Nils Hägerdal, Friend or Foe: Militia Intelligence and Ethnic Violence in the Lebanese Civil War (New York: Columbia University Press, Columbia Studies in Middle East Politics, 2021).

Jadaliyya (J): What made you write this book?

Nils Hägerdal (NH): The basic themes of this book emerged during my time in the Committee on International Relations MA program at the University of Chicago. When I started the program in late summer of 2008, both academics and policymakers were consumed with analyzing the war in Iraq, and especially the troop surge initiated by President Bush in January 2007. Topics such as the causes of ethnic conflict and the dynamics of counterinsurgency were generating intense debates and disagreements. So in an immediate sense I started working on what ultimately became this book as a way of grappling with the wider issue of sectarian conflict in the Arab world. Iraq at that time was still in an ongoing conflict, and not a location where anyone imagined doing fieldwork anytime soon, so I decided to study these themes in another country. While researching other conflicts in the region I settled on Lebanon as a case study because—unlike Iraq—its civil war of 1975-1990 had been concluded for some time; the conflict had generated a rich secondary literature, partly available in English; and it was relatively accessible to researchers for fieldwork, compared to Algeria or Yemen.

In a more general sense, the topic of forced migration in the Middle East had been with me for some time as I grew up in an area of Sweden that absorbed large numbers of refugees in the 1980s and ‘90s. When I finished ninth grade, about a third of the students in my class had been born in the Middle East. In a way, studying these topics was probably always about understanding the deeper forces that have been reshaping the world I grew up in, and that have only grown in intensity ever since.

J: What particular topics, issues, and literatures does the book address?

NH: At its most general level, the book studies the role of ethnicity in civil war violence. When do militant organizations attack people based on their ethnic, sectarian, or religious identity rather than their individual traits, beliefs, or choices? To address this question, I study violence in non-separatist ethnic wars. We have a fairly nuanced understanding of ideological wars that do not revolve around identity cleavages, such as Greece, Spain, or Colombia; and of separatist wars, such as the war in Bosnia or the partition of India and Pakistan. However, I show that many wars fall in a third category: the conflict generally involves a salient ethnic or religious cleavage, but neither side has separatist ambitions. Rather, both sides aim to rule a multiethnic state in its entirety.

As a consequence, I find that combatants in these conflicts generally exercise an element of moderation in their dealings with other ethnic communities. Any leader who wants to rule a plural society can ill afford to appear as a genocidal maniac, but must seek at least a modicum of legitimacy among non-coethnics. Militias that prefer to act with restraint should abstain from acts of indiscriminate violence, such as ethnic cleansing, whenever they have sufficient intelligence to act in more selective ways against specific individuals rather than entire ethnic groups. Yet under the fog of war, when separating friend from foe is a matter of survival, militants may still resort to extensive ethnic violence in situations when they perceive that they have no other alternative.

Interestingly, I find that almost all non-separatist ethnic wars since 1945 have occurred in Africa and the Middle East. I think that scholars have generalized too widely from conflicts in Europe and Asia, which have witnessed the bulk of separatist conflicts, and thereby missed some salient dynamics that largely define regional politics in other parts of the world. In particular, the wars in Bosnia during the 1990s have shaped far too much of general commentary about ethnic conflicts around the world when many wars have radically different parameters. I think it is because that war coincided in time with the end of the Cold War, and a general reorientation of international security scholarship from a focus on Great Power wars to sub-state conflict.

J: How does this book connect to and/or depart from your previous work? 

NH: As is often the case with first books, mine grew out of my doctoral dissertation. So rather than connecting with previous work, this book itself has served as the anchor and foundation for a whole research agenda on the impact of armed conflict on civilian communities in the contemporary Levant. It was the book project that allowed me to spend long stretches of time in Lebanon, as well as to read and think more broadly about the conflicts I was studying. Fieldwork, especially at the doctoral stage, often progresses in a rather chaotic fashion. Through this process you encounter many interesting leads and topics that you want to pursue but that do not fit neatly into the original research endeavor. I have been fortunate to have enough time, during sixteen months of fieldwork and multiple postdoctoral appointments, to explore some of these alternate routes at length, and to publish those as standalone articles.

As a result, research for the book also sparked a series of publications about how armed conflict impacts civilians through forced displacement, return migration, environmental crime, starvation, and cross-border blowback terrorist attacks. These research projects would never have happened if the book project had not enabled me to spend long stretches of time in the field to collect the necessary empirical evidence, whether in the shape of quantitative or qualitative material. As it took longer to publish the book than the standalone articles, the book essentially concludes my first research agenda even though it was the first piece of it I ever worked on.

J: Who do you hope will read this book, and what sort of impact would you like it to have?

NH: It would be immensely gratifying to me if the book found readers outside my own discipline, for instance in history or Middle East studies. However, by both training and vocation I am a political scientist and I expect most readers to come from this community as well. This fact stems partly from the growing methodological divide among the aforementioned fields. The quantitative evidence I compiled in the book constitutes one of its most important contributions, but this currency is barely even legal tender outside of empirical social science circles.

Consequently, my main audience is political scientists studying either ethnic conflict and civil wars, or politics of the contemporary Arab world. I hope that my theoretical argument is general enough to interest the former, while the evidence is specific enough to satisfy the latter. One of the biggest challenges I have faced over the years is trying to speak to two different communities, one where many could barely place Lebanon on a map and a second where many scholars have personally visited the country and have a very sophisticated understanding of its history and politics.

J: What other projects are you working on now? 

NH: At the moment I am working on two rather separate strands of research. First, through my position at the Center for Strategic Studies at Tufts Fletcher School I am involved in the Military Intervention Project (MIP), where we are building a comprehensive dataset of all US military interventions overseas since 1776. During the past academic year my position at Tufts was funded through this project and contingent on the work I conducted supervising research assistants. Going forward, I will be using this data to work on new articles of my own about US foreign policy in the MENA region after 1945. One project studies why the United States has had such radically divergent foreign relations with different oil producers (think Iran, Iraq, and Libya versus Kuwait, Saudi Arabia, and the United Arab Emirates). A second article asks why the frequency of US military intervention in the Middle East accelerated so drastically after the end of the Cold War. Both are still works in progress.

Second, I am working on obtaining data from Sweden about migration from the Middle East over the past several decades. While much of recent commentary fixates on the particularly large wave of migrants arriving in 2015, the Scandinavian countries have had large numbers of refugees arriving since at least the early 1980s. Detailed recordkeeping allows us to study these earlier episodes to get a better understanding of the push- and pull-factors inducing international migration. Much like with the first set of ideas, this new research agenda has had the benefit of not requiring fieldwork: at the moment the pandemic makes it difficult to travel internationally, and while I plan to conduct fieldwork again in future there is at present no indication of when it might be possible to do so. The challenge is to figure out what questions we can make progress on in the meantime.


Excerpt from the book (from the Introduction, pp. 1-5)

In January 1976 Christian forces attacked the Karantina neighborhood in Beirut, a slum area that housed thousands of poor residents who worked as day laborers in the nearby port facilities. While home to a mix of mostly Lebanese Sunni and Shia Muslims and a few Palestinians and stateless Kurds, the area was located in East Beirut which is otherwise the predominantly Christian side of the city. After facing light military resistance, the militias entered the neighborhood and proceeded to systematically raze it to the ground. Most residents survived the attack but were not allowed to stay. Instead, they were placed on trucks and buses and transported to the frontlines. Christian leaders telephoned Palestine Liberation Organization (PLO) chairman Yasser Arafat in advance, and at a pre-arranged time the combatants observed a brief cease-fire so that the civilians from Karantina could cross the frontlines on foot and enter permanent exile in Palestinian-controlled territory. During the attack on Karantina Christian militias thus intentionally and forcibly displaced all Muslim residents, based on their sectarian identity rather than their individual or personal characteristics. This behavior forms the textbook definition of ethnic cleansing.

Nearby Sunni residents in the Baydoun area, named for its eponymous mosque and located no more than a ten minute drive from Karantina, somehow escaped this treatment. The area was controlled by the same Christian militia groups, and also hosted several thousand Sunni residents who lived intermingled with a larger Christian population. This area was just as close to the frontlines as Karantina, and commands a strategic location at the heart of East Beirut. A small number of Muslim residents left Baydoun during the first two years of the war because of political and security issues, possibly including active threats that involved coercion but stopped short of bloodshed. However, an overwhelming majority of Sunni families remained in place during the civil war. The local mosque stayed open for Friday prayers every single week throughout fifteen years of civil war, and local Muslims attended without hiding themselves or their identity. The mosque remains in operation to the present, and was packed from wall to wall with worshippers when I visited during Friday prayers on a warm and sunny spring day in 2014. Unlike in Karantina, it appears that Christian militias in Baydoun forced only select Muslim individuals to leave while permitting the majority to stay. The fact that these Sunni residents remained in place also shows not only that militia forces permitted them to stay, but also that they felt sufficiently safe to choose not to move to Sunni-dominated West Beirut located only a few kilometers away. Why would members of the same sectarian community, residents of two different neighborhoods in close geographic proximity, experience such radically different fates?

This book explores the role of ethnicity in civil war violence. In plural societies, political mobilization typically occurs along ethnic lines, such as through ethnic political parties contesting peaceful elections. Whenever political conflict evolves into armed hostilities, it is therefore quite common to find that its combatants generally hail from separate ethnic communities. As the fog of war thickens during chaotic circumstances, where the ability to separate friend from foe is a matter of life and death, many participants rely on ethnicity as an informational shortcut and start to view non-coethnics in general with suspicion. Ethnicity thereby serves as an informational cue or a heuristic device that signals an individual’s likely political loyalties and potential hostility. Some armed groups engage in ethnic cleansing: comprehensive forced displacement of entire civilian communities based on their ethnic identity. In armed conflicts fought across an ethnic divide, ethnic cleansing may become a rational military strategy because if all hostile militants are non-coethnics then the full set of non-coethnics contains the full set of hostile militants. Even extensive patterns of prewar interethnic trust can rapidly break down and yield ethnic separation, as militants displace non-coethnics and civilians elect to flee from areas where they no longer feel safe.

Sometimes the military-strategic incentives for ethnic cleansing are compounded by political incentives to cause ethnic separation and create homogenous nation states. There is a large body of work on separatist conflict in political science, and on how those conflicts generate political incentives for ethnic cleansing. Many nationalist ideologies contain narratives of how certain territory forms part of an ancestral homeland that should be reserved for members of a particular group only. Bosnian Serb forces employed ethnic cleansing against Muslim Bosniaks in East Bosnia in the mid-1990s because by rendering the area homogenously Serbian, they hoped to fold this territory into the neighboring state of Serbia; their compatriots tried the same dirty trick against Albanians in Kosovo only a few years later. Partition between Greece and Turkey or India and Pakistan generated millions of refugees, who needed to leave their original homes in order to form mono-confessional new polities.

Yet in non-separatist ethnic conflicts contestants generally have political incentives to act with some degree of moderation, and to avoid causing permanent ethnic separation. In such conflicts combatants may regard non-coethnics as enemies, but often still view them as fellow citizens; the armed struggle concerns how to govern multiethnic countries, not how to partition them. Combatants have incentives to exercise restraint because even if they succeed on the battlefield they will need to rule a large population of non-coethnics after the war ends. For instance, national leaders in multiethnic countries like Cote d’Ivoire, Mali, and Senegal have consistently backed away from the precipice of ethnic mass killings to avoid spirals of escalating violence and keep their plural societies intact. While these leaders did not deny the existence of ethnic differences, they genuinely viewed different ethnic communities as full members a diverse society. In fact, research shows that somewhere between one third and one half of all civil wars fought across an ethnic cleavage since 1945 feature non-separatist political goals. Aside from Lebanon, which fought non-separatist civil wars along sectarian lines in 1958 as well as 1975-1990, some other important cases include Jordan (1970-71), Angola (1975-91), Chad (1980-94), and Tajikistan (1992-97). Interestingly, I show that non-separatist ethnic wars have predominantly occurred in the Arab world and Sub-Saharan Africa.

My first contribution in this book is to systematically explore the dynamics of violence in non-separatist ethnic civil wars. Using ethnicity as a proxy for political allegiance is easy. What is more difficult – for those armed groups that do not seek ethnic separation as a political goal – is to correctly identify non-coethnics who are politically loyal, neutral, or apathetic and who do not pose any military threat. How can militants discriminate among neutral and hostile non-coethnics? What kind of information helps them accomplish this task? Under what conditions will they choose to use ethnicity as an informational shortcut, and when might they prefer to make decisions based on other kinds of intelligence? The classic problem of counterinsurgency is how to identify militant fish hiding in a civilian sea. Conversely, in a way the problem I study is how militants can identify friendly non-coethnics in a setting where any non-coethnic could be a hostile militant, and where participants increasingly conflate ethnic identity with political loyalties. My argument considers how armed groups behave when they face divergent political and military incentives with regards to forced displacement of non-coethnics, and when ethnic cleansing as a result becomes a reluctant last resort rather than a primary political goal.

The issue of how militants discriminate among non-coethnics raises the larger question of how they collect and process intelligence during armed conflict. This question forms a central puzzle in the literatures on civil war violence and forced displacement. The most influential approach to micro-level variation in wartime violence centers on how armed groups strive to pacify military resistance in areas they control, and how access to information drives their choice between various forms of selective and indiscriminate violence to accomplish this goal. This literature shows that consensual provision of information by local residents usually constitutes the best source of intelligence, as local residents have better knowledge of their social and geographical environs than outside militants do. Civilians have agency over their own actions, and models of civil war violence usually center on what conditions enable militants to induce local civilians to collaborate or otherwise reveal information. Violence emerges as a joint product of militant organizations and local residents, each participating for their own purposes that often deviate from whatever macro-level political cleavages ostensibly animate the armed conflict they endure.

However, we also know from the literature on industrial organization of violence that many armed groups have deep ties to local residents in areas where they operate, because they mobilize volunteers in the community to create and sustain their military fighting force. Constructing a military organization requires great efforts of both recruitment and logistics. Many armed groups solve these problems by mobilizing preexisting social and political organizations, such as political parties, and rely on volunteer efforts by local residents. Community members may participate in a range of differentiated military and non-military roles, ranging from frontline combat to cooking food. For instance, insurgents in El Salvador could count on as many as up to one third of the civilian population in some areas where they operated to contribute material support for their struggle by making tamale sandwiches for fighters. Large numbers of Ugandan villagers willingly provided rebel groups with simple meals – perhaps peasant staples like cassava, yams, and porridge thickened with cornmeal – to enable the insurgency to fight government forces in their geographic vicinity. Malayan and Vietnamese insurgents operated mostly in areas where locals served them fragrant steamed rice.