previous section
next section



All states use violence, either actual or threatened, to defend their borders and enforce the rule of law. Indeed, it is difficult to imagine any reasonable society existing for long without some kind of centralized coercion. At times, however, the state's coercive apparatus does not focus solely on national defense or public safety, but instead aids the powerful, reproduces inequality, or promotes discrimination. Contemporary Serbia and Israel are two such cases. In each of these states, the military, police, and partially autonomous paramilitaries have used violence to further the interests of one national group over others. In each instance, the state's broader goal has been national exclusion of some kind or another. Actual patterns of nationalist state violence, however, have varied substantially over time and space.

Readers sympathetic to either the Serbian or Jewish national projects may argue there was little choice. To protect basic Serbian and Jewish collective rights, state violence of some sort, while regrettable, was necessary. Others might suggest otherwise, saying Serbia and Israel exaggerated the threats they faced or even helped create their enemies through discrimination. Important as that debate is, my concern here is not with the legitimacy of violence or with each conflict's root cause. Instead, assuming that political violence is already under way, this book seeks to explain why states use some methods and not others in given times and places. Like other nationalist states, Serbia and Israel have occasionally resorted to mass expulsions of unwanted populations, but on

other occasions they have relied on subtler forms of national domination. What explains these variations? Why do states resort to ethnic cleansing in some cases, but use police-style repression in others?

My explanation is straightforward. In times of acute political or military crisis, today's high-capacity states will prefer to police, rather than expel, unwanted groups living in areas of concentrated state power. This is true even in strongly nationalist entities such as Serbia and Israel. At the state's margins, however, military forces and their paramilitary allies enjoy more freedom to maneuver, and it is here that nationalist violence is likely to be most intense. This is so because contemporary nation-states are sensitive to the appearance of legality, in part because of the increased salience of international human rights norms. Areas of concentrated state power are zones of legal density and enhanced state responsibility, but marginal regions are less clearly subject to state authority. On the periphery, official security forces and their unofficial allies operate in a more lawless environment, facilitating their resort to despotism. Note, however, that a region's definition as center or periphery is not set in stone, and that in times of military or political emergency, some areas' status can be rapidly redefined. "Central" or "peripheral" areas are social and political constructs subject to renegotiation.

As the book's title suggests, I use two key spatial metaphors to illustrate my case. In spring 1992, I argue, Bosnia became a "frontier" vis-à-vis Serbia, making it feasible for Serbia to engage, covertly, in awful acts of ethnic cleansing. Frontiers are peripheral regions unincorporated into a powerful state's legal zone of influence, and as such are more prone to acts of lawless nationalist violence. The Palestinian West Bank and Gaza Strip, by contrast, became a "ghetto" vis-à-vis Israel during the 1970s and 1980s. Thus, when the first Palestinian uprising began in December 1987, Israeli forces found themselves engaged in acts of harsh policing, not ethnic cleansing. They did so despite the existence of powerful political forces favoring more radical policies toward Palestinians. Ghettos are repositories of unwanted and marginalized populations, but are nonetheless included within the dominant state's legal sphere of influence, classifying them as quasi-members of the polity. Ghetto populations are more likely to be policed than forcefully deported.

Having advanced my argument in bold terms, I must offer a few caveats. First, the universe of cases to which this argument applies is confined to highly capable states with functioning bureaucracies, internal

coherence, and the ability to enforce laws over much of their territory. These states are also likely to view themselves as part of "civilized" international society. Temporally, my argument refers chiefly to conflicts during the 1970s, 1980s, or 1990s, when norms of international human rights monitoring grew increasingly widespread. My argument, for example, does not apply to Nazi Germany, because it took place during a period of weak international human rights monitoring. Nor, for that matter, does it apply to the Bosnian Serb entity formed in Bosnia during 1992, as it was not a high-capacity, strong state.

If patterns of state violence vary by geographical zone, then borders of both the internal and international sort must also play a key role. Soldiers, secret police, and their paramilitary allies, after all, need some way of differentiating between one zone and another. Thus while borders are socially constructed lines in the sand, they have dramatic realworld significance, marking the transition from zones of high and low state power. State violence takes place in bounded physical spaces termed here "institutional settings." Armed representatives of the state are sensitive to boundaries between these settings, realizing, if only subconsciously, that methods appropriate in one are unthinkable in another. Borders shape state violence in dramatic and abrupt ways, shifting the state's coercive repertoire from ethnic cleansing to policing, and vice versa.

The relevance of all this to today's headlines is substantial. Nationally motivated Serbian violence in the former Yugoslavia ended, at least for now, with NATO's 1999 takeover of Kosovo and the fall of Serbian strongman Slobodan Milosevic. In the Middle East, however, violence rages on. Fighting between Israeli troops and Palestinian militias is escalating, and in the immediate future, Palestinians in the West Bank and Gaza Strip—as opposed to Palestinians living elsewhere—may enter the most physically dangerous chapter of their 120-year battle with Jewish nationalism. In the year 2000, Israel began using warlike methods in the Palestinian territories for the first time in decades, deploying shoot-to-kill ambushes, armored vehicles, and warplanes. In late 2001, Israeli tanks responded to Palestinian suicide bombers by seizing Palestinian autonomous zones and inching toward full-scale war, a trend that accelerated in spring 2002. This shift in repertoires has been a shock to those immersed in the Palestine-Israel conflict, because until recently, Israel relied almost exclusively on harsh police-style methods of control in the West Bank and Gaza. Israeli tactics are now changing dramatically,


Map 1. The Federal Republic of Yugoslavia (Serbia and Montenegro) and its Bosnian "frontier"

[Full Size]
due largely to the region's slide from center to the margins of Israeli state power. As the West Bank and Gaza become increasingly peripheral to the Israeli state, they experience a more despotic regime of domination.

Palestine's reconfiguration to the periphery of Israeli state power


Map 2. Israel and its Palestinian "ghetto" (the occupied West Bank and Gaza Strip)

[Full Size]
began in the mid-1990s. From 1967 until then, Palestinian territories were a de facto ghetto within Israel, securely trapped within the boundaries of a powerful Jewish state. The 1993 Oslo accords, however, created small, partially autonomous Palestinian enclaves scattered throughout the West Bank and Gaza. Although these pockets did not have unfettered access to the outside world, their interior was relatively free from an Israeli military presence. The more these lands escaped direct Israeli control, however, the more precarious their future grew. To restate this book's central thesis, nationalist states tend to be most radical at their margins, not their core. As Palestinians approached a semblance of
territorial autonomy, in other words, they confronted dangers similar to those faced by Kosovo's ethnic Albanian population in 1998, when parts of the contested province slipped from Serbian control.


My interest in geographic settings, nationalism, and repertoires of state violence was sparked by an intense, personal experience. From January 1985 to January 1988, I served as a conscript in the Israeli army, spending much of my time in an infantry unit. At the time, draftees rotated between occupation duty in southern Lebanon and the Palestinian territories. Those were relatively peaceful years, and I experienced little real combat. Still, I did witness remarkably different patterns of state coercion in each zone. Through a mixture of tacit and explicit signals, my colleagues and I learned that violent tactics appropriate to Lebanon were wrong for the West Bank or Gaza, and that methods permitted in Palestinian lands were unthinkable in Israel proper. To take one example, Israeli shoot-on-sight ambushes in the 1980s were common in Lebanon, but the same forces used quite different methods in the occupied Palestinian lands. As soldiers moved from one zone to another, in other words, they changed tactics quite dramatically. These variations were not organized solely along geographic lines, however, as we treated Jews and Arabs differently regardless of locale. As enforcers of Israeli state policy, our actions varied both by region and by nationality. These distinctions were integral to our daily routine, attracting little attention on all sides. When I began thinking more carefully about my military experiences, however, these patterns became intriguing sociological puzzles. Why would the Israeli army treat Lebanon and Palestine so differently?

My interests were strengthened in 1995, when I was sent by Human Rights Watch, a New York–based group, to study Turkey's war against Kurdish insurgents.[1] Turkish forces had burned down dozens of Kurdish villages in the country's southeastern region, using rape, torture, and other intensely violent measures. When Kurdish civilians fled the area defined as the "emergency zone," however, Turkish security forces used quite different methods of control, even though Kurdish squatter neighborhoods often harbored insurgents. What was appropriate for Turkey's southeast was entirely inappropriate in the country's west. As had been true for Israel, Turkey's coercive style varied from one geographical arena to the next. In areas where Turkish state power was less overwhelming, moreover, its methods were more blatantly destructive. While

the Israeli case highlighted variations across international borders, Turkish patterns underlined the role of internal boundaries. I do not include the Turkish case in this book, as this is a focused comparison of Serbia and Israel. My Turkish experiences did convince me, however, of the importance of internal as well as international boundaries.

Working in the world of international human rights documentation, I came to believe that scholars, journalists, and human rights analysis were missing an important aspect of state violence. Contrary to conventional wisdom, security forces and their semi-private allies often distinguished between institutional settings, with vital consequences for states, insurgents, civilian victims, and human rights advocates. These nuances, however, were often overlooked by scholars and state critics, many of whom lumped all coercive policies together under the rubric of "human rights abuses," ignoring geographically specific distinctions. I resolved to develop a way of distinguishing theoretically between different zones of state violence, and of explaining these zones' emergence over time. Most importantly, I wanted to explore an intriguing paradox: Why, when unwanted populations were fully dominated by contemporary ethnonationalist states, were they more likely to be policed than expelled? Why wouldn't states treat areas they controlled fully with even greater violence?

To those familiar with the awful abuses suffered by victims of these and other wars, academic theorizing may seem callous, opportunistic, even obscene. The neutral language of social science can never do justice to articulating the enormity of wartime suffering; but that effort is perhaps best left to journalists, novelists, and poets. As social scientists, our job is more modest. We provide explanatory tools to illustrate the social forces causing and shaping patterns of human misery. Whether or not this provides any tangible benefit to the world is difficult to say.


This book's central case studies are Serbia's 1992–93 campaign against non-Serbs in Bosnia, and Israel's 1988 efforts to put down the first Palestinian uprising, popularly known as the first Intifada.[2] In these periods, both Serbia and Israel were overtly nationalistic states that felt compelled to use violence to defend the security of their national communities, narrowly defined. In each case, however, agents of the state used quite different methods. Serbia responded to Bosnia's 1992 demands for independence with acute violence aimed chiefly at forcing non-Serbs to flee. In

1988, by contrast, Israel responded to similar Palestinian claims with ethnic policing, a pernicious but less destructive policy. Although Serbia and Israel were similar in important respects, their repertoires of violence in these two instances varied dramatically. In explaining this divergence, I draw on the work of organizational and political sociologists, as well as theorists of international norms.

I test and elaborate my argument by studying variations within each case. Serbia did not use ethnic cleansing against minorities in territories that it fully controlled, while Israeli violence in Lebanon, an area beyond Israel's firm and legal control, was far more destructive than in Palestine. The book thus proceeds along two comparative tracks, studying divergent repertoires between Serbia and Israel on the one hand, and among different regions within each case, on the other.


My sources include field interviews, newspaper reports, and scholarly publications.[3] For the Israel-Palestine study I make use of interviews conducted during 1992–94, first for Human Rights Watch, and then for my own academic purposes. In 1992, I wrote a report on the actions of Israeli undercover units seeking to arrest or kill Palestinian activists, interviewing dozens of Palestinians as well as Israeli soldiers, bureaucrats, and journalists.[4] The next year, I analyzed Israeli methods of interrogation in the West Bank and Gaza, interviewing over sixty former detainees and dozens of Israeli and Palestinian lawyers, activists, and officials.[5] I used a translator for Arabic-language interviews, but spoke in Hebrew with Jewish Israelis. To this I have added quotes and insights from fortyfive semi-structured discussions with Israeli military veterans interviewed during 1992–94.

For the Serbian case, my sources are similarly diverse. In early 1996, I traveled to Croatia and Bosnia for the International Committee of the Red Cross, a Geneva-based humanitarian group, studying their efforts to protect civilians and prisoners of war. This project included 100 interviews with expatriate and local employees of international agencies, as well as Croatian and Bosnian officials. Virtually all my informants were involved in efforts to monitor, control, and analyze patterns of state violence. In 1997, I went to Belgrade on my own behalf, asking questions about Serbia's involvement in Bosnia and about violence in ethnically mixed areas of Serbia and Montenegro. Relying again on translators, I interviewed over 100 human rights advocates, academics, former fighters,

journalists, and local politicians. In spring 1999, I traveled to Albania's border with Kosovo for Human Rights Watch, interviewing ethnic Albanian refugees fleeing Serbian ethnic cleansing. For reasons of confidentiality, most informants in both the Israeli and Serbian cases are not identified by their real names. In some cases, I cite a specific interview and identify the informant by a first-name pseudonym. In a handful of instances, I supply the informant's actual full name.

This book also draws on discussions with investigators studying armed conflicts worldwide. Hundreds of researchers travel the globe each year for Human Rights Watch, Amnesty International, the International Committee of the Red Cross, and others, researching state violence and amassing a wealth of practical, empirical knowledge. Some of this appears in written reports, but much remains tucked away in individual memories, available only through discussions and interviews. Finally, my analysis also relies on comparative insights from my work for Human Rights Watch in Turkey, Nigeria, and Chechnya.


For some, the Serbia-Israel comparison may stretch credulity, given their apparently radical differences. Upon closer examination, however, there are some intriguing similarities. First, while the population of both states is multinational, the state apparatus has been captured by one national or ethnic group. As a result, each country's bureaucracy engages in overt and tacit discrimination, prioritizing the interests of one national community over others. Serbs and Jews enjoy more state protection, official respect, and privileges than non-Serbs or non-Jews.

Second, both Serbian and Jewish nationalists claim territories lying beyond their internationally recognized boundaries. In today's world, globally recognized borders are hard to change, but influential Serbian and Israeli nationalists feel strongly that adjacent lands belong only to them.[6] Political scientist Ian Lustick calls Israel an "unsettled state" because of its ambiguous relationship to territory and borders, and this term applies to Serbia in the 1990s as well.[7] Third, the two countries' political discourses share some important themes, with Serbia's concern for its historical roots in Kosovo resembling the attachment many Jews feel toward Judea and Samaria, the biblical term for the West Bank. In both Serbia and Israel, moreover, prominent nationalists have discussed the option of expelling unwanted populations to ensure demographic and military superiority. Without forced population transfer, they say, Serbia

and Israel will always face acute demographic and security crises. Although normatively repugnant, this policy recommendation flows logically, but not inevitably, from Serbia's and Israel's founding principles. Once a state energetically prioritizes one community's rights over others, the notion of ethnic cleansing is bound to arise in one form or another.[8]

Conventional wisdom suggests that Serbia and Israel are not comparable because their methods of repression in Bosnia and the West Bank/Gaza were so radically different. When we examine variations within the Serbian and Israeli cases, however, these sharp distinctions begin to fade. Serbian violence in ethnically mixed areas within Serbia was more restrained than in Bosnia, while Israeli actions in Lebanon were more destructive than in the West Bank/Gaza. Both states, in other words, employed diverse tactics in different arenas, with some overlap. Some Serbian actions within Serbia resemble Israeli ethnic policing efforts in Palestine, while some Israeli methods in Lebanon resemble Serbian actions in Bosnia. Israel was not guilty of genocide in Lebanon, and its soldiers did not engage in mass rape and other war crimes of the sort committed by some Serbian fighters in Bosnia. At the same time, Israel did resort to expulsions and dangerously indiscriminate shelling in Lebanon, and its secret services did work closely with Lebanese paramilitaries guilty of Bosnia-like atrocities. Lebanon and Bosnia are similar in some respects, but they are not parallel cases.

One clear difference between Israel and Serbia is the way in which Western powers and international organizations have responded to each country's territorial and military ambitions. Serbian interventions in Bosnia and elsewhere were harshly condemned by Western powers, who convinced the UN Security Council and NATO to deploy sanctions and, eventually, military force to punish Serbian transgressions. Israel's occupation of the West Bank, Gaza, and Lebanon, on the other hand, has attracted more muted forms of international criticism. Largely due to America's special relationship with Israel, Western powers and the UN Security Council have regarded Israel's actions with greater understanding than Serbia's, with important and unanticipated consequences. In Bosnia, Western sanctions drove Serbian intervention underground, promoting the use of private paramilitaries and underworld thugs, and facilitating Serbia's resort to ethnic cleansing. Israel's control over the West Bank and Gaza, conversely, was done quite openly, relying on Israel's regular security forces, and this resulted in a subtler regime of domination. Different international attitudes, in other words, dramatically shaped each state's coercive style. Greater Western pressure on Serbia provoked more openly destructive Serbian

methods, while greater Western permissiveness led to less acute methods of Israeli control. Contemporary human rights norms being what they are, the West's tolerance for Israel's West Bank/Gaza occupation came attached with some important strings, pushing Israel toward a policing strategy in the occupied lands during the 1980s and early 1990s.

The Serbia-Israel comparison is bound to be controversial, and many readers will remain skeptical. I can only ask that you bear with me, reading as much as you can of the material presented below. If my interpretations push you to think differently about Serbia, Israel, and state violence, I will be content.



I interviewed hundreds of people for this project, many of whom were courageous individuals living through difficult times. I deeply appreciate their willingness to share their views. This book would never have been written, however, without the guidance of Michael Burawoy, the most supportive and wise mentor a student could hope for. My subject demanded a focus on states rather than social class, but my work is deeply influenced by Michael's normative commitments and extended case method. I am similarly indebted to academic advisors Peter Evans and Ken Jowitt of the University of California at Berkeley. Two other mentors, John Meyer and Susan Woodward, went far beyond the call of duty, spending many hours discussing my findings and reviewing my text. Two close personal friends, Peter Andreas and Chuck Call, were supportive and challenging colleagues, and my debt to them is great. Human Rights Watch and the International Committee of the Red Cross hired me as a research consultant at various points over the last decade, and then generously allowed me to draw on those experiences for this book. I am particularly indebted to Carroll Bogert, Eric Goldstein, Ken Roth, Aziz Abu Hamad, Fred Abrahams, Rachel Denber, Jacques Stroun, and Hernan Reyes.

Research for this book was funded by the John D. and Catherine T. MacArthur Foundation and the Social Science Research Council's Program in International Peace and Security. Writing was supported by fellowships from the Brookings Institution and Brown University's Watson

Institute of International Studies. Columbia University's Middle East Institute provided a welcoming environment for months at a time, and Johns Hopkins and McGill universities were my home base during the final stretch.

Many organizations and individuals have helped in the field. In Palestine, I worked extensively with the human rights group al-Haq during 1992–93, as well as with journalist Walid Battrawi, a superb guide and translator. In Israel, the human rights group B'Tselem provided office space and archival access. In Serbia, the Forum for Ethnic Relations, led by Dušan Janjić and his capable assistant, Vesna Vidujević, was my generous host. Journalist Dejan Pavlović was both excellent translator and cheerful traveling companion, while Dejan Anastasijević, Filip Švarm, Nataša Kandić and Sonja Biserko were important sources of wisdom and information. Political scientists Dijana Vukamanović and Zoran Slavujević were tremendous guides to Serbian domestic politics.

Many friends and colleagues have been generous with their time, comments, and support along the way. Jelena Pejić and her family were of special help in Belgrade, while Nitza Berkovitch and Meir Shabat were of great assistance in Israel and abroad. I am similarly indebted to Craig Calhoun, Pierre Englebert, Martha Finnemore, Doug Guthrie, Lisa Hajjar, Robert Hayden, Aliza Marcus, Meghan O'sullivan, Nina Tannenwald, Kellee Tsai, Mona Younis, and Richard Wood. Comments by Roger Haydon were extremely useful during revisions, as was detailed feedback from Michael Barnett, Gay Seidman, Elisabeth Wood, and Erik Olin Wright. An intense session with Chicago University's Comparative Politics workshop was particularly useful in fall 2001, and I am grateful to workshop participants Stathis Kalyvas, Susan Stokes, Mathew Kocher, and others. Dana Bell and Jasmina Burdžović Andreas were superb editors and commentators, while Alma Vardari-Kesler and Howard Ramos helped verify some of my sources. All errors, of course, are entirely my own responsibility. I am also grateful to Naomi Schneider, Executive Editor of University of California Press, for her early and enthusiastic support for this project, as well as to production team members Sue Carter, Annie Decker, Marilyn Schwartz, Mary Severance, and Victoria Kuskowski.

My parents, Dr. Martha and Newton Frohlich, have been true pillars of support throughout the research and writing process. Most importantly, however, this book would never have been written without the love and inspiration of Emma Naughton, to whom I dedicate this book.

previous section
next section