involved a collective action problem. Since political change is a collective outcome, potential sympathizers enjoy the benefit of revolution regardless of whether they contribute or not. Since each individual’s contribution to the final success will tend to be negligible and participation in conflict involves obvious risk, it is difficult to see how political benefits or objectives alone could motivate conflict. Tullock posited instead that conflicts tended to be generated by private benefits, noting how many things called revolutions really amounted to coups carried out by regime insiders, with large potential benefits and better resources to successfully capture the state, rather than revolutions from below. Collier revived attention to the role of private benefits in civil war, and posited greed and grievances as competing explanations for conflict, while highlighting why greed was plausibly more important than the latter. The fact that people tend to provide a political or civic rationale for rebellion is not a good source of insights of their actual motives. More pointedly, Collier (2007, 18) noted that

Rebel movements themselves justify their actions in terms of a catalogue of grievances: repression, exploitation, exclusion. Politically motivated academics have piled in with their own hobbyhorses, which usually cast rebels as heroes. I have come to distrust this discourse of grievances as self-serving.

Collier’s point is also a methodological one, namely the perils of selecting on the dependent variable (see King, Keohane, and Verba 1994). It is not sufficient to point out that actors in civil war seem to have grievances without assessing how common plausible grievances are in non-conflict cases. If grievances are near ubiquitous, then explanations of civil war must provide some account of how rebels manage to overcome collective action problems and achieve mobilization.

3 The World Bank group

While Collier served as director of the Development Research Group of the World Bank from 1998 until 2003, he promoted research on civil war that had an impact far beyond the World Bank and his own research group. Within the Bank, the focus on armed conflict was apparently at first met with some resistance, as it was viewed to be too far afield from development and veering into issues of national sovereignty. Yet by the early 2000s, armed civil conflict and the notion of state fragility had become central aspects of WB development policy.

During this period, the WB group organized a series of meetings, first in Washington, DC and later in other locations such as Princeton, Oslo, and elsewhere, assembling prominent scholars working on civil war. A number of highly-cited academic journal articles and special issues published in the early 2000s were first presented at these meetings, including for example Hegre et al. (2001) and Reynal-Querol (2002).

Most of the quantitative research on civil war in the late 1990s and early 2000s had relied on the Correlates of War civil war dataset, which uses a relatively high threshold of 1000 battle deaths to identify conflicts. The Uppsala Conflict Data Program had started to collect annual date with a lower threshold of 25+ battle deaths, published annually in Journal of Peace Research (starting with Wallensteen and Axell 1993), but the time-series was a short one. The World Bank civil war project put up the bulk of the funding for backdating the time series to 1946, a project undertaken by UCDP and the Peace Research Institute Oslo (PRIO) in collaboration. Thus, Paul Collier played a key role in launching what has become the dataset of choice for most civil war studies (Gleditsch et al. 2002).

4 Research and policy

Collier is also a trail-blazer in crossing the boundary between academia and politics. His engagement with policy-makers and the media brought the development/security nexus to the center of policy debates on international aid, natural resource management and most recently migration. The hallmarks of his policy work are the determination to make academic research accessible to a broad audience and its focus on the analytical evidence for and against specific policy options. All of it is borne out of a deep personal concern for people who have the misfortune to live in societies that are “falling behind and falling apart” (Collier 2007).

The Collier et al. (2003) World Bank report Breaking the Conflict Trap: Civil War and Development Policy introduced the concept of the “conflict trap” and the idea of conflict being “development in reverse” to a wide audience. It documented the devastating consequences of civil wars on civilian health and livelihoods and the life chances of their children – not only in conflict-affected countries but in neighboring countries too. The report pointed out that policy-makers and institutions (such as the World Bank) tasked with developing policy initiatives to tackle poverty must focus on creating sufficient security for investment and trade, or else development will not take place. The report also promotes development as a tool for conflict prevention. These ideas have very much become part of the policy discourse, and the importance of development for conflict and security was one of the reasons cited by the UK government for ringfencing the development budget in times of austerity and budget cuts. ^[1]

In his best-known book The Bottom Billion, Paul Collier took up the complex and thorny issues which keep many of the world’s poorest countries in a development trap. He argues that their problems cannot be “fixed” or even ameliorated by throwing aid money at the problem – in fact aid can have quite the opposite consequences. The book is a call for joint action across governments and – importantly – across governmental departments. Collier does not attempt to simplify complex problems, but rather advocates that governments use a whole range of policy tools to stimulate development. Reading the endorsements of Paul Collier’s books there is a striking diversity of supporters: his unique mixture of passion and analysis strikes a chord with world-leading researchers in both political science and economics, leaders of civil society organizations, philanthropists and politicians from across the political spectrum.

Collier has also had an indirect effect on research and policy through his teaching and supervision activity over the years. Former PhD students and postgraduate collaborators such as Christopher Adam, Stephan Dercon, David Fielding, Anke Hoeffler, Ritva Reinikka, Dominic Rohner and Pedro Vincente are now well-established scholars in their own right and carry the flame for combining cutting edge academic research with policy work into the next academic generation. The Centre for the Study of African Economies (CSAE) has a large number of African and Asian alumni who now have influential position in their home countries’ ministries, NGOs and universities. Many of them retain strong links with the CSAE as research associates. The annual CSAE conference showcases the best research on development to an international audience of practitioners, academics and policy-makers. It is an important forum for dialogue and an important opportunity for researchers to test the political feasibility of their policy advice with a sympathetic and informed audience.

5 Controversy and research progress

Although the influence of Collier is difficult to understate, his work has also generated controversy. Some have challenged the focus on rent-seeking models of conflict and proposed alternative models based on more partisan/preference based conflict (e.g. Acemoglu and Robinson 2006; Esteban and Ray 1999). The dismissal of grievances and inequality has also been challenged, including by one of the authors of this article (see Cederman, Gleditsch, and Buhaug 2013). This is not the place to go into the details of these debates, but it is instructive to see how the influence of Collier has shaped a constructive research program, where his bold and often provocative statements have helped generate responses and reactions that have pushed research on conflict forward. Anti-thesis and synthesis require an initial thesis, and Collier has been a clear prime mover in research on civil war. Again, we see a clear similarity to Richardson himself. Richardson’s (1960b) formal dynamic model of arms races may not be very useful as a description of the data or as an explanation of conflict – indeed, no decision to use force per se appears in the model. Still, it is clear that it has helped move the field ahead and stimulate new research and interest in formal models of conflict. In this sense, the pioneering efforts of Collier are clearly in the spirit of Richardson and highly deserving of this lifetime achievement award.