Contending with Terrorism

Roots, Strategies, and Responses

An International Security Reader

edited by Michael E. Brown Owen R. Cot Jr. Sean M. Lynn-Jones and Steven E. Miller

Contending with Terrorism: Roots, Strategies, and Responses

International Security Readers

Strategy and Nuclear Deterrence (1984) Military Strategy and the Origins of the First World War (1985) Conventional Forces and American Defense Policy (1986) The Star Wars Controversy (1986) Naval Strategy and National Security (1988) Military Strategy and the Origins of the First World War, revised and expanded edition (1991) published by Princeton University Press Soviet Military Policy (1989) Conventional Forces and American Defense Policy, revised edition (1989) Nuclear Diplomacy and Crisis Management (1990) The Cold War and After: Prospects for Peace (1991) Americas Strategy in a Changing World (1992) The Cold War and After: Prospects for Peace, expanded edition (1993) Global Dangers: Changing Dimensions of International Security (1995) The Perils of Anarchy: Contemporary Realism and International Security (1995) Debating the Democratic Peace (1996) East Asian Security (1996) Nationalism and Ethnic Conict (1997) Americas Strategic Choices (1997) Theories of War and Peace (1998) Americas Strategic Choices, revised edition (2000) Rational Choice and Security Studies: Stephen Walt and His Critics (2000) The Rise of China (2000) Nationalism and Ethnic Conict, revised edition (2001) Offense, Defense, and War (2004) New Global Dangers: Changing Dimensions of International Security (2004) Primacy and Its Discontents: American Power and International Stability (2009) Going Nuclear: Nuclear Proliferation and International Security in the 21st Century (2010) Contending with Terrorism: Roots, Strategies, and Responses (2010) published by The MIT Press

Contending with Terrorism

Roots, Strategies, and Responses
edited by Michael E. Brown Owen R. Cot Jr. Sean M. Lynn-Jones and Steven E. Miller

an International Security reader

the mit press cambridge, massachusetts london, england

The contents of this book were rst published in International Security (ISSN 0162-2889), a publication of the MIT Press under the sponsorship of the Belfer Center for Science and International Affairs at Harvard University. Copyright in each of the following articles is owned jointly by the President and Fellows of Harvard College and of the Massachusetts Institute of Technology. Michael Mousseau, Market Civilization and Its Clash with Terror, 27:3 (Winter 2002/03); Audrey Kurth Cronin, Behind the Curve: Globalization and International Terrorism, 27:3 (Winter 2002/03); Assaf Moghadam, Motives for Martyrdom: Al-Qaida, Salafi Jihad, and the Spread of Suicide Attacks, 33:3 (Winter 2008/09); Andrew H. Kydd and Barbara F. Walter, The Strategies of Terrorism, 31:1 (Summer 2006); Max Abrahms, Why Terrorism Does Not Work, 31:2 (Fall 2006); William Rose, Rysia Murphy, and Max Abrahms, Correspondence: Does Terrorism Ever Work? The 2004 Madrid Train Bombings, 32:1 (Summer 2007); Max Abrahms, What Terrorists Really Want: Terrorist Motives and Counterterrorism Strategy, 32:4 (Spring 2008); Erica Chenoweth, Nicholas Miller, Elizabeth McClellan, Hillel Frisch, Paul Staniland, and Max Abrahms, Correspondence: What Makes Terrorists Tick, 33:4 (Spring 2009); Robert F. Trager and Dessislava P. Zagorcheva, Deterring Terrorism: It Can Be Done, 30:3 (Winter 2005/06); Daniel L. Byman, Friends Like These: Counterinsurgency and the War on Terrorism, 31:2 (Fall 2006); Rene de Nevers, NATOs International Security Role in the Terrorist Era, 31:4 (Spring 2007); Mette Eilstrup-Sangiovanni and Calvert Jones, Assessing the Dangers of Illicit Networks: Why al-Qaida May Be Less Threatening Than Many Think, 33:2 (Fall 2008); Audrey Kurth Cronin, How al-Qaida Ends: The Decline and Demise of Terrorist Groups, 31:1 (Summer 2006). Selection and preface, copyright 2010 by the President and Fellows of Harvard College and of the Massachusetts Institute of Technology. All rights reserved. No part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or by any information storage and retrieval system, without permission in writing from the MIT Press. For information, please visit http:/ /mitpress.mit.edu, or please contact the MIT Press, Journals Department, 238 Main Street, Suite 500, Cambridge, MA 02142.

Library of Congress Cataloging-in-Publication Data Contending with terrorism : roots, strategies, and responses / edited by Michael E. Brown ... [et al.]. p. cm. (International security readers) Includes bibliographical references. ISBN 978-0-262-51464-4 (pbk. : alk. paper) 1. Terrorism. 2. TerrorismPrevention. I. Brown, Michael E. HV6431.C6546 2010 363.325dc22 2009049186 10 9 8 7 6 5 4 3 2 1

Contents
vii The Contributors ix Acknowledgments xi Preface

Sean M. Lynn-Jones

PART I: ROOTS OF CONTEMPORARY TERRORISM

3 Market Civilization and Its Clash with Terror 28 Behind the Curve: Globalization and International Terrorism 57 Motives for Martyrdom: Al-Qaida, Salafi Jihad, and the Spread of Suicide Attacks Michael Mousseau Audrey Kurth Cronin Assaf Moghadam

PART II: TERRORIST STRATEGIES: WHY ARE THEY CHOSEN? DO THEY WORK?
93 The Strategies of Terrorism 125 Why Terrorism Does Not Work 162 Correspondence: Does Terrorism Ever Work? The 2004 Madrid Train Bombings 171 What Terrorists Really Want: Terrorist Motives and Counterterrorism Strategy 199 Correspondence: What Makes Terrorists Tick Andrew H. Kydd and Barbara F. Walter Max Abrahms William Rose and Rysia Murphy Max Abrahms Max Abrahms Erica Chenoweth, Nicholas Miller, and Elizabeth McClellan; Hillel Frisch; and Paul Staniland Max Abrahms

PART III: COUNTERING TERRORISM

229 Deterring Terrorism: It Can Be Done 266 Friends Like These: Counterinsurgency and the War on Terrorism Robert F. Trager and Dessislava P. Zagorcheva Daniel L. Byman

Contents vi

303 NATOs International Security Role in the Terrorist Era 336 Assessing the Dangers of Illicit Networks: Why alQaida May Be Less Threatening Than Many Think

Rene de Nevers Mette EilstrupSangiovanni and Calvert Jones

PART IV: THE FUTURE OF TERRORISM

377 How al-Qaida Ends: The Decline and Demise of Terrorist Groups Audrey Kurth Cronin

The Contributors

michael e. brown is Dean of the Elliott School of International Affairs and Professor of International Affairs and Political Science at the George Washington University. owen r. cot jr. is Co-Editor of International Security and Associate Director of the Security Studies Program at the Massachusetts Institute of Technology. sean m. lynn-jones is Co-Editor of International Security and a research associate at the Belfer Center for Science and International Affairs, John F. Kennedy School of Government, Harvard University. steven e. miller is Editor-in-Chief of International Security, CoPrincipal Investigator with the Project on Managing the Atom, and Director of the International Security Program at the Belfer Center for Science and International Affairs, John F. Kennedy School of Government, Harvard University. max abrahms is a postdoctoral fellow at the Center for International Security and Cooperation at Stanford University. daniel l. byman is Director of Georgetown Universitys Center for Peace and Security Studies and is a nonresident senior fellow at the Saban Center for Middle East Policy at the Brookings Institution. erica chenoweth is Assistant Professor of Government at Wesleyan University and Director of Wesleyans Program on Terrorism and Insurgency Research. audrey kurth cronin is Professor of Strategy at the U.S. National War College, National Defense University. rene de nevers is Assistant Professor in Public Administration at the Maxwell School of Citizenship and Public Affairs at Syracuse University. mette eilstrup-sangiovanni is University Lecturer in International Studies at the Centre of International Studies and is a fellow in International Relations at Sidney Sussex College. hillel frisch is a senior research fellow at the Begin-Sadat Center for Strategic Studies and Senior Lecturer at the Department of Political Studies, Bar-Ilan University. calvert jones is a doctoral student in the Department of International Relations at Yale University.

The Contributors viii

andrew h. kydd is Associate Professor of Political Science at the University of Wisconsin. elizabeth mcclellan is a program associate at the Public Interest Research Group. nicholas miller is a graduate student in the Security Studies Program at the Massachusetts Institute of Technology. assaf moghadam is a fellow in National Security at the John M. Olin Institute for Strategic Studies at Harvard University. michael mousseau is Professor of Political Science at Ko University in Istanbul. rysia murphy is a graduate of Connecticut College. william rose is Professor of Government at Connecticut College. paul staniland is a doctoral candidate in MITs Department of Political Science and Security Studies Program, a predoctoral fellow in the Program on Order, Conict, and Violence at Yale University, and a Peace Scholar at the United States Institute of Peace. robert f. trager is Assistant Professor of Political Science at the University of California, Los Angeles. barbara f. walter is Professor of International Relations and Pacic Studies at the University of California, San Diego. dessislava p. zagorcheva is a doctoral candidate in Political Science at Columbia University.

Acknowledgments

The editors gratefully acknowledge the assistance that has made this book possible. A deep debt is owed to all those at the Belfer Center for Science and International Affairs, Harvard University, who have played an editorial role at International Security. Special thanks go to Diane McCree and Katherine Bartel for their invaluable help in preparing this volume for publication.

Preface

Sean M. Lynn-Jones

hen American Airlines ight 11 ew into the north tower of the World Trade Center at 8:46 a.m. on September 11, 2001, the world entered a new era. Terrorism had long been a fact of international life, but previous attacks had not been as spectacularly murderous. Policymakers in the United States and elsewhere immediately began trying to gure out how to prevent future attacks of similar or greater magnitude. U.S. national security policy was soon dominated by counterterrorism strategy, and the United States found itself at war in Afghanistan and Iraq. As national leaders grappled with these dilemmas, the attacks and the U.S. responses also prompted scholars in the eld of international security studies to devote increased attention to the problem of terrorism. The essays in this volume represent some of the research that has been stimulated by the September 11 attacks and by the continuing terrorist threat from al-Qaida. They explore three central topics in terrorism studies. First, what are the sources of contemporary terrorism? Possible explanations offered in this volume include the transformative impact of globalization, the rise of various types of religious beliefs, and the increasing dissatisfaction of the worlds powerless with perceived injustices and the slow pace of reform. Second, what do terrorists want? How are they using violence to achieve their goals? Do their violent campaigns achieve their goals? This volume offers different perspectives on these questions. Some contributors regard terrorists as essentially rational strategic actors who use violence as a means to their ends. These authors generally agree that terrorists often succeed in attaining their political goals. Other essays in this volume dispute whether terrorists act strategically and argue that terrorism rarely succeeds. The debate between these perspectives illuminates what we knowand what we need to know about what motivates individual terrorists and how terrorist organizations behave. The outcome of the debate will have implications for formulating policies to prevent and defeat terrorism. Third, how can the United States and other countries that are targets of contemporary terrorism respond? What policies are most likely to prevent terrorism? The essays here are generally critical of U.S. policies. Contributors to this volume call for improved efforts to deter terrorists, more effective collaboration between the United States and its allies, and improved intelligence capabilities. Several essays offer contending views on the threat posed by al-Qaida. Some authors claim that al-Qaida is a uniquely threatening informal terrorist network, whereas others contend that its status as a network is a source of xi

Preface xii

weakness that the United States can exploit. Most contributors agree that public diplomacy is essential in the ght against terrorist organizations such as alQaida. The rst section of essays in this volume explores the roots of contemporary terrorism. In Market Civilization and Its Clash with Terror, Michael Mousseau argues that the United States and its allies need to understand the motivations of terrorists if they are to succeed in their campaign to prevent terrorism. Rejecting claims that authoritarianism and poverty are the main causes of terrorism, Mousseau contends that terrorism is often a response to the global spread of market-based economies and liberal-democratic values. When globalization and market economies threaten those who benet from clientalist economies in much of the developing world, terrorism is often the result. Mousseau criticizes the claim that people turn to terrorism when political systemsparticularly authoritarian systemsfail to provide opportunities for addressing socioeconomic grievances. He notes that violence and terrorism occur in democracies such as India, so the lack of democracy is not a convincing explanation of terrorism. If poverty were the cause of terrorism, there should have been much more terrorism in previous decades, when virtually every society was poorer. Moreover, individual terrorists, including most of the hijackers involved in the September 11 attacks, are often educated and have good economic prospects. Mousseau also rejects the argument that Islamic religious beliefs provide the cultural foundation for contemporary terrorism. He notes that the Muslim world did not produce suicidal mass murderers in previous decades. NonIslamic societies in Northern Ireland and elsewhere have provided social support for terror, demonstrating that Muslim culture cannot explain all terrorism. In Mousseaus view, the most promising explanation of contemporary terrorism can be found by examining the impact of market-based globalization on clientalist economies. Clientalist economies, which exist in many developing countries, involve complex systems of reciprocal obligations, patronage, and kinship and ethnic ties. Market economies, on the other hand, rely on contracts between strangers and respect for the common law. Clientalist economies tend to be associated with inward-looking, hierarchical social and political arrangements. Market economies foster individualism, universalism, tolerance, equity, and the prevalence of science over faith-based forms of knowledge. Mousseau argues that the rise of market economies has spurred the growth of market

Preface xiii

civilizationa set of stable, prosperous, economically and politically liberal countries that share values and are at peace with one another. Clientalist societies are likely to resent and reject the norms of market economies. They will see participants in market economies as untrustworthy and self-interested. Clientalist societies that are making the transition to market economies may be particularly unstable, as social anarchy may emerge as clientalist norms decline. The populations of such societies may blame the United States or the West for the breakdown of social order. Mousseau points out that the emergence of market economies based on contractual exchange caused major disruptions in Europe. Contractual exchange began in northwestern Europe in the 1450s and precipitated several centuries of upheaval, including the Protestant Reformation, both world wars, and the rise and fall of multiple antimarket movements, including fascism and communism. It took until the twentieth century for contractual exchange to spread throughout Europe. Similar patterns exist in countries that are now in the midst of a similar transition. Mousseau argues that leaders of clientalist structures are most likely to lead resistance to Westernization and marketization, because their positions are threatened by the erosion of clientalist societies. They are the potential terrorist leaders who are willing to engage in or condone the mass murder of Westerners to assert their own power and tap into widespread antimarket fury. As globalization appears to undermine local societies and economies, the social approval of terrorist mass murder grows. Mousseau contends that a sense of rage against market civilization and its shared liberal values and beliefs a rage that can be inamed with the addition of any immediate causelies just beneath the surface in many developing countries. Mousseau asserts that the United States must abandon three myths about the right strategy to use against terrorism: the myth that the United States must increase its economic aid and explain its policies more clearly; the myth that the United States should promote democracy more vigorously; and the myth that people would not hate the United States if they had greater exposure to American values. All three myths fail to take into account the clientalist sources of anti-Americanism. The United States, in Mousseaus view, should wage the war against terrorism by helping developing countries out of the mire of social anarchy and into market development. U.S. policies should encourage developing countries (1) to create and enforce bodies of common law that are vital to the functioning of a market economy, and (2) to equitably subsidize local private

Preface xiv

enterprises with the goal of widespread employment. These policies should be pursued multilaterally so that the United States does not have to shoulder the entire burden. Mousseau concludes that it is through the establishment of market economies that the United States and its allies can be made safe from terror. In Behind the Curve: Globalization and International Terrorism, Audrey Kurth Cronin identies the causes of the current wave of terrorism and criticizes the United States for its inadequate responses. She argues that recent terrorist attacks are a response to globalization. After acknowledging that dening terrorism is complicated and controversial, Cronin adopts the following shorthand denition of terrorism: the threat or use of seemingly random violence against innocents for political ends by a nonstate actor. Thus dened, terrorism has been around for a very long time. In the modern era, terrorism has been directed against empires, colonial powers, and the U.S.-led international system characterized by globalization. Cronin traces the history of terrorism in its various forms. Although contemporary terrorism seems to have religious roots, Cronin regards it as part of a larger phenomenon of antiglobalization and tension between the have and have-not nations, as well as between the elite and underprivileged within those nations. Terrorism exploits the frustrations of the common people in an era when reform comes too slowly. Cronin argues that the latest phase of terrorism has deep causes and is likely to last at least a generation. Terrorism is now rooted in widespread alienation and religious identity and doctrine. Globalization has had few benets for much of the world. Countries with poor governance and widespread poverty create an environment in which terrorists can win support from people who feel powerless in a globalizing world. According to Cronin, there are four types of contemporary terrorist organizations: left-wing terrorists, right-wing terrorists, ethnonationalist/separatist terrorists, and religious terrorists. Each type has different motivations and agendas. Left-wing terrorists tend to prefer revolutionary, antiauthoritarian, antimaterialistic agendas. They often engage in kidnapping, murder, arson, and bombing directed against elite targets. Many such organizations especially in Europehave been brutal, but ephemeral. Right-wing terrorists have tended to choose their targets on the basis of race, ethnicity, religion, or immigrant status. Their organizations tend to be less cohesive, impetuous, and hard to track. Ethnonationalist/separatist terrorists usually have clear political or territorial aims that are negotiable. They generally have some popular support and often engage in conventional insurgencies as well as terrorist attacks.

Preface xv

Religious terrorists have distinctive characteristics that make them especially dangerous, for ve reasons. First, they believe that they are engaged in a struggle of good against evil that justies indiscriminate attacks. Second, they believe that they are acting to please a deity, which makes their behavior unpredictable and insensitive to the reactions of others. Third, they do not feel a need to follow secular laws and values and may even be attacking law-based secular society. If they aspire to overturn the current state system, they are a much more fundamental threat than, for example, separatist terrorists who seek their own state. Fourth, their alienation from the existing system may incline religious terrorists to stage highly destructive and apocalyptic attacks, such as those of September 11, 2001. Fifth, religious terrorism has dispersed popular support and can spread globally. Attempts to ght religious terrorists also may alienate those who share their faith. Cronin identies four trends in modern terrorism that had emerged by the late 1990s. First, religious terrorism was increasing. Second, the overall number of attacks was falling. Third, the number of casualties per terrorist attack was increasing, partly because more attacks were religiously motivated. Fourth, Americans were increasingly targeted. These trends prompted fears that the next step in the evolution of terrorism would be attacks with chemical, biological, nuclear, or radiological weapons. At the same time, terrorist organizations, including al-Qaida, have become more capable of acting globally, drawing funds, recruits, and support from multiple countries and staging attacks far from where the organizations are based. Cronin suggests that globalization, dened as a gradually expanding process of interpenetration in the economic, political, social, and security realms, uncontrolled by (or apart from) traditional notions of state sovereignty,1 is an important inuence on contemporary patterns of terrorism. It has contributed to changes in terrorist methods. Information technologies have expanded the global reach of terrorist organizations, including al-Qaida, which rely on the internet to communicate, recruit, and raise funds. The removal of barriers to the ow of goods and people has enabled terrorists to move freely and to establish cells in many countries. Formal and informal international nancial networks make it easier for terrorist groups to collect, disburse, and hide their funds. Globalization has also changed the objectives of terrorist organizations. It can bring disruptive changes to conservative cultures, or it can anger those who feel that they are not receiving their share of its benets. In either case,
1. Cronin attributes this definition to Victor D. Cha, Globalization and the Study of International Security, Journal of Peace Research, Vol. 37, No. 3 (March 2000), pp. 391393.

Preface xvi

the United States is likely to be blamed. The political incentives to attack the United States have increased as some cultures feel threatened by the United States and the secular vision of modernization it embraces. Cronin points out that the disenfranchised areas of the Arab world increasingly resent the United States as they lag behind the West in their human and economic development. She recognizes, however, that anti-American terrorism is also motivated by a desire to change U.S. policies in the Middle East and the Persian Gulf. Cronin recommends that the United States continue to improve its military capabilities while recognizing that nonmilitary instrumentsintelligence, public diplomacy, cooperation with allies, international legal instruments, and economic assistance and sanctionswill be more effective in the long run. She calls for the United States to create an incentive structure that rewards countries with good governance and intervenes to assist failed states. The United States also must project a vision of sustainable developmentof economic growth, equal access to basic social needs such as education and health, and good governancefor the developing world. The nal essay in this section explains the rise of one type of contemporary terrorism: suicide attacks. From the World Trade Center to Iraq and Afghanistan, suicide terrorist attacks have become more frequent in recent years. What explains the rise of this particular form of terrorism? In Motives for Martyrdom: Al-Qaida, Sala Jihad, and the Spread of Suicide Attacks, Assaf Moghadam examines the causes and characteristics of suicide terrorism. He argues that two factors explain the proliferation of suicide terrorism: the evolution of al-Qaida into a global terrorist actor and the growing appeal of Sala jihad, the guiding ideology of al-Qaida and its associated movements. According to Moghadams data, suicide attacks have increased dramatically since 2000. There were relatively few attacks in the 1980s and early 1990s, followed by a slight increase in the mid-1990s. Suicide attacks then rose from 54 in 2001 to 535 in 2007. During the same period, greater numbers of terrorist organizations perpetrated suicide attacks. More than 20,000 people were killed in suicide attacks between December 1981 and March 2008. Moghadam assesses existing explanations of suicide terrorism and nds them inadequate. Studies that have focused on the motivations of individual suicide terrorists have found multiple motivations, including a commitment to a cause, a desire for revenge, expectations of benets after death, and personal crisis. These studies have not, however, explained why some individuals in these categories become suicide terrorists and others do not. They also cannot explain the globalization of suicide terrorism. Other studies emphasize organizational-strategic explanations of suicide ter-

Preface xvii

rorism. Some analysts claim that suicide attacks enable a terrorist organization to achieve its political goals and strengthen itself. Moghadam suggests that these studies may exaggerate the strategic effectiveness of suicide terrorism. In addition, they do not explain why, if suicide terrorism is effective, so many terrorist groups do not employ it as a tactic, or why its use has increased greatly since 2000. Another set of explanations argues that terrorists will resort to suicide attacks when there is social support for the tactic. The veneration of suicide terrorists as martyrs in, for example, Lebanon, is said to account for the rise in suicide attacks there. Moghadam points out, however, that suicide attacks are becoming more frequent in places such as Afghanistan, where there is no tradition of domestic support for such attacks. One of the most prominent explanations of the rise of suicide terrorism has been offered by Robert Pape, who argues that suicide terrorists are often responding to foreign occupation.2 Moghadam disagrees, however, and points out that attacks occur in countries where there is no occupation, such as Pakistan. He also notes that suicide attacks are directed against internal groups that are not foreign occupiers (e.g., Shiites in Iraq), and are perpetrated by citizens of countries that are not under occupation. Moghadam argues that al-Qaida encourages suicide attacks for religious and ideological reasons, not because it is reacting to foreign occupations. Another prominent explanation of suicide terrorism has been advanced by Mia Bloom, who argues that terrorist groups launch suicide missions as part of a process of outbidding in which they compete with other groups for the support of the local population.3 Moghadam acknowledges that this thesis may apply to some groups, such as the Popular Front for the Liberation of Palestine, the Fatah al-Aqsa Martyrs Brigades, and Amal in Lebanon. The outbidding explanation does not appear to apply in other cases, however. In Sri Lanka the Tamil Tigers adopted suicide terrorism only after they eliminated their main rival. Moreover, in Sri Lanka there is little evidence that the population supports suicide attacks and would be more likely to back a group that used this tactic. Moghadam argues that the best explanation of the global rise of suicide terrorism rests on (1) al-Qaidas emergence as a global actor, and (2) the growing

2. Robert A. Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (New York: Random House, 2005). 3. Mia Bloom, Dying to Kill: The Allure of Suicide Terror (New York: Columbia University Press, 2005).

Preface xviii

appeal of its ideology of Sala jihad. Al-Qaida was initially intended to be a rapid reaction force that would assist threatened Muslims anywhere. Its globalization accelerated when foreign ghters who joined the war against Soviet forces in Afghanistan returned to their home countries or headed to other countries. In the mid-1990s, al-Qaida also decided to focus its efforts on the far enemythe United Statesinstead of local Arab regimes. Al-Qaida made suicide attacks its primary tactic. Its leaders developed a cult of martyrdom based on the morality of self-sacrice and the promise that those who gave their lives would be rewarded in paradise. When the Egyptian terrorist group al-Jihad merged with al-Qaida, its emphasis on suicide attacks inuenced Osama bin Ladens decision to make this tactic central in al-Qaidas operations. Suicide missions also were adopted because they offered a means of imposing heavy costs on powerful adversaries. The ideology of Sala jihad is a radical offshoot of Salasm, which upholds a strict interpretation of Islam based on emulating the lifestyle of the Prophet Mohammed and his disciples. Unlike ordinary Salas, Sala jihadists advocate violent jihad, including suicide operations. They label some fellow Muslims as indels (a practice known as takr) and condone the killing of civilians. Sala jihadist preachers in mosques and on the internet have instilled a cult of martyrdom in their followers. Their teachings have become increasingly popular in Africa, Central Asia, the Middle East, and Southeast Asia. Moghadam nds that Sala jihadist groups carried out 37.7 percent of all suicide terrorist attacks between December 1981 and March 2008more than any other type of group. He claims it is likely that many of the 57.6 percent of suicide attacks that were perpetrated by unknown groups also can be attributed to Sala jihadists, because many of these attacks occurred in Iraq, where Sala jihadists conducted numerous suicide attacks. Sala jihadist groups also make up an increasing proportion of the groups that conduct suicide attacks. They have placed special emphasis on global suicide attacks in which terrorists connected to transnational networks attack targets outside their own countries. How can Western countries respond to suicide terrorism? Moghadam recommends that moderate Muslims must take the lead in arguing against Sala jihad. He writes, Western states can underscore what most Muslims already know: the credibility of Sala jihad suffers from a fundamental contradiction. On the one hand, Sala jihadists claim to act for the benet of Muslins. On the other hand, Muslims suffer the consequences of Sala jihadist ideology and terrorism more than any other group.

Preface xix

The next section of essays in this volume examines the strategies that terrorists choose to achieve their goals, whether these strategies are likely to succeed, and how the targets of terrorist attacks should respond. There is a rich debate over the effectiveness of terrorism and the motivations of terrorists. Multiple perspectives on these questions are included in these pages. In The Strategies of Terrorism, Andrew Kydd and Barbara Walter argue that terrorism often works. Terrorism instills fear in target populations, forces governments to make concessions, and provokes governments to respond in ways that aid the terrorists cause. Kydd and Walter seek to explain terrorists choice of goals and strategies, the effectiveness of the strategies chosen, and how to respond to terrorism. Kydd and Walter argue that terrorism is a form of costly signaling that attempts to persuade audiences that the terrorists are committed to their cause. Terrorist strategies may include (1) attrition, (2) intimidation, (3) provocation, (4) spoiling, and (5) outbidding. Understanding these ve distinct strategic logics is crucial for devising effective counterterrorism policies. Kydd and Walter observe that terrorism is usually not senseless violence, but a set of actions intended to achieve one or more political goals. They identify ve goals that terrorists often pursue: regime change, territorial change, policy change, social control, and status quo maintenance. A given group may have multiple goals. Of the forty-two groups on the U.S. State Departments list of Foreign Terrorist Organizations, thirty-one seek regime change, nineteen seek territorial change, four seek policy change, and one seeks to maintain the status quo. In international politics, uncertainty about states power, resolve, and trustworthiness means that states often must rely on costly signals to communicate. Verbal statements may not be credible. In a crisis, for example, states order military mobilizations, and sometimes war itself is necessary to communicate resolve. Kydd and Walter contend that a similar logic applies to terrorists, who need to provide credible information to the audiences whose behavior they hope to inuence. Those audiences include governments whose policies they wish to change, as well as the terrorists supporters. Kydd and Walter analyze each of the ve potential terrorist strategies to determine when each strategy can succeed and how each can be countered. Attrition strategies attempt to demonstrate that terrorists are strong and resolute and will inict serious costs to achieve their goals. For example, Palestinians during the second intifada attempted to impose high costs on Israel in an effort

Preface xx

to persuade Israel to withdraw from the occupied territories. Suicide terrorism is often a key element of an attrition strategy. Kydd and Walter argue that attrition strategies are more likely to be effective when the target state has a relatively low level of interest in the disputed issue, when the target states ability to retaliate is constrained, and when the target state is highly sensitive to the costs of violence. They identify ve counterstrategies as the best responses to terrorist attrition strategies: (1) conceding inessential issues in exchange for peace; (2) retaliating against terrorist leaders, their followers, their assets, and other objects of value; (3) hardening likely targets to minimize the costs of attacks; (4) denying terrorists access to nuclear and biological weapons; and (5) minimizing the psychological costs of terrorism by educating the public about the low risk of being killed by terrorists. Intimidation strategies aim to show that terrorists can punish those who disobey them and that the government is powerless to stop them. Such strategies may consist of killing those who support a regime opposed by the terrorists. Terrorists might adopt an intimidation strategy in cases where the government has refused to yield to their demands and it is easier to impose their policy directly on the population. The Taliban has adopted such a strategy in Afghanistan since the U.S.-backed government of Hamid Karzai came to power. Intimidation strategies are more likely to succeed against weak states and states with large territories or rough terrain. The best response to an intimidation strategy is for the government to retake and control territory. The clear-and-hold strategy pursued by U.S. forces in Iraq is a prime example. Strengthening law enforcement can also deny terrorists the ability to exercise social control. Provocation strategies are intended to incite governments into retaliating against terrorists in a way that harms the population and thereby builds support for the terrorists and opposition to the regime. The strategy recognizes that the population may not initially support the terrorists, but that it will rally to their cause if the government becomes more repressive or murderous. Indiscriminate attacks reveal that a government does not care about its citizens. Kydd and Walter suggest that provocation works against governments that are capable of middling levels of brutality. Provoking a highly brutal government may produce a massive response that will annihilate the constituency the terrorists hope to represent. At the other end of the spectrum, provocation may fail against a government that is so committed to human rights and the rule of law that it will not retaliate indiscriminately. The best response to provocation is a discriminating strategy that kills terrorists without inicting collateral

Preface xxi

damage. Such a strategy requires excellent intelligence, which is not always available. Spoiling strategies aim to prevent governments from reaching peace agreements with moderate leaders who share some of the terrorists goals. Terrorists oppose negotiated agreements that might undermine support for their more radical goals. Spoiler attacks are intended to persuade the enemy that moderates on the terrorists side cannot be trusted to abide by a peace deal. Such attacks have hampered negotiations between Arabs and Israelis and between Protestants and Catholics in Northern Ireland. Spoiling attacks work when the other side believes that moderates on the terrorists side are strong and capable of controlling the terrorists. In such cases, the attacks are seen as evidence that the moderates do not want to control the terrorists. The best responses to spoiling are to build trust and reduce vulnerabilities. Outbidding strategies are adopted when two groups seek the support of their population. Each group tries to present itself as a zealous guardian of the populations interests, so that it will not be accused of being willing to sell out to the enemy. One way to be seen as a zealot is to commit violent acts against the enemy. The competition between Hamas and Fatah for the allegiance of the Palestinian population is a classic case. Outbidding emerges when multiple groups are competing for the allegiance of the population. One of the best responses to outbidding is to encourage competing groups to consolidate. Another response is to make concessions to the moderate groups. Kydd and Walter conclude by arguing that information is a vital part of counterterrorism strategy. Costly signaling, they argue, is pointless in the absence of uncertainty on the part of the recipient of the signal. States with accurate information will, for example, avoid indiscriminate retaliation for terrorist attacks, because they will have the information that enables them to discriminate. Kydd and Walter also emphasize the importance of regime type. Democracies seem more susceptible to attrition and provocation strategies. They may be more sensitive to the costs of terrorist attacks, but they also may face popular pressure to respond to such attacks. Finally, Kydd and Walter note that intelligence gathering and target defense make sense regardless of the strategy adopted by the terrorists, as do attempts to eradicate the root causes of terrorism. Understanding the strategic logic of terrorism is just one part of an effective counterterrorism strategy. In Why Terrorism Does Not Work, Max Abrahms argues that terrorist attacks on civilians rarely coerce governments into making policy concessions. In contrast to what he regards as the prevailing view, Abrahms contends that

Preface xxii

there is little evidence to support the pessimistic claim that terrorists usually succeed. Abrahms analyzes the record of twenty-eight terrorist organizations and nds that they achieved their policy objectives only 7 percent of the time. He suggests that the act of attacking civilians undermines the ability of terrorists to achieve their aims. The targets of terrorist attacks conclude that the terrorists only want to destroy them and their societies and thus refuse to make concessions. Abrahms believes that the conventional wisdom is that terrorism is an effective coercive strategy. He cites many authors who have concluded that terrorism often works. He points out, however, that this conclusion rests on limited empirical support from a few cases, including apparent victories by Hezbollah, the Tamil Tigers, and Palestinian groups. In his view, even some of those victories have been limited. For example, the limited Israeli withdrawals from parts of the Gaza Strip and West Bank in 1994 were not a victory for Palestinian terrorists, because Israel also increased the number of settlers in the West Bank during the same period. According to Abrahms, terrorists can engage in strategic terrorism, which attempts to change government policies, and redemptive terrorism, which is intended to obtain resources such as prisoners or money. He attempts to measure the strategic effectiveness of terrorismwhether it achieves its policy objectivesby examining the record of the twenty-eight terrorist groups described by the U.S. State Department as foreign terrorist organizations since 2001. After determining the objectives of each group by examining their ofcial statements, Abrahms codes the outcomes of each terrorist campaign as a total success, partial success, limited success, or no success. He nds that the terrorist groups had a total of forty-two objectives, but achieved only three of them, for a 7 percent total or partial success ratelower than the success rate for economic sanctions, which are often seen as ineffective. When and why does terrorism succeed? Abrahms considers two explanations: type of objective and target selection. He argues that terrorist campaigns with limited territorial objectives (e.g., ending a foreign military occupation) succeed more often than campaigns with maximal objectives (e.g., destruction of a state), but they still fail most of the time. Target selection is a much better explanation of the success or failure of terrorist campaigns. When terrorist groups attack civilians, they rarely achieve their objectives. Abrahms argues that attacking civilians is an ineffective strategy for terrorists because it convinces the targets of such attacks that the terrorists want to destroy their values, society, or both. He suggests that this outcome is consis-

Preface xxiii

tent with social psychologys correspondent inference theory. This theory suggests that targets of terrorist attacks infer that the terrorists have maximal objectives when such attacks target civilians. The targets of these attacks do not believe that limited concessions will prevent further attacks, and thus they refuse to concede to any terrorist demands. Abrahms examines three cases that support this argument: the Russian response to the September 1999 apartment bombings; the U.S. response to the September 11 attacks; and the Israeli response to the rst Palestinian intifada. In each case, citizens of the target state concluded that terrorists who attacked and killed civilians had maximal objectives. All three countries refused to agree to terrorist demands, and the campaigns of the terrorist groups failed to achieve their objectives. Abrahms concludes that terrorists who attack the U.S. homeland and otherwise kill American civilians are unlikely to achieve their goals. In 2005, surveys in Muslim countries revealed declining support for al-Qaida, although support for guerrilla warfare against U.S. forces in Iraq remained high. He recommends further research on why terrorists attack civilians even though this strategy almost always fails. In a response to Why Terrorism Does Not Work, William Rose and Rysia Murphy agree that terrorists rarely achieve their political objectives, but they argue that the March 11, 2004, Madrid train bombings reveal that terrorists who attack civilians can sometimes achieve their political goals. The case shows that Abrahmss theory does not always apply. The Madrid bombings inuenced the outcome of the subsequent Spanish elections and brought to power a new government that decided to withdraw Spains troops from Iraq. The jihadist network that carried out the bombings was attempting to compel Spain to withdraw its troops from Iraq. After the attack, public opinion polls showed that the Spanish public believed that the attack was a response to Spains involvement in the Iraq War and that withdrawing Spains forces from Iraq would make future attacks less likely. Three days after the attack, Spains citizens voted into power the Socialist Party, which had called for pulling out of Iraq. Polls suggested that the bombings inuenced the outcome of the election. The governments attempts to blame the Basque separatist group ETA were seen as a refusal to acknowledge the connection between Spains role in the Iraq War and the attacks. Rose and Murphy acknowledge that this case may be unusual, but it demonstrates that terrorists sometimes succeed. They argue that the Spanish case suggests that policies intended to prevent terrorist attacks should focus on countries that are most likely to be vulnerable to ter-

Preface xxiv

rorist coercion. Governments also will need to retain the trust of their citizens by not engaging in partisan interpretations of terrorist attacks. Abrahms replies that the Madrid bombings case does not prove that terrorist coercion succeeded, for three reasons. First, the Socialist Party may have won the election even if the attacks had not taken place. Polls showed the parties were virtually tied before the attacks and many undecided voters were leaning toward the Socialists. Second, the voters may have turned against the government for its apparent attempt to mislead the public about the attacks, not because of its policies on Iraq. Third, the Madrid case is unusual because terrorism against democracies usually shifts the electorate toward the right. Abrahms acknowledges, however, that the attacks in Spain did not induce the Spanish public to conclude that the terrorists had maximalist objectives. The public already believed that Spanish involvement in Iraq made Spain vulnerable to terrorism, and therefore it was inclined to believe that the terrorists had the limited objective of persuading Spain to withdraw from Iraq. Max Abrahms analyzes what motivates terrorists and why people become terrorists in What Terrorists Really Want: Terrorist Motives and Counterterrorism Strategy. He argues that determining what terrorists want is the key to devising effective counterterrorism strategies. In his view, the strategic model, which holds that terrorists are rational actors who attack civilians for political ends, has been the standard explanation of what terrorists want. Abrahms tests the strategic model, nds it awed, and contends that most terrorists are motivated by the socialnot politicalbenets of terrorism and that counterterrorism strategies should be reoriented accordingly. The strategic model, according to Abrahms, rests on three assumptions. First, terrorists are motivated by relatively stable and consistent political goals. Second, terrorist groups weigh their political options and resort to terrorism only after determining that alternative political avenues are blocked. Third, terrorists expect to achieve their political objectives by attacking civilians and abandon armed struggle when it fails or when better options arise. Abrahms questions whether the strategic model actually applies to terrorist groups. He identies seven empirical puzzles that seem to contradict the premises of the strategic model: (1) terrorist groups do not achieve their goals; (2) terrorist organizations do not use terrorism as a last resort and rarely become nonviolent political parties; (3) terrorists reject compromise proposals with signicant concessions; (4) terrorist organizations have protean political platforms; (5) terrorist groups attack anonymously, preventing their targets from making political concessions; (6) terrorist groups with identical political

Preface xxv

platforms often attack one another; and (7) terrorist organizations do not disband when their goals have been achieved or when they consistently fail to achieve them. If the strategic model is awed, what explains terrorist behavior? Why do people become terrorists? Abrahms suggests that people participate in terrorist organizations for the social solidarity, not for their political return. He draws on organization theory for the natural systems model, which holds that individuals join organizations on the basis of personal inducements, which are often unconnected to the organizations stated goals. Abrahms offers evidence that people join terrorist groups to develop strong affective ties with other terrorists. Socially alienated individuals are disproportionately represented in terrorist organizations. Terrorists from many groups, including the Irish Republican Army (IRA), ETA, the Red Brigades, and Turkish terrorist organizations, say they joined to develop or maintain social relations with other members. Studies also show that al-Qaida, Hamas, and Hezbollah members often joined because a friend or relative was already a member. Many members of terrorist organizations never develop an understanding of their groups political purpose. Even leaders of such groups often cannot explain their organizations goals. Terrorist groups recruit from the ranks of the socially alienated. Their tight-knit structures attract new recruits and boost morale. Viewed in the light of natural systems theory, the puzzles for the strategic model become explicable. Abrahms argues that the puzzles demonstrate that terrorist groups are regularly prioritizing the maintenance of the terrorist organization over the advancement of its political agenda, just as the natural systems model predicts. He concludes that counterterrorism policies need to be reoriented. Instead of attempting to prevent terrorists from achieving their goals, encouraging accommodation to remove grievances, or promoting democracy, countries that want to prevent terrorism should focus on identifying and monitoring socially alienated individuals and break the social bonds within terrorist organizations. Abrahmss analysis of terrorist motivations has provoked a lively debate over what makes terrorists tick. In their response, Erica Chenoweth, Nicholas Miller, and Elizabeth McClellan welcome Abrahmss critique of rational choice interpretations of terrorist behavior, but argue that he goes too far in rejecting the strategic model. Drawing on a 2008 RAND study, they contend that a strategic perspective can account for each of the seven puzzles identied by Abrahms. In their view, Abrahms does not adequately address the strategic

Preface xxvi

calculations of the leaders of terrorist groups, although he does contribute to our understanding of why terrorist foot soldiers become and remain terrorists. Hillel Frisch offers a different critique. He argues that Abrahms errs by lumping all terrorist organizations together; some terrorist groups are dedicated to achieving their political goals, whereas others are not. Frisch also argues that terrorist groups behave more rationally and strategically than Abrahms suggests. Groups that, for example, ght other groups with similar goals, may be strategically pursuing hegemony. Terrorist groups may alter their goals or adopt protean platforms, but so do nonviolent political parties. Many political organizations modify their objectives as they change their assessment of what can be achieved. Like Chenoweth, Miller, and McClellan, Frisch questions whether the same analytical framework can be applied to individual terrorists and the strategies of the organization as a whole and its leaders. He recommends that further research explore why some terrorist organizations act more strategically than others. Paul Stanilands reply to Abrahms includes three criticisms. First, he argues that there are multiple reasons why individuals join terrorist organizations. Second, he points out that the motivations of the foot soldiers in a group may be different from the political goals of the group, as dened by its leadership. Individuals may join a group for many different reasons, but the leaders generally harness these motivations toward a common goal. Third, Staniland argues that terrorist groups do not behave in the way predicted by Abrahms. He offers evidence from the cases of the IRA and the Tamil Tigers. Staniland concludes that the empirical record does not mean that either militants or armed groups are purely strategic or rational actors, but it does suggest that they are not simplistic solidarity maximizers. Abrahms responds to these criticisms. In reply to Chenoweth, Miller, and McClellan, he argues that studies nd that terrorist recruits are motivated by social factors and that terrorist leaders act in ways to keep their groups intact, instead of focusing on their political objectives. He suggests that the RAND study cited by Chenoweth, Miller, and McClellan supports his conclusions. The study, for example, nds that only 4 percent of terrorist groups since 1968 have achieved their objectives. He also argues that terrorist groups behave rationally, if rational behavior is dened as taking steps to perpetuate their own existence. They do not, however, act strategically in pursuit of their political objectives. Abrahms replies to Frisch by arguing that Frisch shows only that terrorist

Preface xxvii

groups do not always engage in all of the behavioral tendencies that Abrahms enumerated. In response to Frischs suggestion that some terrorist groups are motivated by social benets but others are motivated by political benets, Abrahms argues that the terrorist groups of greatest concern to the United States and its allies are not rational political actors. Abrahms responds to Staniland by pointing out that he focused on terrorist organizations, whereas Staniland draws on evidence from other types of militant groups. He also points out that he was not trying to offer a single answer to the question, What do terrorists want? Instead, he compared social and political explanations of terrorist behavior to see which was more powerful. Abrahms argues that Stanilands empirical critique is awed methodologically and rests on a misinterpretation of the history of the IRA. He contends that Irish republican terrorist groups did not even remotely behave as political maximizers, notwithstanding the widespread belief in the strategic model. The essays in the following section assess various aspects of how to respond to the terrorist threat. The authors analyze whether terrorists can be deterred, discuss how the United States can work more effectively with its allies, and consider whether al-Qaidas networked structure is an asset or a liability. Many analysts and observers doubt that terrorists can be deterred. Such claims rest on assumptions that terrorists are irrational, unafraid of punishment, and impossible to locate and target. In Deterring Terrorism: It Can Be Done, Robert Trager and Dessislava Zagorcheva argue that terrorists can be deterred by strategies that hold their political goals at risk. In addition, even if terrorists are hard to nd and deter, they rely on support systems that may be more vulnerable to traditional forms of deterrence. Trager and Zagorcheva note that deterrence is only one strategy for countering terrorism. Other strategies include a focus on winning hearts and minds, economic aid and democratization, appeasement, and military force. They dene a deterrence strategy as one that has two elements: (1) a threat or action designed to increase an adversarys perceived costs of engaging in particular behavior, and (2) an implicit or explicit offer of an alternative state of affairs if the adversary refrains from that behavior. Deterrence includes threats of punishment in response to an actual or a threatened terrorist attack, as well as denial involving hardening of targets to make them too costly to attack. For deterrence to work, the threatened party must understand the threat and calculate the costs and benets of its actions. In addition, the deterrer must be able to hold at risk something that the adversary values;

Preface xxviii

the adversary must place a higher value on what is held at risk than the value of taking action; and both parties must be able to make credible promises. Trager and Zagorcheva assert that the three main arguments posited by other scholars and analysts for why terrorists cannot be deterred are unpersuasive. First, they claim that the problem of irrationality does not make it impossible to deter terrorists. Recent studies suggest that terrorist groups often have a set of hierarchically ordered goals and choose strategies that best advance them. Even if terrorists combine rational and nonrational behavior, they can be deterred because they are sufciently sensitive to cost-benet calculations. Second, Trager and Zagorcheva argue deterrence can work even when a state is threatened by highly motivated terroriststhe sort of terrorists who would be labeled fanatics by most observers. They note that terrorist networks often contain less motivated elements, such as nanciers, who can be deterred. It also may be possible to deter state sponsors of terrorism. In addition, they argue that deterrence can work against a terrorist group that has local goals but also may cooperate with transnational terrorist networks that have other objectives. Groups such as the IRA, ETA, and the Tamil Tigers, for example, might be less likely to cooperate with anti-American terrorists if they knew that the United States would retaliate against them and make it harder for them to achieve their objectives. In some cases, states may be able to deter major attacks while tacitly accepting minor attacks. Regardless of the motivations of terrorist groups, deterrence by denial (hardening targets and increasing security) can be a successful strategy. Deterring multiple terrorist groups may require focusing deterrent threats on the most dangerous group. Third, Trager and Zagorcheva question whether the return address problem actually prevents deterrence of terrorists. They argue that members of terrorist groups usually can be found if states invest sufcient resources in intelligence. Even when terrorists cannot be found, states can make it harder for terrorists to attain their political goals. Trager and Zagorcheva advocate providing economic and military aid to governments targeted by insurgents, pressuring targeted states not to make concessions to terrorists, aiding other groups with goals that are opposed to those of the terrorist group, and imposing travel and fundraising restrictions on terrorist group members. They point out that less motivated terrorists will be deterred, even if the chances of catching all of them are low. Trager and Zagorcheva test their central argument with evidence from the cases of the Moro Islamic Liberation Front (MILF) and the Abu Sayyaf Group.

Preface xxix

Both of these groups operate in the southern Philippines. Coercion was attempted against the MILF, and force was used against the Abu Sayyaf Group. The MILF is a rebel group that seeks greater autonomy for the Moros in the Philippines. It has engaged in terrorism, although it denies such charges, and it has cooperated with al-Qaida and Jemaah Islamiah, a radical Islamist group in Southeast Asia. Western intelligence reports indicate that the MILF operated camps for foreign jihadists at al-Qaidas request. The MILF, however, condemned the attacks on the World Trade Center and rejected the Talibans call for a jihad against the United States. Although the MILF is highly motivated in its battle for Moro autonomy, coercive threats may deter it from cooperating with al-Qaida. After the September 11 attacks, the United States threatened to include the MILF on its list of Foreign Terrorist Organizations if it did not cease its violence against civilians. It also warned that it would cut aid earmarked for areas in the southern Philippines if the MILF did not cut its ties with Jemaah Islamiah. The MILF responded by promising to help Philippine authorities arrest alQaida and Jemaah Islamiah operatives and to assist the government in its ght against the Abu Sayyaf Group. Military action against the Abu Sayyaf group had mixed results. A joint Filipino-U.S. operation in 2002 killed several hundred Abu Sayyaf members and evicted the group from its stronghold on Basilan Island. On the other hand, the groups leadership remained intact and the group moved to Jolo Island. One American hostage was rescued, but two were killed. The Abu Sayyaf Group then began a series of bombings, one of which killed a U.S. soldier. Trager and Zagorcheva speculate that members of the group who survive U.S. attacks may become more radicalized and may even join global terrorist groups when they cannot achieve their local objectives. Trager and Zagorcheva conclude by offering ve recommendations for U.S. counterterrorism policy. First, it is sometimes possible to deter critical elements of terrorist networks from participating in terrorist enterprises. The United States should pursue those elements vigorously after any attacks to increase the chances of deterring other terrorists. In particular, the United States should pursue terrorist nanciers and their assets. Second, even highly motivated groups can be deterred. The United States may be able to deter groups that might collaborate with al-Qaida if it threatens to deny such groups the ability to achieve their local goals. Trager and Zagorcheva write, By holding at risk the local agendas of local groupsthe United States can often more effectively achieve its ends of preventing cooperation between groups and denying sanc-

Preface xxx

tuary to those against which force will have to be used. Third, the United States may be able to deter many terrorist groups from cooperating with alQaida, because al-Qaida is an unusual terrorist group with fanatical members and sweeping goals. Other groups may want to differentiate themselves from al-Qaida so that they are not prime U.S. targets in the war on terrorism. Fourth, the United States should concentrate its resources on al-Qaida, which represents the greatest threat to the United States. It should also try to deter groups from cooperating with al-Qaida. Fifth, the United States should harden soft terrorist targets and demonstrate its resolve not to back down in the face of terrorist threats. These steps will decrease the coercive leverage of terrorism and reduce terrorists motivation to carry out attacks. Since the September 11 attacks, the United States has aligned itself with states that are ghting Islamist insurgents who often have ties to al-Qaida. In Friends Like These: Counterinsurgency and the War on Terrorism, Daniel Byman examines the reasons why it is so difcult for the United States to work with such allies. He argues that governments ghting al-Qaidalinked insurgencies often have multiple problems that make it difcult to mount effective counterinsurgency campaigns. Such governments are often illegitimate, rent by civil-military tension, economically backward, and prone to social discrimination. To overcome these problems, the United States needs better intelligence on its allies and their security forces. In addition, it should increase its leverage by acting more like a third party in local conicts, instead of aligning too closely with government forces. The United States will need to pick its battles more carefully to avoid unwinnable wars. Byman points out that al-Qaida is a terrorist organization, but it also actively encourages Islamist insurgencies. Such insurgencies benet al-Qaida in many ways, providing recruits, building unity, and adding legitimacy. Defeating alQaida will require defeating or inhibiting its ability to engage in and benet from insurgencies in countries such as Afghanistan, Algeria, India, Iraq, Somalia, and Uzbekistan. Counterinsurgency against groups linked to al-Qaida depends on the military forces of U.S. allies. In many cases, local allies will have an advantage over U.S. forces, because they know the local culture and language and are less likely to provoke a nationalist backlash. Local allies, however, do not have the same interests as the United States. Their armed forces and security agencies may have the mission of propping up an undemocratic regime, and they may use brutal tactics to do so. Byman argues that classic principal-agent dynamics come into play, because the United States depends on its local allies yet cannot control them.

Preface xxxi

Fighting counterinsurgency wars is always difcult, because the security forces involved must combine military skills and political savvy. The forces of U.S allies may be garrison militaries that have trouble ghting small-unit engagements, taking the initiative, gathering and analyzing adequate intelligence, and coordinating different services and units. They also tend to have poor leadership, motivation, and creativity. Their political weaknesses often include tensions between the civilian leadership and military, dishonesty and corruption, penetration by insurgents, and little support from the population. Byman attributes many of the weaknesses of U.S. allies to a set of structural weaknesses: illegitimate and repressive regimes, civilian suspicions that the military will launch a coup, economic backwardness, and social exclusion. One or more of these weaknesses can be found in U.S. counterinsurgency allies such as Afghanistan, Algeria, Egypt, India, Indonesia, Pakistan, the Philippines, Saudi Arabia, and Uzbekistan. Byman argues that the United States cannot easily improve its counterinsurgency allies. The United States tends to ally itself with governments that, through their mistakes and policy failures, have provoked insurgencies. Once the United States backs a government, its support enables the government to avoid reforms. The United States cannot abandon allies that are deemed to be vital in the war against terrorism, so it has little leverage over them. Moreover, reforming some U.S. allies would require an enormous effort that would entail a massive commitment of U.S. resources. Byman recommends that the United States adopt three new policies to make its counterinsurgency alliances more effective. First, the United States should gather more intelligence on its allies and, in particular, on its allies information and collection and dissemination activities. Because allies may use information they gather to portray local insurgents as al-Qaida afliates or to exaggerate the effectiveness of their security services, the United States needs to know more about such information and its sources. Second, the United States needs to gather information from sources other than allied governments, including the international media and independent investigative groups such as the International Crisis Group. Third, the United States sometimes must distance itself from its allies and act as a third party, ghting the insurgents but also demanding that the government reform itself. Most important, the United States needs to adopt realistic expectations about its allies and the limits of U.S. efforts to change them. Byman offers two concluding observations. First, it will be difcult for the government of Iraq to take over the counterinsurgency war waged by the United States in that country. Second, although assisting allies in their counter-

Preface xxxii

insurgency wars will often be in the U.S. national interest, there will be cases in which it would make more sense for the United States to invest its resources elsewhere. In addition to local allies in counterinsurgency wars, the United States may turn to its NATO allies in the struggle against terrorists. In NATOs International Security Role in the Terrorist Era, Rene de Nevers examines how NATO has contributed to counterterrorism missions. She nds that NATO has played a limited role in the ght against terrorism. The United States and its European allies are engaged in multiple counterterrorism operations, but almost all of them are conducted outside NATOs alliance structures. Shortly after the September 11 terrorist attacks, NATO invoked article 5 of its charter and declared that the attacks on the United States represented an attack on all its members. It subsequently agreed to assist the U.S.-led coalition in its war in Afghanistan. NATO then adopted a new concept for responding to terrorism that states that the alliances goal should be to help deter, defend, disrupt, and protect against terrorist attacks. The United States, on the other hand, has emphasized more offensive measures that would attempt to confront terrorists overseas and to hunt down and kill terrorists outside the United States. According to de Nevers, NATOs deeply institutionalized, consensus-based model is not the United States preferred approach for multilateral cooperation in the war on terror. The United States has thus sought looser coalitions of the willing to assist it in its response to terrorist threats. De Nevers assesses four categories of NATOs contribution to the U.S. war on terror: (1) prevention and defense, (2) denial, (3) counterterrorism, and (4) consequence management. NATOs efforts at prevention and defense include intelligence sharing and surveillance operations, such as Operation Active Endeavor in the Mediterranean. Because much of NATOs intelligence is gathered by the United States, there are limits to how much NATO can contribute to the war on terror in this area. The most vital terrorist-related intelligence gathered in Europe comes from police and domestic intelligence agencies, so it is outside the scope of NATO. Denial includes preventing terrorists from acquiring weapons of mass destruction (WMD), sanctuaries, and state support. NATO has done little as an alliance to deny terrorists access to WMD, although its members have been involved in other efforts to secure the Russian nuclear arsenal and to prevent nuclear proliferation in general. The United States has opted to pursue its counterproliferation initiatives through informal coalitions. The United States attitude toward preventive and preemptive attacks differs from that of its

Preface xxxiii

European allies, which makes it harder for NATO to play an active role in denial. NATO has contributed substantially to the goal of denying terrorists sanctuaries, most notably through its contribution of forces in Afghanistan. NATO members have not, however, contributed as many troops as desired, and there are limits on how some of the forces can be deployed. NATO has played a limited offensive counterterrorism role, because the alliance has focused on defense. Only in Afghanistan has NATO taken on combat missions. Even there, NATO members have questioned whether their forces should be engaged in war ghting on such a large scale. NATO has attempted to enhance its capabilities for consequence management so that it can respond to terrorist attacks after they occur and minimize their effects. It has, for example, developed capabilities to respond to chemical and biological terrorism, and it has a Euro-Atlantic Disaster Response Center. NATOs consequence management programs have two limitations, however. First, individual countries must request NATO assistance. Neither Spain nor Britain requested help after the 2004 and 2005 bombings in those countries. Second, there is disagreement over whether NATO or the European Union should play the leading role in emergency response. According to de Nevers, three factors explain why the United States has not cooperated more closely with NATO in the war against terror. First, the emergence of U.S. hegemony and the development of a security community in Europe since the end of the Cold War have caused NATOs members to differ on threat perceptions and the role of military force. Second, U.S. military capabilities are much more advanced and sophisticated than those of other NATO countries, which makes it hard for the allies to coordinate military operations. Third, much of the war against terrorism is fought by nonmilitary means, such as law enforcement and intelligence gathering, in which NATO has a limited role. NATOs future military role in the war on terrorism is in doubt, de Nevers concludes. Its ability to contribute to the war in Afghanistan may be a crucial test of its viability. De Nevers notes, however, that there are signs that the United States is beginning to adopt European policies on terrorism, which stress intelligence, law enforcement, and quiet engagement with the Muslim world. This convergence may ease transatlantic tensions, but it would not necessarily revive NATO. Many observers have claimed that al-Qaida is a uniquely threatening terrorist group because of its decentralized network structure. This perspective emphasizes the difculty of ghting a network that is resilient, exible, and able

Preface xxxiv

to recruit from many different sources and areas. In Assessing the Dangers of Illicit Networks: Why al-Qaida May Be Less Threatening Than Many Think, Mette Eilstrup-Sangiovanni and Calvert Jones argue against this conventional wisdom. They contend that networks such as al-Qaida have signicant disadvantages. Eilstrup-Sangiovanni and Jones dene a network as a collection of actors that pursue exchange relations with one another but lack an organization to resolve disputes that arise in the course of such exchanges.4 Unlike traditional hierarchical organizations, networks are at and decentralized. Decisionmaking is dispersed, and local actors enjoy much autonomy and form lateral links with one another. These links involve high levels of trust and reciprocity. Networks have several apparent advantages over traditional hierarchical and centralized organizations. First, information can ow quickly from one part of the network to another. Actors can process and act on information quickly. The information itself may be of higher quality, because the trust between networked actors enables people to share and collectively interpret information. Second, networks have scalability, the ability to grow sideways by adding links to new actors. The advent of the internet has facilitated this type of expansion. Third, networks are adaptable. The absence of a centralized structure enables them to change their form quickly. They also can easily expand into new geographical areas. Fourth, networks are resilient. They cannot be destroyed quickly, because they have multiple nodes and can reconstitute easily if one node is destroyed. Fifth, networks have a high learning capacity, because they transfer information quickly and encourage experimentation. Eilstrup-Sangiovanni and Jones argue that the advantages of networks have been exaggerated, at least partly because little research has been done on failed networks and researchers have tended to overlook earlier examples of networks. They point out that networks also suffer from multiple disadvantages. Networks may not be superior structures for gathering and processing information. Decentralization and compartmentalization can make it hard to locate and share information. Individual cells of terrorists with loose ties to their leadership may communicate poorly. Loose networks also may make poor decisions and take excessive risks. Instead of being highly exible and adaptive, they may make decisions slowly and the absence of centralized leadership may reduce respect for those deci4. This definition is based on one in Joel M. Podolny and Karen L. Page, Network Forms of Organization, Research in Organizational Behavior, Vol. 12 (1990), pp. 5859.

Preface xxxv

sions. Too much local autonomy can mean too little strategic planning. Insular underground organizations also may fall victim to groupthink and take excessive risks. Although networks are said to be scalable, in reality their reliance on trust and interpersonal relations may limit their growth. Criminal and terrorist networks often depend on preexisting bonds of family or friendship. When networks try to link with other networks, they often splinter into separate groups. This pattern has emerged with Egyptian militants, Palestinian terrorists, and the Aum Shinrikyo cult. When networks manage to grow, they often lose their sense of unity and shared purpose. If multiple centers of power emerge, they may issue conicting messages and undermine the groups legitimacy and popular support. Networks also may be less adaptable than the conventional wisdom suggests, because networks built on personal ties and reciprocity are not always exible. Collective action problems often plague networks. With much local autonomy, networks are beset by inghting between competing centers of power. The Palestine Liberation Organization is a prime example of this pathology. As a loose federation of different factions, the PLO has a long history of internal rivalry. Centralized hierarchical organizations, such as Shining Path in Peru, have avoided this level of fractiousness. Networks also have trouble coordinating their actions across great distances. Even in the age of the internet, groups can miscommunicate when there are limited opportunities for face-to-face interaction. It becomes harder to instill loyalty and commitment in a dispersed network. Networks may be more vulnerable to security breaches. Once one cell is uncovered and destroyed, the network may unravel as other participants blame one another. When recruitment is dispersed, it becomes harder to monitor and screen new recruits. The conventional wisdom holds that networks learn efciently, but in practice limits on information ows may inhibit learning. In some networks, compartmentalization means individual cells often absorb only those lessons they have learned directly rather than benet from the experience of others. Networks supposedly facilitate the exchange of tacit knowledge, but the absence of centralization and hierarchy may mean that members do not receive necessary training. Militants often try to address this problem by establishing training camps, but such camps can be detected and become a security liability. Unlike centralized organizations, networks also have minimal institutional

Preface xxxvi

memory and may not be able to translate lessons learned into solid improvements in organizational practice. Eilstrup-Sangiovanni and Jones look at the record of al-Qaida to determine whether that organization has beneted from its loose networked structure. Many observers see al-Qaida as a prime example of a network that has been resilient and adaptable. Eilstrup-Sangiovanni and Jones, however, make three arguments that cast doubt on claims that al-Qaida is an effective decentralized terrorist network. First, al-Qaida was a hierarchical organization when it carried out its most successful missionsincluding the September 11 attacks. The central leadership of al-Qaida was very structured and hierarchical in the 1990s, with a tidy organization of committees with well-dened positions and responsibilities. The most spectacular al-Qaida attackers were closely managed by al-Qaidas central leadership. Moreover, al-Qaida was not a dispersed network. It had a headquarters in Afghanistan under the Taliban. Second, as al-Qaida has evolved into a more loosely structured network, it has lost unity and cohesion. When al-Qaida became a diffuse transnational network, local afliates staged attacks in, for example, Bali, Madrid, and London, without central direction. Rifts may be emerging within al-Qaida as the network expands. Third, many al-Qaida attacks have been foiled in recent years, suggesting that an apparently effective network is vulnerable to counterterrorism measures. Many of al-Qaidas local afliates have adopted poor security procedures, making it easy for domestic police and security services to monitor and ultimately arrest their members. The dispersed elements of the network seem to be slow learners when it comes to adapting to new law enforcement techniques. Eilstrup-Sangiovanni and Jones conclude that al-Qaida may have become a less signicant threat since it evolved into a networked structure. The group has not been able to plan and execute complex attacks since September 11, 2001. Most of its attacks have been initiated by local afliates. Although the ideology of al-Qaida can inspire continuing violence, this danger cannot be attributed to its purported strength as a network. Eilstrup-Sangiovanni and Jones recommend that law enforcement agencies exploit the vulnerabilities of al-Qaida and other terrorist networks by targeting networks repeatedly, forcing actors to change their practices abruptly, or sowing doubt and mistrust through inltration and manipulation of information.

Preface xxxvii

Is the world entering an age of continuing massive terrorist attacks? Will al-Qaida remain a major terrorist threat? Why do terrorist groups decline and disband? The nal essay in this volume explores these questions. In How al-Qaida Ends: The Decline and Demise of Terrorist Groups, Audrey Kurth Cronin argues that particular terrorist groups and the campaigns against them do not last forever; even if terrorism, like war, is an enduring feature of the international system. Al-Qaida in many ways is similar to previous terrorist groups, even if it has evolved into a unique network. Cronin recommends that U.S. policies be changed to take into account the lessons of previous counterterrorism campaigns and research on the rise and demise of terrorist groups. Previous research on how terrorism ends has included analyses of how the causes and motivations of terrorism affect the rise and fall of terrorist groups. Cronin notes that research on the organizational dynamics and ideological commitment of terrorist groups may be less relevant to groups that rely on decentralized cells and the internet. She also points out that terrorist groups motivated by religion may have greater staying power. Other research on the rise and decline of terrorism attempts to identify waves or cycles of terrorist activity, as well as stages through which terrorist groups pass. Cronin points out that cyclical hypotheses are hard to prove and that the current wave of jihadist terrorism may not resemble previous terrorist campaigns. She also notes that comparative case studies of terrorist groups and counterterrorism campaigns are often awed, because researchers do not have access to all the relevant data or they choose to focus only on some variables. Cronin identies seven explanations for why terrorist groups decline and cease operations. First, capturing or killing a leader may cripple a terrorist organization. Recent examples of groups that suffered this fate include the Shining Path in Peru, the Kurdistan Workers Party, the Real IRA in Northern Ireland, and Aum Shinrikyo in Japan. Cronin suggests that killing a groups leader may backre if he or she is then seen as a martyr and new members are inspired to become terrorists. Arresting a leader may be more effective, provided that the leader is not allowed to send encouraging messages to his or her followers while incarcerated. Second, terrorist groups may decline if they fail to pass their cause to a new generation of members. In the 1970s, for example, left-wing and anarchist groups such as the Baader-Meinhof group and the Japanese Red Army failed to inspire successor generations to adopt their goals, ideologies, and violent

Preface xxxviii

tactics. Right-wing groups may have suffered a similar fate, although their racist ideas sometimes endure as their organizations go underground and become a network of decentralized cells. Third, terrorist groups may cease to exist once they have achieved their goals. The Irgun, which fought to establish the state of Israel, and the African National Congress, which fought to end apartheid in South Africa, are two examples. Cronin notes that terrorist groups rarely succeed in their campaigns, but achieving their goals may lead them to disband. Fourth, terrorist groups may end their terrorist activities when they enter into negotiations and become legitimate political actors. The Provisional IRA and the Palestine Liberation Organization are two prominent organizations that have followed this path. In other cases, a terrorist groups entry into negotiations may cause the group to splinter. Some of the emerging splinter groups may be more radical and more violent than their mother organization. Cronin observes that territorially based terrorist groups may be more likely to enter into negotiations than ideological or religious groups. Fifth, terrorist groups may decline or disband when they lose popular support. Terrorists often need active or passive support from the population. They may lose that support if governments are able to offer rewards and punishments that induce the population to end their support. Populations also may become uninterested in a terrorist groups ideology or revolted by its indiscriminate violence. The Real IRA and ETA have alienated potential supporters by engaging in brutal bombing campaigns. Sixth, terrorist groups may decline when they are repressed with military force. This approach has contributed to the decline of Shining Path and the Kurdistan Workers Party. Cronin argues that the results of such repression may only be temporary. In some cases, such as the Russian repression of rebels in Chechnya, military success in one region may export the problem to another area. Military repression is also likely to be costly and may be a particularly difcult strategy for democracies that profess respect for human rights. Seventh, terrorist groups may abandon terrorism and become criminal enterprises or pursue conventional military operations. Terrorist organizations that have increasingly focused on prot-making activities instead of the pursuit of political ends may include Abu Sayyaf in the Philippines and some narco-terrorist groups in Colombia. Terrorists that have escalated to insurgency or conventional war include groups in Kashmir and Maoists in Nepal. Is al-Qaida a unique terrorist organization? Does it have some of the characteristics that have caused previous terrorist groups to decline? Cronin argues

Preface xxxix

that al-Qaidas uid organization, recruitment methods, funding, and means of communication distinguish it from previous terrorist organizations. Al-Qaida has made the transition from being a visible organization with training camps in Afghanistan and is now a much more uid network of multiple afliated groups in many countries. These entities share a commitment to Sunni jihad, but they are hard to track because they rely on internet communications and consist of clandestine cells. It is difcult to estimate the number of al-Qaida members who might engage in terrorist operations. Al-Qaidas recruitment pattern is more like that of a social network. New members are not recruited, but instead volunteer and become part of a social network based on ties of kinship and friendship. As a result, there are often no direct connections between al-Qaidas leadership and its operatives. It is extremely difcult for intelligence agencies to trace the command and control relationships. Funding for al-Qaida comes from money channeled through charitable organizations and the prots from businesses that Osama bin Laden owns. Al-Qaida provides grants for local terrorist groups that often are able to conduct attacks with limited funds. Although some of al-Qaidas funds have been frozen, it is very difcult to control informal networks for nancial transfers or to prevent the ow of funds through charities. Al-Qaida communicates using the full range of technologies of globalization: mobile phones, text messaging, email, websites, blogs, and chat rooms. The internet has provided easy access to the media and potential recruits. Al-Qaida has little difculty distributing its message and practical advice on how to conduct terrorist attacks. Cronin suggests that al-Qaida no longer needs to be an organization in the traditional sense, because individuals can participate with the stroke of a few keys. Despite al-Qaidas innovations, it follows patterns established by earlier terrorist groups, many of which also had international links, sophisticated public communications, and a taste for mass-casualty attacks. Cronin argues that some lessons from the ways in which previous terrorist groups have declined may apply to al-Qaida. First, in Cronins view, killing bin Laden is unlikely to end al-Qaida. Previous organizations that have been crippled by the death of their leader have been hierarchically structured and have lacked a successor. Al-Qaida is not in these categories. It is not driven by a cult of personality. If bin Laden were captured or killed, a successor would emerge and al-Qaida might even be strengthened by his martyrdom.

Preface xl

Second, Cronin argues that al-Qaida will not decline out of a failure to pass on its cause to a second generation. Indeed, it has already inspired a second, third, and fourth generation through its ability to attract radicalized followers who are connected to existing local networks. Cronin contends that the third and fourth ways in which terrorist groups declineachieving the groups goal or entering into negotiationsare not relevant to al-Qaida today. Al-Qaidas goals, including establishing a new Islamic caliphate, removing existing regimes in Muslim countries, and expelling indels from the Muslim worldare so broad that they are unlikely to be achieved or negotiated. The United States may, however, be able to drive a wedge between the disparate groups that have joined al-Qaidas coalition, provided that it does not treat al-Qaida as a monolith. The fth avenue to the decline of terrorist groupsthe erosion of popular supportmay contribute to the decline of al-Qaida. Cronin argues that the U.S. campaign to spread democracy in the Muslim world is the wrong way to erode support for al-Qaida. Democratization will take decades, and there is little evidence to support claims that democratization reduces terrorism. Cronin advocates increased efforts to control or interdict al-Qaidas use of the internet to send messages, as well as a much more robust effort to publicize the human cost of al-Qaidas murderous attacks. The United States should not focus public diplomacy on attempts to promote its values and bolster its image, but should tap into a growing international norm against killing innocent civilians to increase revulsion directed at al-Qaida. The sixth factor that has ended terrorist groups, military repression, will have limited effectiveness against al-Qaida. The movement has shown its ability to evolve in the face of such action. Al-Qaida is already following the seventh pathway to the end of terrorist groups, the transition to criminal activity and conventional military campaigns (at least in Iraq). Cronin notes that this is not a welcome development, because criminal enterprises are being used to fund al-Qaida terrorist operations. They therefore support terrorism instead of replacing it. Cronin concludes that the United States must learn from experience with previous terrorist groups and devise a counterterrorist campaign to accelerate al-Qaidas demise. She argues that al-Qaida will end when the West removes itself from the heart of [a civil war within the Muslim world], shores up international norms against terrorism, undermines al-Qaidas ties with its followers, and begins to exploit the movements abundant missteps. The essays in this volume address some of the most prominent questions in

Preface xli

the analysis of contemporary terrorism. They do not cover every topic related to terrorism. There is little discussion, for example, of the danger that terrorists will acquire nuclear weapons and use them to launch catastrophic attacks.5 It also would have been well beyond the scope of this volume to offer detailed assessments of the many different terrorist groups that are active on every continent except Antarctica. We hope, however, that this collection will stimulate further research and analysis of one of the most important security issues of the early twenty-rst century.

5. For further discussion of nuclear terrorism, see Matthew Bunn, Nuclear Terrorism: A Strategy for Prevention, in Michael E. Brown, Owen R. Cot Jr., Sean M. Lynn-Jones, and Steven E. Miller, eds., Going Nuclear: Nuclear Proliferation and International Security in the 21st Century (Cambridge, Mass.: MIT Press, 2010).

Part I: Roots of Contemporary Terrorism

Market Civilization and Its Clash with Terror

Market Civilization and Its Clash with Terror

Michael Mousseau

lausewitzs dictum that war is politics by other means is a reminder that the primary goal of the war against terror is not to defeat and eliminate those who aim to attack the United States and its allies. Rather it is to enhance the security of the American people and their allies. These goals are the same only if terrorist organizations such as al-Qaeda are isolated groups of criminals that need only be found and dealt with swiftly. But if al-Qaeda and its associated groups represent the values and beliefs of substantial numbers of people, and all signs indicate that this is the case, then defeating these groups will not end the struggle against terror. Only by changing the values and beliefs of supporters of terrorist groups can the United States and its allies expect to achieve this objective. To win the war against terror, the United States and its allies must have both a military strategy and a political strategy. Achieving political victory requires an understanding of the social basis of terrorthat is, the values and beliefs that legitimate the use of extreme and indiscriminate violence against the civilian populations of out-groups. Such understanding will not reveal much about terror groups that seem to lack social support, such as the Basque terrorists in Spain, but it will help to reduce the inuence of those groups that appear to enjoy widespread support, such as al-Qaeda. Seeking to understand the motivations of terrorists, however, should not be confused with empathizing with them or acquiescing on issues that terrorists and their supporters claim motivate them. Some scholars have sought to link poverty with terror. Poverty, they argue, fosters terror because it creates a sense of hopelessness, restricts educational opportunity, and produces frustration over inequality.1 The direct causal linkages between poverty and terror are more elusive than scholars suggest, howMichael Mousseau is Associate Professor of International Relations at Ko University in Istanbul, Turkey. It is with profound gratitude that the author dedicates this article to the memory of Stuart A. Bremer (19442002), whose brilliance, affection, and scholarly leadership will be deeply missed by his many students and colleagues. 1. For a summary of these views, see Martha Crenshaw, The Causes of Terrorism, in Charles W. Kegley Jr., ed., International Terrorism: Characteristics, Causes, Controls (New York: St. Martins, 1990), pp. 113126. For recent examples, see Samuel P. Huntington, The Age of Muslim Wars, Newsweek, December 17, 2001, pp. 4248; and James D. Wolfensohn, Making the World a Better and Safer Place: The Time for Action Is Now, Politics, Vol. 22, No. 2 (May 2002), pp. 118123.
International Security, Vol. 27, No. 3 (Winter 2002/03), pp. 529 2003 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

Contending with Terrorism 4

ever. Indeed I am unaware of any comprehensive explanation in print for how poverty causes terror. Nor has there been any demonstrated correlation between the two.2 Nevertheless, there has been a chorus of calls to increase foreign aid as a tool in the ght against terror.3 Absent an understanding of the social origins of this phenomenon, however, there is little reason to believe that greater foreign aid will have any signicant positive effect. It may even increase the terrorist threat. In this article, I argue that the social origins of terror are rooted less in povertyor in growing discontent with U.S. foreign policyand more in the values and beliefs associated with the mixed economies of developing countries in a globalizing world. I show how liberal-democratic values and beliefs are embedded in the economic infrastructure that prevails in market democracies, and how collective-autocratic values and beliefs are embedded in clientalist economies. As a result of globalization, these values and beliefs are increasingly clashing in the mixed marketclientalist economies of the developing world, triggering intense antimarket resentment directed primarily against the epitome of market civilization: the United States. This study builds on several generations of research in anthropology, economics, political science, and sociology; it explains much of the historical record of sectarian terror around the globe and, most important, suggests how the United States and its allies can combat it. The article is organized as follows. After reviewing the literature on rational and cultural explanations for terror, I show how market democracies constitute a global civilization based not on interstate trade but on common liberal values and beliefs that thrive in market economies. I then discuss the clash of these liberal values and beliefs with the values and beliefs embraced in many parts of the developing world. I next demonstrate how clientalist values are a necessary condition for the resort to terrorist violence. I conclude with recommendations for developing a political strategy to win the war on terror.

Rational and Cultural Explanations for Terror

The academic literature offers two explanations, one rational and the other cultural, for why some societies support terrorism. The rst view holds that ter2. Alan B. Krueger and Jitka Maleckova, The Economics and the Education of Suicide Bombers: Does Poverty Cause Terrorism? New Republic, June 24, 2002, pp. 2733. 3. Many of these calls are made in the media. In the academic literature, see Ivo H. Daalder and James M. Lindsay, Nasty, Brutish, and Long: Americas War on Terrorism, Current History, December 2001, pp. 403408; and Wolfensohn, Making the World a Better and Safer Place.

Market Civilization and Its Clash with Terror 5

rorism is a rational strategy for dealing with particular socioeconomic grievances in societies where the paths to legal expression of opposition are blocked.4 Governments that fail to meet the basic welfare and economic needs of their peoples and suppress their liberties, argues Samuel Huntington, generate violent opposition to themselves and to Western governments that support them.5 In the context of the current war on terror, the Arab world is said to need a managed political opening . . . that introduces pluralism into . . . political life.6 Rational explanations of the origins and social support of terror accord well with mainstream views in academia. Realism, for instance, assumes that values and beliefs play no role in the origin or resolution of conict,7 and thus the resort to terror is a predictable strategy of the weak. Liberal institutionalists argue that democracies are more likely than other kinds of states to resolve their internal (and external) differences through peaceful means.8 In addition, they predict that societies in autocracies are more likely to experience violence and to support terror as an acceptable political tool. Although many developing countries have not produced widespread support for terrorism, such support does seem to be more pervasive in the developing world, especially in those countries lacking stable democratic institutions (e.g., Egypt, Indonesia, and Pakistan). Rational models for explaining the social support of terror have several major weaknesses. The historical record, for instance, does not accord with the proposition that democracies are less likely to condone terror. India, as some observers suggest, has been democratic for more than half a century, yet the threat of sectarian violence seems omnipresent. Nor does the evidence support the notion that poverty or illiteracy increases the threat of terror.9 If economic deprivation were the culprit, then a century or two ago most societies around the world should have supported terrorist activity, because they were generally worse off (in terms of diet, health care, leisure time, and material wealth) than most societies are today. In addition, it is perhaps noteworthy that fteen of the nineteen hijackers who struck at the World Trade Center and the Penta4. Crenshaw, The Causes of Terrorism, p. 116. 5. Huntington, The Age of Muslim Wars, p. 48. 6. Larry Diamond, as cited in Thomas L. Friedman, The Free-Speech Bind, New York Times, Mach 27, 2002, p. A23. 7. Kenneth N. Waltz, Theory of International Politics (New York: McGraw-Hill, 1979). 8. Larry Diamond, Introduction: In Search of Consolidation, in Diamond, Marc F. Plattner, Yunhan Chu, and Hung-mao Tien, eds., Consolidating the Third Wave Democracies (Baltimore, Md.: Johns Hopkins University Press, 1997), pp. xiiixlvii. 9. Krueger and Maleckova, The Economics and the Education of Suicide Bombers.

Contending with Terrorism 6

gon on September 11, 2001, were from Saudi Arabia, one of the richest countries in the world. Most of them were highly educated and appeared to have had ample opportunities for building materially rewarding lives. These facts suggest that rational explanations for the social origins and support of terrorism are inadequate. The September 11 hijackers were motivated by something deepersomething that fundamentally distinguished them from their victims. Put simply, terrorists and their supporters do not think like their victims. From the cultural perspective, terrorists are not merely engaged in a rational strategy of the weak. Rather there is something about ingrained habits and historical traditions that renders terrorism a socially acceptable method for addressing grievances in some societies, but not others.10 When such traditions are combined with social, economic, or political grievances, individuals can be socialized into violence from early childhood, particularly when they experience violence in their formative years.11 From a cultural perspective, the creation of a political strategy to combat terror must begin with an examination of terrorists values and beliefs. What motivates them? What values do they claim justify their actions? Because all the September 11 hijackers were from Islamic countries, and all seemed to express religious motivations, the cultural approach would suggest that there is something inherent in Islamic beliefs and values that yields the social approval of terror. For instance, some observers argue that because the Koran offers instructions for even the minutiae of everyday life, Islamic culture has tremendous difculty dealing with change and lacks a tradition of self-criticism. As a result, some analysts suggest that Muslims tend to be defensive and insecure; they are also likely to blame bad news on exterior, malevolent powers.12 Like rational theories, cultural theories that seek to explain terrorisms origins and base of support have signicant weaknesses. For instance, traditional cultural mores are a constant, not a variable, and thus cultural explanations cannot sufciently account for variation in levels of social support for terror across time and place. More specically, Islamic values and beliefs cannot explain why the Muslim world did not produce suicidal mass murderers in, for instance, the 1950s, or why millions of Muslims around the world joined others in expressing shock and horror at the events of September 11. Further,
10. Crenshaw, The Causes of Terrorism, p. 115. 11. Martha Crenshaw, Thoughts on Relating Terrorism to Historical Contexts, in Crenshaw, ed., Terrorism in Context (University Park: Pennsylvania State University Press, 1992), p. 74. 12. Hume Horan, Those Young Arab Muslims and Us, Middle East Quarterly, Vol. 9, No. 4 (Fall 2002), pp. 5354.

Market Civilization and Its Clash with Terror 7

the social support of terror has a tradition in non-Islamic societies (e.g., Catholics and Protestants in Northern Ireland), demonstrating that Muslim culture alone does not sufciently explain this phenomenon. Rational approaches have an advantage over cultural approaches because they focus on observable circumstancespoverty, economic inequality, illiteracy, and lack of democracythat allow scholars to predict when and where social support for terror is likely to emerge. Explanations linking poverty and its related conditions with terror, however, are nebulous. Cultural approaches have an advantage over rational approaches because they are based on the seemingly apparent fact that those who engage in or support suicidal mass murder do not think like people in out-groups (in this case, people in the United States and the rest of the Western world). None of these approaches, however, helps scholars to predictand thus expose and eradicatethe kinds of values and beliefs that support terror. To grasp the origins of socially approved terror, scholars need an approach that combines the rationalist identication of observable circumstances with the culturalist emphasis on learning why people think and act as they do. In short, scholars must be able to predict when and where the use of indiscriminate violence against out-groups is likely to be socially approved and when and where it is not. Only then can potential terrorist targets devise a political strategy for eliminating this growing menace.

The Rise of Market Civilization

Many scholars of politics have suggested that there is a growing need to be able to predict variation in peoples values and beliefs,13 a need illustrated most dramatically by the September 11 terrorist attacks. In this view, political scientists do not have to start from scratch: Anthropologists, economic historians, and sociologists have been at this task for years. Anthropologists have long sought to explain the relationship between economic conditions and values and beliefs;14 economic historians have for years linked certain economic conditions with particular sets of values and beliefs, identifying at least two primary kinds of socioeconomic integration in historyclientalism and mar13. Robert O. Keohane, Governance in a Partially Globalized World, American Political Science Review, Vol. 95, No. 1 (March 2001), pp. 115; and Alexander Wendt, Social Theory of International Politics (Cambridge: Cambridge University Press, 1999). 14. See Marvin Harris, Cultural Materialism: The Struggle for a Science of Culture (Walnut Creek, Calif.: AltaMira Press, 2001 [1979]); and Maxine L. Margolis, Introduction to the Updated Edition, in Marvin Harris, The Rise of Anthropological Theory: A History of Theories of Culture, updated ed. (Walnut Creek, Calif.: AltaMira Press, 2001), pp. viixiii.

Contending with Terrorism 8

kets;15 and sociologists have documented the social implications of clientalist exchange.16 In clientalist economies, the obligations of cooperating parties are implied (rather than made explicit) and take the form of reciprocity, or gift giving. Exchange occurs through the giving of gifts, which reinforces a sense of trust and enduring obligation among the parties. Enforcement of obligations comes with the threat of punishment: Violations of trust lead to severed relationships. Clientalist economies can be complex;17 and with specialization, patrons emerge who have more to give than others, creating a surplus of obligations accompanied by increased inuence. Because reciprocal obligations are only implied and are socially enforced, patrons rather than states regulate economic cooperation. Examples of clientalist socioeconomies include feudal Europe, and in the contemporary period, maas and the complex systems of patronage that characterize the politics of redistribution in most developing countries.18 Because economic relations are enduring, clientalist economies are based on explicit social linkages, such as kinship and ethnicity. These linkages render ingroups more important than out-groups, making clientalist communities more inward looking than market communities in terms of identity, values, and beliefs. Clientalist communities are also organized hierarchically: Patrons, such as lords, dons, and uncles, receive gifts from clients as expressions of loyalty in exchange for life-long protection.19 In market economies, in contrast, the mutual obligations of cooperating parties are made explicit in the form of contracts. The quid pro quo nature of the
15. A third mode of integration, sharing, is common among hunting and gathering societies, but is not discussed here because it has not been a prominent mode of exchange in any state. See Janet L. Abu-Lughod, Before European Hegemony: The World System, A.D. 12501350 (New York: Oxford University Press, 1989); Karl Polanyi, The Great Transformation: The Political and Economic Origins of Our Time (Boston: Beacon, 1957 [1944]); Marshall D. Sahlins, Stone Age Economics (Hawthorne: Aldine de Gruyter, 1972); and David W. Tandy and Walter C. Neale, Karl Polanyis Distinctive Approach to Social Analysis and the Case of Ancient Greece: Ideas, Criticisms, Consequences, in Colin A.M. Duncan and Tandy, eds., From Political Economy to Anthropology: Situating Economic Life in Past Societies (London: Black Rose, 1994), pp. 1920. 16. See, for instance, Marcel Mauss, The Gift: The Form and Reason for Exchange in Archaic Societies (New York: W.W. Norton, 2000 [1924]). 17. Polanyi, The Great Transformation, pp. 4950. 18. See S.N. Eisenstadt and Ren Lemarchand, Political Clientalism: Patronage and Development (Thousand Oaks, Calif.: Sage, 1981); and Luis Roniger and Aye Gne-Ayata, eds., Democracy, Clientelism, and Civil Society (Boulder, Colo.: Lynne Rienner, 1994). 19. For further discussion of the rules and norms of gift exchange, see Christopher A. Gregory, Gifts and Commodities (San Diego, Calif.: Academic Press, 1983); and Monica Prasad, The Morality of Market Exchange: Love, Money, and Contractual Justice, Sociological Perspectives, Vol. 42, No. 2 (Summer 1999), pp. 181214.

Market Civilization and Its Clash with Terror 9

cooperation implies no obligation among the parties beyond that expressed in the contract. Unlike in clientalist economies, therefore, in market economies, strangers and even enemies can cooperate in prescribed ways.20 Because contracts cannot be negotiated without explicit assertions of self-interest, their extensive use renders such assertions socially approved. Moreover, a contract imposes an equitable relationship on the parties.21 The implications of this are profound: The norm of cooperating with strangers on the basis of legal equality is the logical prerequisite for respecting the rule of common law. Because contractual obligations are explicit, a state can enforce them, and a market economy can emerge if a state is willing and able to enforce contracts with impartiality. In these ways, markets develop and the liberal values of individualism, universalism, tolerance, and equity emerge concurrently with the rule of common law and democratic governance.22 Examples of market economies include classical Athens and, in the contemporary period, Sweden and the United States.23 The market economy and its liberal belief system also account for the rise of science over faith-based forms of knowledge. Science is anchored in the notion that (1) some facts are universal (universalism), (2) any person can challenge anothers assertions of fact, including those of his or her leader (freedom and equity), and (3) truth is sought through the competition of ideas (tolerance). The opposite of science is truth determined by an authority sanctioned by loyalty and faiththe norm in clientalism. All societies have some combination of clientalist and market exchange. For markets to prevail, howeverfor a majority of people to engage regularly in making contractsa complex division of labor associated with economic development is necessary. At lower levels of development and thus incomes, individuals engage in fewer exchanges, and the few big-ticket exchanges that do

20. In this way, the initial emergence of market norms allows for increased specialization and thus greater economic production, which in turn can stimulate a mutually reinforcing cycle of marketintegrated growth. See Michael Mousseau, Market Prosperity, Democratic Consolidation, and Democratic Peace, Journal of Conict Resolution, Vol. 44, No. 4 (August 2000), p. 478. 21. William J. Booth, On the Idea of the Moral Economy, American Political Science Review, Vol. 88, No. 3 (September 1994), pp. 653667; and Ronald Inglehart, Culture Shift in Advanced Industrial Society (Princeton, N.J.: Princeton University Press, 1990), p. 46. 22. For further discussion of this process, see Mousseau, Market Prosperity, Democratic Consolidation, and Democratic Peace; and Michael Mousseau, Globalization, Markets, and Democracy: An Anthropological Linkage, in Mehdi Mozaffari, ed., Globalization and Civilizations (London: Routledge, 2002), pp. 97124. 23. Rondo Cameron, A Concise Economic History of the World: From Paleolithic Times to the Present, 3d ed. (New York: Oxford University Press, 1997), pp. 3235.

Contending with Terrorism 10

occursuch as getting a job, buying a home, or purchasing expensive consumer goodsare less likely to be mediated by the market (with price determined by supply and demand): More often than not, these will be seen as exchanges of gifts among members of an in-group (with price determined by privileged discount). As a result, developing countries tend to have political cultures characterized by intergroup conict (deep in-group/out-group feelings), less respect for individual freedom, stronger religious beliefs, greater respect for loyalty and hierarchy than for the rule of law, and extensive informal patronage networks (known for, among other things, high levels of corruption).24 Sociologists and economic historians have documented the association of gift giving and contracting norms with, respectively, collectivist and individualist value orientations.25 Anthropologists and archaeologists have long considered economic conditions to be a leading inuence on cultural mores and institutional structures.26 Rational choice theorists and others acknowledge that values affect political behavior27; and most agree that, for stability, democracy requires a liberal political culture.28 The chain of causation is well established: The evidence linking economic development with liberal values is so overwhelming that the proposition has no serious detractors,29 nor does the

24. Mousseau, Market Prosperity, Democratic Consolidation, and Democratic Peace; and Mousseau, Globalization, Markets, and Democracy. 25. Abu-Lughod, Before European Hegemony; Fernand Braudel, Afterthoughts on Material Civilization and Capitalism, trans. Patricia Ranum (Baltimore, Md.: Johns Hopkins University Press, 1979), p. 63; Emile Durkheim, The Division of Labour in Society (Basingstoke, U.K.: Macmillan, 1984 [1893]); Polanyi, The Great Transformation; and Tandy and Neale, Karl Polanyis Distinctive Approach. 26. Harris, Cultural Materialism; and Margolis, Introduction to the Updated Edition. 27. Keohane, Governance in a Partially Globalized World; Margaret Levi, Consent, Dissent, and Patriotism (Cambridge: Cambridge University Press, 1997); James D. Morrow, Game Theory for Political Scientists (Princeton, N.J.: Princeton University Press, 1994); Elinor Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action (New York: Cambridge University Press, 1990); and Tom R. Tyler, Why People Obey the Law (New Haven, Conn.: Yale University Press, 1990). 28. Gabriel A. Almond and Sidney Verba, The Civic Culture: Political Attitudes and Democracy in Five Nations (Princeton, N.J.: Princeton University Press, 1963); Robert Alan Dahl, Democracy and Its Critics (New Haven, Conn.: Yale University Press, 1989); Samuel P. Huntington, Will More Countries Become Democratic? Political Science Quarterly, Vol. 99, No. 2 (Summer 1984), pp. 193218; and Seymour Martin Lipset, Some Social Requisites of Democracy: Economic Development and Political Legitimacy, American Political Science Review, Vol. 53, No. 1 (March 1959), pp. 69105. 29. Braudel, Afterthoughts on Material Civilization and Capitalism; Geert Hofstede, Cultures Consequences: Comparing Values, Behaviors, Institutions, and Organizations across Nations, 2d ed. (Thousand Oaks, Calif.: Sage, 2001 [1980]); and Ronald Inglehart and Wayne E. Baker, Modernization, Cultural Change, and the Persistence of Traditional Values, American Sociological Review, Vol. 65, No. 1 (February 2000), pp. 1952.

Market Civilization and Its Clash with Terror 11

stabilizing impact of development on democracy.30 Indeed virtually every economically developed democracy in history has been a market democracy. Although the disciplines of anthropology, economics, political science, and sociology have all addressed different aspects of the relationship between market economies and society, none has examined this relationship in its entirety. Anthropologists and archaeologists typically link cultural mores not to modes of exchange but to environmental conditions;31 some economists have argued that the social implications of markets invalidate the core assumptions of neoclassical liberalism,32 and others have addressed the role of social capital in economic growth;33 political scientists have focused on how development, not the market economy, stabilizes democracy;34 and sociologists have highlighted the social, but apparently not the political, consequences of gift exchange.35 Findings from these four disciplines help to explain the rise of market civilization and its supremacy in the contemporary era. Surveys and other works have established that the inhabitants of high-income countriesmost of which have developed market economiesshare common liberal values;36 other studies conrm that elected leaders seek to promote domestic values in making foreign policy.37 If median voters in market democracies have liberal values

30. Ross E. Burkhart and Michael S. Lewis-Beck, Comparative Democracy: The Economic Development Thesis, American Political Science Review, Vol. 88, No. 4 (December 1994), pp. 111131; and Adam Przeworski and Fernando Limongi, Modernization: Theories and Facts, World Politics, Vol. 49, No. 2 (January 1997), pp. 155183. 31. See, for example, R. Brian Ferguson, Yanomami Warfare: A Political History (Sante Fe, N.M.: School of American Research Press, 1995). 32. See Polanyi, The Great Transformation. 33. Stephen Knack and Philip Keefer, Does Social Capital Have an Economic Payoff? A Crosscountry Investigation, Quarterly Journal of Economics, Vol. 112, No. 4 (November 1997), pp. 1251 1288. 34. Lipset, Some Social Requisites of Democracy; and Dietrich Rueschemeyer, Evelyne Huber Stephens, and John D. Stephens, Capitalist Development and Democracy (Chicago: University of Chicago Press, 1992). 35. Mauss, The Gift; and Prasad, The Morality of Market Exchange. 36. Braudel, Afterthoughts on Material Civilization and Capitalism; Yun-han Chu, Fu Hu, and Chungin Moon, South Korea and Taiwan: The International Context, in Diamond et al., Consolidating the Third Wave of Democracies, pp. 267294; Hofstede, Cultures Consequences; and Inglehart and Baker, Modernization, Cultural Change, and the Persistence of Traditional Values. 37. George C. Edwards III and B. Dan Wood, Who Inuences Whom? The President, Congress, and the Media, American Political Science Review, Vol. 93, No. 2 (June 1999), pp. 327345; Ronald H. Hinckley, Peoples, Polls, and Policymakers: American Public Opinion and National Security (New York: Lexington, 1992); Jeffrey W. Knopf, How Rational Is the Rational Public? Evidence from U.S. Public Opinion on Military Spending, Journal of Conict Resolution, Vol. 42, No. 5 (October 1998),

Contending with Terrorism 12

and median voters in all other types of democracies do not, then only the elected leaders of the market democracies are likely to have liberal values and a political incentive to pursue a liberal foreign policy course. In this way, the common liberal values of their electorates constrain leaders of market democracies (but not leaders of other types of democracies) to pursue common aims in foreign affairs: for instance, to respect and promote international law, human rights, and an equitable global order. Proponents of the democratic peace note the apparent dearth of militarized conict among democratic nations.38 It now appears, however, that this peace is limited to the advanced market democracies.39 Democratic dyads where at least one state lacked a developed market economy and that have had a history of militarized confrontation include India and Pakistan, Greece and Turkey, and Ecuador and Peru. Moreover, market democraciesbut not other types of democraciestend to cooperate with each other against other states.40 They also tend to express common positions in the United Nations General Assembly.41 Of course, leaders of market democracies do not agree on everything, but they do agree on the fundamentals: how the world should be organized politically, economically, and sociallyand what constitutes proper governmental behavior both internally and externally. When differences surface among market democracies, the discourse is bounded by mutual respect for state rights (equity) and the primacy of international lawjust as the domestic political behavior of the governments of these democracies is culturally
pp. 544571; Timothy J. McKeown, The Cuban Missile Crisis and Politics as Usual, Journal of Politics, Vol. 62, No. 1 (February 2000), pp. 7087; Benjamin I. Page and Robert Y. Shapiro, The Rational Public: Fifty Years of Trends in Americans Policy Preferences (Chicago: University of Chicago Press, 1992); and Douglas A. Van Belle and Steven W. Hook, Greasing the Squeaky Wheel: News Media Coverage and U.S. Development Aid, 19771992, International Interactions, Vol. 26, No. 3 (July September 2000), pp. 321346. 38. Stuart A. Bremer, Dangerous Dyads: Conditions Affecting the Likelihood of Interstate War, 18161965, Journal of Conict Resolution, Vol. 36, No. 2 (June 1992), pp. 309341; Bruce M. Russett, Grasping the Democratic Peace: Principles for a PostCold War World (Princeton, N.J.: Princeton University Press, 1993); and James Lee Ray, Democracy and International Conict: An Evaluation of the Democratic Peace Proposition (Columbia: University of South Carolina Press, 1995). 39. Mousseau, Market Prosperity, Democratic Consolidation, and Democratic Peace; and Michael Mousseau, Hvard Hegre, and John R. Oneal, How the Wealth of Nations Conditions the Liberal Peace, European Journal of International Relations, Vol. 9, No. 4 (June 2003), in press. 40. Michael Mousseau, An Economic Limitation to the Zone of Democratic Peace and Cooperation International Interactions, Vol. 28, No. 2 (AprilJune 2002), pp. 137164. 41. Michael Mousseau, The Nexus of Market Society, Liberal Preferences, and Democratic Peace: Interdisciplinary Theory and Evidence, International Studies Quarterly, Vol. 47, No. 3 (September 2003), in press.

Market Civilization and Its Clash with Terror 13

bounded by respect for individual rights and the primacy of democratic law. There is, in short, a market civilization.

The Clash against Market Civilization

Few if any states have predominantly clientalist economies. Most economies are heavily integrated with the market (market democracies) or include some mixture of clientalism and markets (developing countries). Although in many developing countries contracts are ofcially enforced and regulated, in-group linkages can diminish impartiality. In addition, because clientalist exchange is informal, it lies beyond the regulatory capacity of the state. In this mixed economy, the clash of clientalist and market cultures can lead to illiberal and unstable democracy, military dictatorship, state failure, sectarian violence, or some combination thereofand bitter anti-Americanism. In clientalist societies, cooperation occurs with the exchange of gifts, and trust is based on life-long friendships within in-groups. In market societies, loyalty to the in-group is downgraded, as cooperation with strangers is encouraged; trust is based not on friendship but on the perceived universal principle of the sanctity of contractual exchange. Individuals from market cultures thus seek out cooperation universally. From the clientalist perspective, however, those with market values are from out-groups and thus are untrustworthy. Moreover, by expressing self-interest, individuals with market values are viewed as selsh; they appear to have no culture and are seemingly interested in little beyond the crude pursuit of material gain. Cultures change slowly; so when endogenous factors cause a rise in contractual exchange, a clientalist societys economic norms diverge from prevailing cultural values and beliefs. When this happens, individuals with deeply embedded clientalist values have difculty grasping new market norms; they perceive that those who are driven by self-interest not only lack strong social ties but have no values at all. This perception is partly true: A society that undergoes economic change may experience a period when there is no common culture, as clientalist linkages break down before market values emerge. During this period of social anarchy, a zero-sum culture may emerge in which strangers pursue their interests without any regard for shared values market or clientalist. This explains the circumstances in many developingworld societies today: that is, widespread disrespect for the rule of law (everyone wants the law to apply to someone else); social chaos, as many act without

Contending with Terrorism 14

regard for others (e.g., unwillingness to wait in line or obey rules); and the apparent lack of empathy for anyone outside ones in-groups (family, friends, and coworkers).42 From the market perspective, these conditions seem uncivil and are often assumed to be a consequence of local indigenous culture (i.e., a supposed function of Arab culture, Asian culture, and so on). Academics from market cultures have assumed that what people in these countries need is more education,43 a democratic form of government,44 or time to develop.45 As I have sought to show, however, this behavior may not be associated with any particular indigenous culture, form of government, or inherent backwardness. Rather, it may reect the breakdown of clientalist linkages in economies that, facing severe and persistent economic shocks, have not replaced their clientalist values with market values. Although great differences remain across the developing world, traditional clientalist protections tend to be strongest in rural areas. Urban communities, on the other hand, are more likely to be in ux, with new patron-client networks (e.g., political parties, unions, and maas) increasingly replacing traditional patron-client networks (e.g., clans and villages). Strangers in these communities, lacking in both empathy and mutual respect, frequently interact on the basis of few if any common values and beliefs. Meanwhile, in-groups compete over state resources in a zero-sum waywith winners taking all. This helps to explain (1) the high frequency of political violence in developing countries; (2) why democratic institutions in such countries seem to do so poorly in producing public goods, such as roads and security; and (3) why the absence of a strong state often results in chronic instability, civil conict, and in some cases state failure. No economic transition can erase a societys collective history or memory, nor can it eliminate the role of external inuences, ethnic diversity, and historical animosity among competing factions. Moreover, the breakdown of traditional clientalist linkages is not the only source of social anarchy in developing countries: War and state failure can also be factors, as witnessed in Afghanistan and Somalia, where both rural and urban areas remain in tremendous

42. As documented by Hofstede, Cultures Consequences. 43. Daniel Lerner, The Passing of Traditional Society: Modernizing in the Middle East (New York: Free Press, 1958); and Alex Inkeles and David Smith, Becoming Modern: Individual Change in Six Developing Countries (Cambridge, Mass.: Harvard University Press, 1974). 44. Diamond, Introduction: In Search of Consolidation. 45. Talcott Parsons, Evolutionary Universals in Society, American Sociological Review, Vol. 29, No. 3 (June 1964), pp. 339357.

Market Civilization and Its Clash with Terror 15

ux. Likewise, mineral wealth in a developing economy with weak market norms probably works to reinforce the inuence of traditional clientalist ingroups, as patrons spread their riches in return for pledges of loyalty. For the majority of countries without mineral wealth, however, the mire of underdevelopment and economic displacement has meant a rise in social anarchy and civil insecurity. For many individuals living in this rough-and-tumble Hobbesian world, the new zero-sum culture has a thoroughly Western or American character, as seen on television, in movies, and in other forms of popular culture exported from Europe and the United States. Lacking market values and beliefs, millions of people in developing countries believe that the breakdown of traditional clientalist relationships and the emergence of zero-sum anarchy are results of a growing Westernization or Americanization of their societies, and they deeply resent it. Moreover, a society with clientalist values and beliefs but with fading protections from in-groups is extremely vulnerable to any in-group system that promises to put an end to its deep sense of insecurity. This explains the allure of alternative value systems in developing countries that support ethnic sectarianism, extreme nationalism, or various types of religious fundamentalism. A brief examination of the impact of economic change in the contemporary period conrms this view. Contractual exchange in the modern period began in northwestern Europe in the 1450s,46 precipitating for the next 200 years the social and institutional changes brought about during the Protestant Reformation. For three centuries after that, many states in Europe (e.g., England and Holland) began to develop market economiesby enforcing contracts, subsidizing private enterprise, and breaking up clientalist linkages.47 Only in the twentieth century, however, did the majority of Europeans possess the resources to engage regularly in contractual exchange. The combination of nineteenth-century industrialization and mass migration to the United States greatly increased the demand for, and thus the wages of, labor in Europe. As a result, Europes majority, once clients in a clientalist world, became buyers in a new market world.48 The political repercussions of this socioeconomic transition were vast: Euro-

46. Braudel, Afterthoughts on Material Civilization and Capitalism, p. 24. 47. As documented by Polanyi, The Great Transformation. 48. See Simona Piattoni, ed., Clientelism, Interests, and Democratic Representation: The European Experience in Historical and Comparative Perspective (Cambridge: Cambridge University Press, 2001).

Contending with Terrorism 16

pean peasants gave up their way of lifeincluding traditional in-group protectiononly to arrive in cities in the midst of rapid economic change and seemingly devoid of common values and beliefs. Like their counterparts in the nineteenth century, todays migrants confront a bewildering array of zero-sum conditions and Hobbesian anarchy. In both cases, the refugees created by these socioeconomic disruptions sought economic and political protection by joining new forms of clientalist in-groups; and in both cases, they seem to have perceived the rise of market exchange as lacking any redeeming social value. The consequences were the same in both cases: Just as many Europeans in the last century were drawn to clientalist in-groups that championed antimarket (i.e., socialist, communist, or fascist) values, many of todays refugees have been pulled toward antimarket socialist, nationalist, or religious political organizations. I say organizations because these clientalist in-groups are not civicoriented political parties: They offer all-encompassing social, economic, and political programs in exchange for absolute loyalty. In the midst of industrial change, many Europeans joined ethnic sectarian groups, including some that identied European Jewry as the cause of their social anarchy. Frequently, European Jews were merchants and thus tended to behave according to market norms by, among other things, expressing selfinterest through the use of contracts. Faced with the increasing destruction of their traditional clientalist linkages and rising social anarchy, many other Europeans began to equate the proliferation of zero-sum values with Jewish values. Seeking support in socially collapsing societies, some political leaders unleashed antimarket passions by encouraging pogroms against the seemingly cultureless (but really just liberal) Jews. One such leader, Adolf Hitler, was himself from a poor migrant section of Viennaas were many of his followers. While Germany was in the midst of a rapid transition toward a market economy in the 1920s, hyperination eliminated the savings of the nascent middle class. This caused a widespread loss of faith in contracts, a revival of clientalist values, and an antimarket fury that legitimated the mass murder of outgroups. This explains why the Nazis replaced the failing market with a statedirected economy, and why the Germans (and others) became Hitlers willing executioners.49 In fact, across Europe and across time, the strength of antiSemitism seems to correlate negatively, and the stability of democracy positively, with the intensity of the market economy. One indication of this is the
49. Daniel J. Goldhagen, Hitlers Willing Executioners: Ordinary Germans and the Holocaust (New York: Alfred A. Knopf, 1996).

Market Civilization and Its Clash with Terror 17

availability of jobs that offer a living wage. Signicantly, just one generation after the U.S. imposition and subsidization of a market economy in West Germany following the end of World War II, West Germans were well on their way toward developing a liberal political culture.50 Europes transition to a market economy in the nineteenth and early twentieth centuries led to the rise of antimarket socialist, communist, and fascist movements as well as sectarian terror. Similarly, the transition toward a market economy in many contemporary developing countries is associated with antimarket socialist, ethnofascist, hypernationalist, and religious fundamentalist movementsas well as sectarian terror. Examples include the Marxist guerrillas in Latin America, such as the FARC in Colombia and the Shining Path in Peru; increasing ethnic identication, and popularity of hypernationalist political parties, in parts of Russia and Turkey; and the rise of religious fundamentalism in India and much of the Islamic world. Although the character of these movements varies, the catalyst is the same: bitter opposition to market (liberal) values. Herein lays the source of todays widespread anti-Americanism and anti-Westernism: The liberal way of life in the United States and the rest of the Westits cold materialism, from the clientalist perspectiveis being broadcast to homes around the world, many of which are transitioning to market economies. In this way, just as the Jews symbolized emerging market norms in Europe a century ago, today, with modern technology, American and Western culture symbolizes the dreaded market norms linked with globalization.

The Resort to Terror

Those on the lowest rung of the economic ladder are the most vulnerable to the negative consequences associated with globalization. Those with the most to lose, however, are patrons and their lieutenants who hold privileged positions in the old clientalist hierarchies. This is why leaders of terrorist organizations frequently come from privileged backgrounds. To maintain the clientalist structure that carries with it higher social status, these leaders seek to rally their client base by appealing to some antimarket ideology. Because it is in a

50. See Ronald Inglehart, Modernization and Postmodernization: Cultural, Economic, and Political Change in Forty-three Societies (Princeton, N.J.: Princeton University Press, 1997), p. 175. Of course, the socioeconomic transition cannot explain the long history of anti-Semitism in Europe, much of which predates the rise of markets.

Contending with Terrorism 18

clients interest to have a powerful patron, leaders attract and maintain followers by demonstrations of strength. In this way, the mass murder of Westerners serves two purposes: It reects the leaders power, and it taps into widespread antimarket fury. Islam itself is not responsible for the social approval of terror. Patrons fearing the loss of their privileged statussuch as Osama bin Ladennd an antimarket ideology useful to attract followers. They manipulate Islam to serve their own ends, just like their counterparts in Europe did a century ago by contorting Christianity to justify terror and mass murder.51 In fact, Islam emerged in Mecca, the center of sixth-century Mediterranean and South Asian trade, and the Koran stress the market values of universalism, equity, contractual exchange, and a degree of tolerance toward outsiders (non-Muslims).52 The market economy in this region declined before market normsand liberal cultureintensied and expanded throughout the Islamic world, but the liberal origins of Islam demonstrate that religion can be interpreted, and manipulated, to suit anyones purposes. In societies steeped in market values, it is difcult to comprehend how anyone can engage in the mass murder of out-groups, or how anyone can support it. Individuals with market values believe that each person is responsible only for his or her actions. Just as those who are not parties to contracts cannot be made obligated to them, individuals cannot be assumed to be responsible for any and all behavior of other members of their apparent in-group. It therefore seems absurd to blame individuals for the alleged bad behavior of others, and this is the social origin of the presumption of individual innocence in market societies. From the clientalist perspective, in contrast, no one is innocent: Individuals share responsibility for the actions of others within the in-group; if followers do not support their leaders, then they are betraying the entire ingroup. From the clientalist perspective, all in-group members are privileged and all out-group members are enemies or, at best, outsiders unworthy of em51. Although Osama bin Laden is from Saudi Arabia, I do not contend that Saudi Arabia has an emerging market economy. On the contrary, its oil wealth has served to reinforce its clientalist linkages, as patron sheiks spread their wealth in return for loyalty. With globalization and satellite television, however, patrons have reason to feel threatened by the perceived omnipresence of zerosum norms and Americanization, a fear that fuels resentment toward the West and, more specically, the presence of U.S. troops on Saudi soil. Support for al-Qaeda appears in tribal linkages in Saudi Arabia and Yemen, as well as in poor Muslim countries facing the social anarchy of development, such as Egypt, Indonesia, and Pakistan. 52. Ali A. Mazrui, Cultural Forces in World Politics (Portsmouth, N.H.: Heinemann, 1990).

Market Civilization and Its Clash with Terror 19

pathy. A paucity of empathy is necessary for doing harm to, and tolerating the suffering of, all out-group members. This is why international human rights are a concern promoted mostly by market democracies. It is also why widespread social support for both terrorism and sectarian violence frequently arises in developing countries but not in countries with deeply integrated markets.53 Clientalist values also lie at the core of the social approval of suicidal mass murder. From the market perspective, all behavior should have some immediate utility for the parties to a contract. It is thus difcult to comprehend the efcacy of suicide. But in cultures where the individual is less important than the group and the absence of science increases devotion to insular beliefs, suicideunder conditions of extreme socioeconomic disruptionmay emerge as a socially approved way of expressing ultimate loyalty to the in-group. In this way, cultural insularism, characterized by the absence of a market economy, is a necessary condition for the social approval of suicidal mass murder and sectarian violence. Cultural insularism combined with a particular grievancesuch as the negative consequences associated with globalizationcan create a deadly mix for Americans and other Westerners. Although latent anti-Americanism and antiWesternism exist throughout much of the developing world, these are most likely to rise to the surface during economic criseswhen nascent middle classes lose their status and turn against emerging liberal values. This is what is happening, for example, in Indonesia where the recent collapse of the local currency has eliminated the savings of the middle class, just as hyperination devastated the savings of Germanys middle class seventy-ve years ago. Recent terrorist acts against Indonesian Christians (as symbols of the West) and Westerners directly (the November 2002 bombing of a disco in Bali) are reminiscent of Germanys middle class turning against those it identied with market values, such as European Jews and the West. The West, in this sense, means market civilization.

53. The closest possible exceptions that I am aware of are the socially approved lynchings of African Americans by white Southerners in the 1920s and 1930s and the sectarian murders during the Troubles in Northern Ireland. In my view, however, the economies of neither the Southern states in the United States nor Northern Ireland were primarily integrated with contractsand this helps to explain the sectarian terror. Of course, this is an empirical issue that could be explored in future research.

Contending with Terrorism 20

The Eradication of Terror

Terrorism has both expressed and underlying causes. Expressed causes are those that terrorists assert themselves. Emic analysis, in which subjects are asked to explain why they behave as they do, identies expressed causes. It does not explain, however, why some acts iname passions while others go unnoticed. For instance, hundreds have died in recent violence between Hindus and Muslims in India, including many Muslims. Yet these killings have elicited an emotionally muted headline in the Arab media. When Israelis kill Muslims, however, as has occurred in the most recent round of Middle East violence, it inames the entire Muslim world.54 To understand these different responses, scholars must engage not only in emic but also in etic analysis: They need to be able to interpret the behavior of their subjects. Why do so many Indonesians, for instance, empathize with the plight of the Palestinians but seem to express little outrage over the deaths of Indian Muslims? The reason is Israels identication with the United States and emerging markets. Although identication with Islam may be an expressed cause of this rage, the underlying cause is not Islam but rather a deeply embedded antimarket and thus antiAmerican passiona fury that extends beyond the Islamic world and whose origins are not understood even by those espousing hatred for the West. This sense of rage against market civilization and its shared liberal values and beliefsa rage that can be inamed with the addition of any immediate causelies just beneath the surface in many developing countries. This is not to say that all or even the majority of people living in the developing world share this wrath, but that the potential for a clash is ever-present. Once policymakers understand this, they can begin to develop the kinds of political strategies needed to eliminate the terrorist threat. Since the terrorist attacks of September 11, three myths have emerged regarding the direction that these strategies should take; all three threaten to derail efforts to eradicate terror. The rst myth is that to win the hearts and minds of people around the world in the struggle against terror, the United States must do more to signal its friendly intentionsfor instance, by increasing economic aid and explaining U.S. policies more clearly.55 This view is mistaken.

54. Thomas L. Friedman, The Core of Muslim Rage, New York Times, March 6, 2002, p. A21. 55. Articles representing inuential American think tanks that take this view include Daalder and Lindsay, Nasty, Brutish, and Long; and Peter G. Peterson, Diplomacy and the War on Terrorism, Foreign Affairs, Vol. 81, No. 5 (SeptemberOctober 2002), pp. 7496.

Market Civilization and Its Clash with Terror 21

The rage against the United States as the leading symbol of the West is so deeply embedded in some societies that many will interpret whatever the United States does with malign intent. If the United States offers to increase economic aid, it is seen as imperialist; if it does not, it is neglectful. If the United States intervenes to protect Muslims, as it did in the 1999 Kosovo conict, critics will rail against U.S. imperialism (there must be oil there) or, at best, charge that the United States intentionally delayed the intervention because Americans really hate Muslims. Consider that in clientalist cultures the notion of scienceuniversal truthis incomprehensible.56 For this reason, even many educated people in the developing world believe in such nonsense as the notion that 4,000 Jews were warned not to go to work at the World Trade Center on September 11:57 These people believe what they want to believe, regardless of the evidence before them. In fact, for those enraged against the United States for its perceived zero-sum values, friendly acts will have no positive effect. The implication of this is liberating: In terms of underlying causes, the United States need not worry about how societies that produce or harbor terrorists perceive its actions in the war against terrorism. Given that whatever the United States (and other market democracies) do will be interpreted as malevolent, they may as well behave as they see t. The second myth associated with September 11 is that terror arises in the absence of democracy,58 and therefore the United States should push harder for democratic change in developing countries. This view is also awed. Stable democracies emerge when people want them to, when they share the liberal values and beliefs that prevail in market economies. It is understandable that scholars, policymakers, and pundits in market democracies value democracy and consider it a cure-all against evil: In market civilization, democratic institutions are a deeply embedded value. There is little evidence, however, that democracy causes liberal values. History shows that democracy without lib56. The notion of science should not be confused with the use of advanced technology or education. Science is a process of discovery that assumes that some facts are universal, anyone can challenge anothers assertions of fact, and truth is sought through the free competition of ideas. One can be taught the discoveries of advanced physics but still have no concept of challenging assertions of fact scientically. This is why communist and developing nations can import and modify advanced technology, but the market democracies will always be in the avant-garde of developing knowledge. 57. Thomas L. Friedman, Global Village Idiocy, New York Times, May 12, 2002, sec. 4, p. 15; see also Horan, Those Young Arab Muslims and Us, p. 54. 58. For example, Crenshaw, The Causes of Terrorism; Friedman, The Free-Speech Bind; Huntington, The Age of Muslim Wars; and Nicholas D. Kristof, What Is Democracy Anyway? New York Times, May 3, 2002, p. A23.

Contending with Terrorism 22

eral values results in illiberal democracy and the rise to power of antidemocratic regimes that frequently display antimarket and clientalistand therefore terroristorientation, such as the Bolsheviks in Russia and the Nazis in Germany. The third myth to emerge after September 11 is that if people who detest the United States only had greater exposure to American values, their hatred would dissipate.59 This view is premised on the assumption that because Americans know they are nice people, others will feel the same way if only they get to know them better. This view is also inaccurate. Anti-American rage is the result of people knowing Americans too well. The problem is that they just do not like what they see, because from the clientalist perspective, American values reect a degeneration of culture and the ascendance of zero-sum norms. Ironically, the notion that modern culture means no culture is also a common assumption of many academic models of global politics.60 As I have deduced from my analysis of the market economy, however, modern culture does possess valuesthe values of contractual exchange. The task for the United States in the struggle against terrorism then is not to expose more of itself but to counteract the ill effects of too much exposure by more subtly demonstrating the redeeming aspects of market culture. To win the war against terrorism, the United States and other market democracies must remove the underlying cause of terror: the deeply embedded antimarket rage brought on by the forces of globalization. To do this, the market democracies have only one option: to boost developing countries out of the mire of social anarchy and into market development. Most developing countries cannot make this transition alone, because their leaders are likely to hold clientalist rather than market values and beliefs. Furthermore, maintaining their grasp on power typically involves redistributing state resources among winning coalitions of clientalist in-groups. In this way, current forms of foreign aid may actually reinforce values and beliefs that condone terror, as recipient governments use the aid to pay off supporters and reinforce clientalist linkages. In fact, studies report that much foreign aid pays the salaries of bureaucrats and those working for aid agencies.61 Because these jobs are frequently

59. See, for instance, Friedman, Global Village Idiocy. 60. See, for instance, Bruce Bueno de Mesquita, James D. Morrow, Randolph M. Siverson, and Alastair Smith, An Institutional Explanation of the Democratic Peace, American Political Science Review, Vol. 93, No. 4 (December 1999), pp. 791807; and Waltz, Theory of International Politics. 61. See Dubious Aid, Canada and the World Backgrounder, Vol. 65, No. 6 (May 2000), p. 27.

Market Civilization and Its Clash with Terror 23

obtained through clientalist linkages, current forms of aid can actually promote the very clientalist values that can legitimate the resort to terror. Because governments of developing countries are unlikely to get out of the mire of social anarchy and into market development themselves, an outside power is needed to act as a sort of Leviathan: to push the governments of target countries to establish the prerequisites of a market economy. These include impartial enforcement of contracts and common law; destruction of clientalist linkages (corruption); subsidization of private enterprises (with fair bidding practices); widespread equitable subsidization of small loans so people can purchase homes or start small businesses; and redistribution to widen the scope of opportunities for market engagement. In the 1980s, Ronald Reagans administration encouraged cuts in the number of state-owned enterprises in a variety of countries. State ownership of enterprises is not the problem, however: The problem is when state ownership prevents an enterprise from competing fairly in the market. In recent years the International Monetary Fund has begun to enforce rules of equity in banking practices.62 For the most part, however, policymakers have placed greater emphasis on balancing budgets, supporting democratization,63 and reducing poverty.64 It is not deregulated markets, democracy, or an absence of poverty that produces liberal values, however, but rather a market economy.65 Thus, to reduce the social support of terror, market democracies should use economic aid as both a means and an incentive for governments in developing countries (1) to create and enforce bodies of common law that are vital to the functioning of a market economy, and (2) to equitably subsidize local private enterprises with the goal of widespread employment. The latter is critical during the transition period: The availability of living-wage jobs in the market alleviates insecurity and prevents antimarket rage. Given the deep distrust of U.S. motives among the millions living in the social anarchy of underdevelopment, other market democracies must share the burden of pulling them out of this mire. One option would be to create an in62. Ajit Singh, Aid, Conditionality, and Development, Development and Change, Vol. 33, No. 2 (2002), pp. 299300. 63. James K. Boyce, Unpacking Aid, Development and Change, Vol. 33, No. 2 (2002), p. 242. 64. Graham Bird, A Suitable Case for Treatment? Understanding the Ongoing Debate about the IMF, Third World Quarterly, Vol. 22, No. 5 (October 2001), pp. 823848. 65. A market economy is not a free market. A market economy is one in which the majority of people routinely engage in contractual exchange. Thus a market economy may be highly regulated (e.g., Sweden) and, in theory at least, be publicly owned. A free market, in contrast, refers to a deregulated or partially regulated economy that can coexist with underdevelpment (e.g., Kenya).

Contending with Terrorism 24

ternational organization with substantial powers to monitor compliance with aid conditions, run by the donor states yet unconstrained by their independent interests. In this way, multilateralism could legitimate the indirect external control of the economies of recipient states during their transitions. Multilateral action would also allow the United States to keep a lower prole and include its allies as partners in the war against terror. The historical record shows that market democracies easily cooperate and establish legal regimes among themselves.66 The European Union and the North Atlantic Treaty Organization are just two of the many regimes that bind these countries together. Although differences do occur, they are mostly at the level of tactics and not over major goals. Sharing preferences and bounded by the logic of contractual exchange, market democracies manage their relationships and resolve their disputes with other market democracies through a combination of mutual respect (equity), common law, and in the absence of law, negotiation and compromise. As German Chancellor Gerhard Schrder described his countrys recent rift with the United States, Between friends, there can be factual differences.67 Like West Germany after World War II, developing countries whose market economies are subsidized are likely to one day have market economies, at which time their newly emerged liberal values will reinforce their market-democratic institutions, and there will be no further need of foreign assistance. The social basis of terror against the United States and its allies could thus be eradicated.

Conclusion
Until now there have been two general approaches to understanding the motivations behind terrorism, one rational and the other cultural. Rational explanations focus on the role of political and economic grievances and assume that certain observable factors associated with poverty such as economic inequality, illiteracy, and lack of democracy cause terror. None of these approaches, however, has established a direct causal link between any of these factors and terror. Nor does there appear to be a correlation between poverty and terror. Cultural explanations, in contrast, focus not on political or economic conditions but on the notion that the values and beliefs of terrorists and their sup-

66. Mousseau, An Economic Limitation to the Zone of Democratic Peace and Cooperation. 67. Quoted in Steven Erlanger, Moves by Germany to Mend Relations Rebuffed by Bush, New York Times, September 24, 2002, p. A1.

Market Civilization and Its Clash with Terror 25

porters are vastly different from those of their targets. Typically, however, these explanations identify indigenous culture as the causal variable, which makes this approach unsuitable for predicting variation in social support for terror within cultures across both time and place. To grasp the origins of terror and why some support it, scholars need an approach that combines the rationalist identication of observable circumstances with the culturalist emphasis on the way people think. In this article I drew on several generations of research in anthropology, economics, political science, and sociology to show how the values and beliefs that support terrora lack of empathy for out-groups, an emphasis on community over the individual, and an incomprehension for objective truth and individual innocencearise from the clientalist economic linkages that are commonplace in many developing countries. In contrast, values that work against terrorindividualism, tolerance, equity, and the rule of common law arise with a market economy. Because all market economies in the contemporary period have been developed economies, there appears to be a link between underdevelopment and terror. As I have argued, however, the real culprit is social anarchy produced by globalization and the difculties attending the transition to a market economy. Just as millions in the last century turned to antimarket and sectarian values during the rise of market economies in Europe, today millions in the developing world support antimarket and sectarian values reected in support for ethnofacism, sectarian murder, and fundamentalist religionsanything that offers psychic comfort in the face of volatile social anarchy. It follows that there is a market civilization based on common liberal values and beliefs, and that this civilization is in conict with much of the developing world. Direct and expressed causes bring this conict to the surface at particular times and places. Beneath the surface, however, lies a deeply embedded clash of cultures: market civilization versus the rest. A number of scholars have noted signs of this conict but have typically identied indigenous culture, not the market economy, as exogenous.68 The problem with this view is that it
68. The most prominent argument for this view in recent years appears in Samuel P. Huntington, The Clash of Civilizations and the Remaking of World Order (New York: Simon and Schuster, 1996). For empirical challenges to this thesis, see Errol Anthony Henderson and Richard Tucker, Clear and Present Strangers: The Clash of Civilizations and International Conict, International Studies Quarterly, Vol. 45, No. 2 (June 2001), pp. 317338; and Bruce M. Russett, John R. Oneal, and Michaelene Cox, Clash of Civilizations, or Realism and Liberalism Dj Vu? Some Evidence, Journal of Peace Research, Vol. 37, No. 5 (September 2000), pp. 583609. For Huntingtons response to Russett, Oneal, and Cox, see Samuel P. Huntington, Try Again: A Reply to Russett, Oneal, and Cox, ibid., pp. 609610.

Contending with Terrorism 26

assumes that liberal values emerge from Western indigenous culture. Although this view is pervasive,69 it runs contrary to the historical record. For instance, a generation after Max Weber wrote about the virtues of the Protestant ethic,70 millions of Protestants in the West conspired to murder millions of Jews.71 The Nazis are just one prominent example of Western barbarism; white Southern Protestants in the United States who participated in lynchings in the early twentieth century are another. To many, it may seem as though liberal values are inherently Western, but this notion rests on a biased selection of the evidence. It ignores cases of Western barbarity. Empirical research across several disciplines demonstrates that it is market development that correlates with liberal values.72 Although this conclusion may be unsettling for many scholars of global politics, it offers a better accounting of global history. More important, it carries a liberating implication for progressive leaders in the developing world: The rise of markets and liberal culture will not make a developing country any more Western than the rise of a market economy in England made the British any more Dutch. Nevertheless, there is nothing in this thesis that argues against other possible sources of anti-Americanism and anti-Westernism. Realists and world systems theorists might focus on the projection of U.S. military power from the core into the periphery as the source of anti-Americanism; liberal institutionalists may focus on what they consider the unilateralist turn that U.S. foreign policy recently seems to have taken. These sources of anti-Americanism, however, exist primarily on the surface and are present mostly in the West and at universities. Antiglobalization protesters within market democracies, for instance, frequently express anti-American and antimarket sentiments. These protesters, however, call for greater global equalitya deeply embedded lib69. See, for example, Francis Fukuyama, The Primacy of Culture, Journal of Democracy, Vol. 6, No. 1 (January 1995), pp. 714; and Ronald Inglehart, The Renaissance of Political Culture, American Political Science Review, Vol. 82, No. 4 (December 1988), pp. 12031230. 70. Max Weber, The Protestant Ethic and the Spirit of Capitalism, trans. Talcott Parsons (New York: Charles Scribners Sons, 1958 [190405]). 71. See Goldhagen, Hitlers Willing Executioners. 72. For the observation that a rise in markets liberalizes values, see Braudel, Afterthoughts on Material Civilization and Capitalism. There is extensive cross-national data linking economic development and liberal values, with the overwhelming majority of observed cases of development being cases of market-oriented development. See Hofstede, Cultures Consequences; and Inglehart and Baker, Modernization, Cultural Change, and the Persistence of Traditional Values. For specic empirical conrmation that it is market developmentand not other kinds of developmentthat promotes liberal values, see Michael Mousseau, Market Culture and Peace among Nations: Its the Market Democracies That Ally, paper presented at the annual meeting of the American Political Science Association, Boston, Massachusetts, August 26September 1, 2002.

Market Civilization and Its Clash with Terror 27

eral preference. Although these protesters may also express resentment of the market, they do so for the same reason that many people in developing countries resent the market: They are typically young students with little direct experience in the marketplace. Regardless, such anti-Americanism appears only among a minority of the West, and few would suggest that antiglobalization protesters would support the mass murder of Americans and other Westernersa preference that requires a radically different set of values than those associated with market democracies. Once the rise of market civilization and its clash with the rest is understood, political strategies for winning the war against terror can be developed. Just as the United States imposed and subsidized the emergence of market economies in Germany and Japan after World War IIeffectively liberalizing their culturesmarket democracies today must subsidize the rise of markets in developing countries. This does not mean deregulating their economies, which would do little to inhibit clientalist linkages or encourage trust in contractual exchange. Nor is wealth the source of liberal values: Saudi Arabia is one of the worlds wealthiest states, but it has a predominantly clientalist economy, which is why it produces terrorists. Rather, it is through the establishment of market economies that the United States and its allies can be made safe from terror.

Behind the Curve

Behind the Curve

Globalization and International Terrorism

Audrey Kurth Cronin

he coincidence between the evolving changes of globalization, the inherent weaknesses of the Arab region, and the inadequate American response to both ensures that terrorism will continue to be the most serious threat to U.S. and Western interests in the twenty-rst century. There has been little creative thinking, however, about how to confront the growing terrorist backlash that has been unleashed. Terrorism is a complicated, eclectic phenomenon, requiring a sophisticated strategy oriented toward inuencing its means and ends over the long term. Few members of the U.S. policymaking and academic communities, however, have the political capital, intellectual background, or inclination to work together to forge an effective, sustained response. Instead, the tendency has been to fall back on established bureaucratic mind-sets and prevailing theoretical paradigms that have little relevance for the changes in international security that became obvious after the terrorist attacks in New York and Washington on September 11, 2001. The current wave of international terrorism, characterized by unpredictable and unprecedented threats from nonstate actors, not only is a reaction to globalization but is facilitated by it; the U.S. response to this reality has been reactive and anachronistic. The combined focus of the United States on statecentric threats and its attempt to cast twenty-rst-century terrorism into familiar strategic terms avoids and often undermines effective responses to this nonstate phenomenon. The increasing threat of globalized terrorism must be met with exible, multifaceted responses that deliberately and effectively exploit avenues of globalization in return; this, however, is not happening.
Audrey Kurth Cronin is Specialist in International Terrorism at the Congressional Research Service at the Library of Congress. The article was written when she was Visiting Associate Professor at the Edmund A. Walsh School of Foreign Service and a Research Fellow at the Center for Peace and Security Studies, Georgetown University. I am grateful for helpful comments and criticisms on previous drafts from Robert Art, Patrick Cronin, Timothy Hoyt, James Ludes, and an anonymous reviewer. I have been greatly inuenced by conversations and other communications with Martha Crenshaw, to whom I owe a huge debt. None of these people necessarily agrees with everything here. Also benecial was a research grant from the School of Foreign Service at Georgetown University. My thanks to research assistants Christopher Connell, William Josiger, and Sara Skahill and to the members of my graduate courses on political violence and terrorism. Portions of this article will be published as Transnational Terrorism and Security: The Terrorist Threat to Globalization, in Michael E. Brown, ed., Grave New World: Global Dangers in the Twenty-rst Century (Washington, D.C.: Georgetown University Press, forthcoming).
International Security, Vol. 27, No. 3 (Winter 2002/03), pp. 3058 2003 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

28

Behind the Curve 29

As the primary terrorist target, the United Sates should take the lead in fashioning a forward-looking strategy. As the worlds predominant military, economic, and political power, it has been able to pursue its interests throughout the globe with unprecedented freedom since the breakup of the Soviet Union more than a decade ago. Even in the wake of the September 11 terrorist attacks on the World Trade Center and the Pentagon, and especially after the U.S. military action in Afghanistan, the threat of terrorism, mostly consisting of underfunded and ad hoc cells motivated by radical fringe ideas, has seemed unimportant by comparison. U.S. strategic culture has a long tradition of downplaying such atypical concerns in favor of a focus on more conventional state-based military power.1 On the whole, this has been an effective approach: As was dramatically demonstrated in Afghanistan, the U.S. military knows how to destroy state governments and their armed forces, and the American political leadership and public have a natural bias toward using power to achieve the quickest results. Sometimes it is important to show resolve and respond forcefully. The United States has been far less impressive, however, in its use of more subtle tools of domestic and international statecraft, such as intelligence, law enforcement, economic sanctions, educational training, nancial controls, public diplomacy, coalition building, international law, and foreign aid. In an ironic twist, it is these tools that have become central to the security of the United States and its allies since September 11. In an era of globalized terrorism, the familiar state-centric threats have not disappeared; instead they have been joined by new (or newly threatening) competing political, ideological, economic, and cultural concerns that are only supercially understood, particularly in the West. An examination of the recent evolution of terrorism and a projection of future developments suggest that, in the age of globalized terrorism, old attitudes are not just anachronistic; they are dangerous. Terrorism as a phenomenon is not new, but for reasons explained below, the threat it now poses is greater than ever before. The current terrorist backlash is manifested in the extremely violent asymmetrical response directed at the United States and other leading powers by terrorist groups associated with or inspired by al-Qaeda. This backlash has the potential to fundamentally threaten the international system. Thus it is not just an American problem. Unless the United States and its allies formulate a more comprehensive re1. The issue of U.S. strategic culture and its importance in the response to international terrorism is explored in more depth in Audrey Kurth Cronin, Rethinking Sovereignty: American Strategy in the Age of Terror, Survival, Vol. 44, No. 2 (Summer 2002), pp. 119139.

Contending with Terrorism 30

sponse to terrorism, better balanced across the range of policy instruments, the results will be increasing international instability and long-term failure. The article proceeds in ve main sections. First, it provides a discussion of the denition, history, causes, and types of terrorism, placing the events of September 11, 2001, in their modern context. Second, it briey describes key trends in modern terrorism, explaining how the phenomenon appears to be evolving. Third, it analyzes the implications of these trends for the stability and security of the international community generally, and the United States and its allies more specically. Fourth, the article outlines the prospects of these trends. It concludes with a range of policy recommendations suggested by the analysis.

Denition, Origins, Motivations, and Types of Modern Terrorism

The terrorist phenomenon has a long and varied history, punctuated by lively debates over the meaning of the term. By ignoring this history, the United States runs the risk of repeating the plethora of mistakes made by other major powers that faced similar threats in the past. This section begins with an explanation of the denition of terrorism, then proceeds to an examination of terrorisms origins, major motivations, and predominant types. definition of terrorism Terrorism is notoriously difcult to dene, in part because the term has evolved and in part because it is associated with an activity that is designed to be subjective. Generally speaking, the targets of a terrorist episode are not the victims who are killed or maimed in the attack, but rather the governments, publics, or constituents among whom the terrorists hope to engender a reactionsuch as fear, repulsion, intimidation, overreaction, or radicalization. Specialists in the area of terrorism studies have devoted hundreds of pages toward trying to develop an unassailable denition of the term, only to realize the fruitlessness of their efforts: Terrorism is intended to be a matter of perception and is thus seen differently by different observers.2 Although individuals can disagree over whether particular actions constitute terrorism, there are certain aspects of the concept that are fundamental.
2. On the difculty of dening terrorism, see, for example, Omar Malik, Enough of the Denition of Terrorism! Royal Institute of International Affairs (London: RIIA, 2001); and Alex P. Schmid, Political Terrorism: A Research Guide (New Brunswick, N.J.: Transaction Books, 1984). Schmid spends more than 100 pages grappling with the question of a denition, only to conclude that none is universally accepted.

Behind the Curve 31

First, terrorism always has a political nature. It involves the commission of outrageous acts designed to precipitate political change.3 At its root, terrorism is about justice, or at least someones perception of it, whether man-made or divine. Second, although many other uses of violence are inherently political, including conventional war among states, terrorism is distinguished by its nonstate charactereven when terrorists receive military, political, economic, and other means of support from state sources. States obviously employ force for political ends: When state force is used internationally, it is considered an act of war; when it is used domestically, it is called various things, including law enforcement, state terror, oppression, or civil war. Although states can terrorize, they cannot by denition be terrorists. Third, terrorism deliberately targets the innocent, which also distinguishes it from state uses of force that inadvertently kill innocent bystanders. In any given example, the latter may or may not be seen as justied; but again, this use of force is different from terrorism. Hence the fact that precision-guided missiles sometimes go astray and kill innocent civilians is a tragic use of force, but it is not terrorism. Finally, state use of force is subject to international norms and conventions that may be invoked or at least consulted; terrorists do not abide by international laws or norms and, to maximize the psychological effect of an attack, their activities have a deliberately unpredictable quality.4 Thus, at a minimum, terrorism has the following characteristics: a fundamentally political nature, the surprise use of violence against seemingly random targets, and the targeting of the innocent by nonstate actors.5 All of these attributes are illustrated by recent examples of terrorismfrom the April 2000 kidnapping of tourists by the Abu Sayyaf group of the Philippines to the various incidents allegedly committed by al-Qaeda, including the 1998 bombings of the U.S. embassies in Kenya and Tanzania and the September 11 attacks. For the purposes of this discussion, the shorthand (and admittedly imperfect) denition of terrorism is the threat or use of seemingly random violence against innocents for political ends by a nonstate actor.
3. Saying that terrorism is a political act is not the same as arguing that the political ends toward which it is directed are necessarily negotiable. If violent acts do not have a political aim, then they are by denition criminal acts. 4. The diabolical nature of terrorism has given resonance to Robert Kaplans view that the world is a grim landscape littered with evildoers and requiring Western leaders to adopt a pagan ethos. But such conclusions deserve more scrutiny than space allows here. See Steven Mufson, The Way Bush Sees the World, Washington Post, Outlook section, February 17, 2002, p. B1. 5. R.G. Frey and Christopher W. Morris, Violence, Terrorism, and Justice, in Frey and Morris, eds., Violence, Terrorism, and Justice (Cambridge: Cambridge University Press, 1991), p. 3.

Contending with Terrorism 32

origins of terrorism Terrorism is as old as human history. One of the rst reliably documented instances of terrorism, however, occurred in the rst century b.c.e. The ZealotsSicarri, Jewish terrorists dedicated to inciting a revolt against Roman rule in Judea, murdered their victims with daggers in broad daylight in the heart of Jerusalem, eventually creating such anxiety among the population that they generated a mass insurrection.6 Other early terrorists include the Hindu Thugs and the Muslim Assassins. Modern terrorism, however, is generally considered to have originated with the French Revolution.7 The term terror was rst employed in 1795, when it was coined to refer to a policy systemically used to protect the edgling French republic government against counterrevolutionaries. Robespierres practice of using revolutionary tribunals as a means of publicizing a prisoners fate for broader effect within the population (apart from questions of legal guilt or innocence) can be seen as a nascent example of the much more highly developed, blatant manipulation of media attention by terrorist groups in the mid- to late twentieth century.8 Modern terrorism is a dynamic concept, from the outset dependent to some degree on the political and historical context within which it has been employed. decolonization and antiglobalization: drivers of terrorism? Although individual terrorist groups have unique characteristics and arise in specic local contexts, an examination of broad historical patterns reveals that the international system within which such groups are spawned does inuence their nature and motivations. A distinguishing feature of modern terrorism has been the connection between sweeping political or ideological concepts and increasing levels of terrorist activity internationally. The broad political aim has been against (1) empires, (2) colonial powers, and (3) the U.S.led international system marked by globalization. Thus it is important to understand the general history of modern terrorism and where the current threat ts within an international context.
6. Walter Laqueur, Terrorism (London: Weidenfeld and Nicolson, 1977, reprinted in 1978), pp. 78; and David C. Rapoport, Fear and Trembling: Terrorism in Three Religious Traditions, American Political Science Review, Vol. 78, No. 3 (September 1984), pp. 658677. 7. David C. Rapoport, The Fourth Wave: September 11 in the History of Terrorism, Current History, December 2001, pp. 419424; and David C. Rapoport, Terrorism, Encyclopedia of Violence, Peace, and Conict (New York: Academic Press, 1999). 8. Ironically, Robespierres tactics during the Reign of Terror would not be included in this articles denition of terrorism, because it was state terror.

Behind the Curve 33

David Rapoport has described modern terrorism such as that perpetuated by al-Qaeda as part of a religiously inspired fourth wave. This wave follows three earlier historical phases in which terrorism was tied to the breakup of empires, decolonization, and leftist anti-Westernism.9 Rapoport argues that terrorism occurs in consecutive if somewhat overlapping waves. The argument here, however, is that modern terrorism has been a power struggle along a continuum: central power versus local power, big power versus small power, modern power versus traditional power. The key variable is a widespread perception of opportunity, combined with a shift in a particular political or ideological paradigm. Thus, even though the newest international terrorist threat, emanating largely from Muslim countries, has more than a modicum of religious inspiration, it is more accurate to see it as part of a larger phenomenon of antiglobalization and tension between the have and have-not nations, as well as between the elite and underprivileged within those nations. In an era where reforms occur at a pace much slower than is desired, terrorists today, like those before them, aim to exploit the frustrations of the common people (especially in the Arab world). In the nineteenth century, the unleashing of concepts such as universal suffrage and popular empowerment raised the hopes of people throughout the western world, indirectly resulting in the rst phase of modern terrorism. Originating in Russia, as Rapoport argues, it was stimulated not by state repression but by the efforts of the czars to placate demands for economic and political reforms, and the inevitable disappointment of popular expectations that were raised as a result. The goal of terrorists was to engage in attacks on symbolic targets to get the attention of the common people and thus provoke a popular response that would ultimately overturn the prevailing political order. This type of modern terrorism was reected in the activities of groups such as the Russian Narodnaya Volya (Peoples Will) and later in the development of a series of movements in the United States and Europe, especially in territories of the former Ottoman Empire. The dissolution of empires and the search for a new distribution of political power provided an opportunity for terrorism in the nineteenth and twentieth centuries. It climaxed in the assassination of Archduke Franz Ferdinand on June 28, 1914, an event that catalyzed the major powers into taking violent action, not because of the signicance of the man himself but because of the sus-

9. Rapoport, The Fourth Wave.

Contending with Terrorism 34

picion of rival state involvement in the sponsorship of the killing. World War I, the convulsive systemic cataclysm that resulted, ended the rst era of modern terrorism, according to Rapoport.10 But terrorism tied to popular movements seeking greater democratic representation and political power from coercive empires has not ceased. Consider, for example, the Balkans after the downfall of the former state of Yugoslavia. The struggle for power among various Balkan ethnic groups can be seen as the nal devolution of power from the former Ottoman Empire. This postimperial scramble is also in evidence elsewhere for example, in Aceh, Chechnya, and Xinjiang, to mention just a few of the trouble spots within vast (former) empires. The presentation of a target of opportunity, such as a liberalizing state or regime, frequently evokes outrageous terrorist acts. According to Rapoport, a second, related phase of modern terrorism associated with the concept of national self-determination developed its greatest predominance after World War I. It also continues to the present day. These struggles for power are another facet of terrorism against larger political powers and are specically designed to win political independence or autonomy. The mid-twentieth-century era of rapid decolonization spawned national movements in territories as diverse as Algeria, Israel, South Africa, and Vietnam.11 An important by-product was ambivalence toward the phenomenon in the international community, with haggling over the denition of terrorism reaching a fever pitch in the United Nations by the 1970s. The question of political motivation became important in determining international attitudes toward terrorist attacks, as the postWorld War II backlash against the colonial powers and the attractiveness of national independence movements led to the creation of a plethora of new states often born from violence. Arguments over the justice of international causes and the designation of terrorist struggles as wars of national liberation predominated, with consequentialist philosophies excusing the killing of innocent people if the cause in the long run was just. Rapoport sees the U.S. intervention in Vietnam, and especially the subsequent American defeat by the Vietcong, as having catalyzed a third wave of modern terrorism; however, the relationship between the Vietnam conict and other decolonization movements might just as easily be considered part of the same phase. In any case, the victory of the
10. Ibid., pp. 419420. 11. Ibid., p. 420.

Behind the Curve 35

Vietcong excited the imaginations of revolutionaries throughout the world and, according to Rapoport, helped lead to a resurgence in terrorist violence. The Soviet Union underwrote the nationalist and leftist terrorist agendas of some groups, depicting the United States as the new colonial poweran easy task following the Vietnam interventionand furthering an ideological agenda oriented toward achieving a postcapitalist, international communist utopia. Other groups, especially in Western Europe, rejected both the Soviet and capitalist models and looked admiringly toward nationalist revolutionaries in the developing world.12 Leftist groups no longer predominate, but the enduring search for national self-determination continues, not only in the areas mentioned above but also in other hot spots such as the Basque region, East Timor, Sri Lanka, and Sudan. Terrorism achieved a rmly international character during the 1970s and 1980s,13 evolving in part as a result of technological advances and partly in reaction to the dramatic explosion of international media inuence. International links were not new, but their centrality was. Individual, scattered national causes began to develop into international organizations with links and activities increasingly across borders and among differing causes. This development was greatly facilitated by the covert sponsorship of states such as Iran, Libya, and North Korea, and of course the Soviet Union, which found the underwriting of terrorist organizations an attractive tool for accomplishing clandestine goals while avoiding potential retaliation for the terrorist attacks. The 1970s and 1980s represented the height of state-sponsored terrorism. Sometimes the lowest common denominator among the groups was the concept against which they were reactingfor example, Western imperialismrather than the specic goals they sought. The most important innovation, however, was the increasing commonality of international connections among the groups. After the 1972 Munich Olympics massacre of eleven Israeli athletes, for example, the Palestinian Liberation Organization (PLO) and its associated groups captured the imaginations of young radicals around the world. In Lebanon and elsewhere, the PLO also provided training in the pre12. Adrian Gulke, The Age of Terrorism and the International Political System (London: I.B. Tauris, 1995), pp. 5663. 13. This is not to imply that terrorism lacked international links before the 1970s. There were important international ties between anarchist groups of the late nineteenth century, for example. See David C. Rapoport, The Four Waves of Modern Terrorism, in Audrey Kurth Cronin and James Ludes, eds., The Campaign against International Terrorism (Washington, D.C.: Georgetown University Press, forthcoming).

Contending with Terrorism 36

ferred techniques of twentieth-century terrorism such as airline hijacking, hostage taking, and bombing. Since the September 11 attacks, the world has witnessed the maturation of a new phase of terrorist activity, the jihad era, spawned by the Iranian Revolution of 1979 as well as the Soviet defeat in Afghanistan shortly thereafter. The powerful attraction of religious and spiritual movements has overshadowed the nationalist or leftist revolutionary ethos of earlier terrorist phases (though many of those struggles continue), and it has become the central characteristic of a growing international trend. It is perhaps ironic that, as Rapoport observes, the forces of history seem to be driving international terrorism back to a much earlier time, with echoes of the behavior of sacred terrorists such as the Zealots-Sicarii clearly apparent in the terrorist activities of organizations such as al-Qaeda and its associated groups. Religious terrorism is not new; rather it is a continuation of an ongoing modern power struggle between those with power and those without it. Internationally, the main targets of these terrorists are the United States and the U.S.-led global system. Like other eras of modern terrorism, this latest phase has deep roots. And given the historical patterns, it is likely to last at least a generation, if not longer. The jihad era is animated by widespread alienation combined with elements of religious identity and doctrinea dangerous mix of forces that resonate deep in the human psyche. What is different about this phase is the urgent requirement for solutions that deal both with the religious fanatics who are the terrorists and the far more politically motivated states, entities, and people who would support them because they feel powerless and left behind in a globalizing world. Thus if there is a trend in terrorism, it is the existence of a two-level challenge: the hyperreligious motivation of small groups of terrorists and the much broader enabling environment of bad governance, nonexistent social services, and poverty that punctuates much of the developing world. Al-Qaeda, a band driven by religious extremism, is able to do so much harm because of the secondary support and sanctuary it receives in vast areas that have not experienced the political and economic benets of globalization. Therefore, the prescription for dealing with Osama bin Laden and his followers is not just eradicating a relatively small number of terrorists, but also changing the conditions that allow them to acquire so much power. Leaving aside for the moment the enabling environment, it is useful to focus on the chief motivations of the terrorists themselves, especially the contrasting secular and spiritual motivations of terrorism.

Behind the Curve 37

leftist, rightist, ethnonationalist/separatist, and sacred terrorism There are four types of terrorist organizations currently operating aound the world, categorized mainly by their source of motivation: left-wing terrorists, right-wing terrorists, ethnonationalist/separatist terrorists, and religious or sacred terrorists. All four types have enjoyed periods of relative prominence in the modern era, with left-wing terrorism intertwined with the Communist movement,14 right-wing terrorism drawing its inspiration from Fascism,15 and the bulk of ethnonationalist/separatist terrorism accompanying the wave of decolonization especially in the immediate postWorld War II years. Currently, sacred terrorism is becoming more signicant.16 Although groups in all categories continue to exist today, left-wing and right-wing terrorist groups were more numerous in earlier decades. Of course, these categories are not perfect, as many groups have a mix of motivating ideologiessome ethnonationalist groups, for example, have religious characteristics or agendas17but usually one ideology or motivation dominates. Categories are useful not simply because classifying the groups gives scholars a more orderly eld to study (admittedly an advantage), but also because different motivations have sometimes led to differing styles and modes of behavior. Understanding the type of terrorist group involved can provide insight into the likeliest manifestations of its violence and the most typical patterns of its development. At the risk of generalizing, left-wing terrorist organizations, driven by liberal or idealist political concepts, tend to prefer revolutionary, antiauthoritarian, antimaterialistic agendas. (Here it is useful to distinguish between the idealism of individual terrorists and the frequently contradictory motivations of their sponsors.) In line with these preferences, left-wing organizations often engage in brutal criminal-type behavior such as kidnapping, murder, bombing, and arson, often directed at elite targets that symbolize authority. They have difculty, however, agreeing on their long-term
14. Groups such as the Second of June Movement, the Baader-Meinhof Gang, the Red Brigades, the Weathermen, and the Symbionese Liberation Army belong in this category. 15. Among right-wing groups would be other neo-Nazi organizations (in the United States and Europe) and some members of American militia movements such as the Christian Patriots and the Ku Klux Klan. 16. The list here would be extremely long, including groups as different as the Tamil Tigers of Sri Lanka, the Basque separatist party, the PLO, and the Irish Republican Army (IRA) and its various splinter groups. 17. Bruce Hoffman notes that secular terrorist groups that have a strong religious element include the Provisional IRA, Armenian factions, and perhaps the PLO; however, the political/separatist aspect is the predominant characteristic of these groups. Hoffman, Terrorist Targeting: Tactics, Trends, and Potentialities, Technology and Terrorism (London: Frank Cass, 1993), p. 25.

Contending with Terrorism 38

objectives.18 Most left-wing organizations in twentieth-century Western Europe, for example, were brutal but relatively ephemeral. Of course, right-wing terrorists can be ruthless, but in their most recent manifestations they have tended to be less cohesive and more impetuous in their violence than leftist terrorist groups. Their targets are often chosen according to race but also ethnicity, religion, or immigrant status, and in recent decades at least, have been more opportunistic than calculated.19 This makes them potentially explosive but difcult to track.20 Ethnonationalist/separatist terrorists are the most conventional, usually having a clear political or territorial aim that is rational and potentially negotiable, if not always justiable in any given case. They can be astoundingly violent, over lengthy periods. At the same time, it can be difcult to distinguish between goals based on ethnic identity and those rooted in the control of a piece of land. With their focus on gains to be made in the traditional state-oriented international system, ethnonationalist/separatist terrorists often transition in and out of more traditional paramilitary structures, depending on how the cause is going. In addition, they typically have sources of support among the local populace of the same ethnicity with whom their separatist goals (or appeals to blood links) may resonate. That broader popular support is usually the key to the greater average longevity of ethnonationalist/ separatist groups in the modern era.21
18. An interesting example is Frances Action Directe, which revised its raison dtre several times, often altering it to reect domestic issues in Franceanarchism and Maoism, dissatisfaction with NATO and the Americanization of Europe, and general anticapitalism. See Michael Dartnell, Frances Action Directe: Terrorists in Search of a Revolution, Terrorism and Political Violence, Vol. 2, No. 4 (Winter 1990), pp. 457488. 19. For example, in the 1990s Germany and several other European countries experienced a rash of random arson attacks against guest houses and ofces that provided services to immigrants, many of whom were Middle Eastern in origin. Other examples include the violence associated with groups such as Europes football hooligans. A possible American example of the opportunistic nature of right-wing terrorism may be the anthrax letter campaign conducted in October 2001. See Susan Schmidt, Anthrax Letter Suspect Proled: FBI Says Author Likely Is Male Loner; Ties to Bin Laden Are Doubted, Washington Post, November 11, 2001, p. A1; and Steve Fainaru, Ofcials Continue to Doubt Hijackers Link to Anthrax: Fla. Doctor Says He Treated One for Skin Form of Disease, Washington Post, March 24, 2002, p. A23. 20. It is interesting to note that, according to Christopher C. Harmon, in Germany, 1991 was the rst year that the number of indigenous rightist radicals exceeded that of leftists. Harmon, Terrorism Today (London: Frank Cass, 2000), p. 3. 21. For example, in discussing the longevity of terrorist groups, Martha Crenshaw notes only three signicant terrorist groups with ethnonationalist ideologies that ceased to exist within ten years of their formation (one of these, EOKA, disbanded because its goalthe liberation of Cypruswas attained). By contrast, a majority of the terrorist groups she lists as having existed for ten years or longer have recognizable ethnonationalist ideologies, including the IRA (in its many

Behind the Curve 39

All four types of terrorist organizations are capable of egregious acts of barbarism. But religious terrorists may be especially dangerous to international security for at least ve reasons. First, religious terrorists often feel engaged in a Manichaean struggle of good against evil, implying an open-ended set of human targets: Anyone who is not a member of their religion or religious sect may be evil and thus fair game. Although indiscriminate attacks are not unique to religious terrorists, the exclusivity of their faith may lead them to dehumanize their victims even more than most terrorist groups do, because they consider nonmembers to be indels or apostatesas perhaps, for instance, al-Qaeda operatives may have viewed Muslims killed in the World Trade Center. Second, religious terrorists engage in violent behavior directly or indirectly to please the perceived commands of a deity. This has a number of worrisome implications: The whims of the deity may be less than obvious to those who are not members of the religion, so the actions of violent religious organizations can be especially unpredictable. Moreover, religious terrorists may not be as constrained in their behavior by concerns about the reactions of their human constituents. (Their audience lies elsewhere.) Third, religious terrorists consider themselves to be unconstrained by secular values or laws. Indeed the very target of the attacks may be the law-based secular society that is embodied in most modern states. The driving motivation, therefore, is to overturn the current post-Westphalian state systema much more fundamental threat than is, say, ethnonationalist terrorism purporting to carve out a new secular state or autonomous territory. Fourth, and related, religious terrorists often display a complete sense of alienation from the existing social system. They are not trying to correct the system, making it more just, more perfect, and more egalitarian. Rather they are trying to replace it. In some groups, apocalyptic images of destruction are seen as a necessityeven a purifying regimenand this makes them uniquely dangerous, as was painfully learned on September 11.22
forms), Sikh separatist groups, Euskadi Ta Askatasuna, the various Palestinian nationalist groups, and the Corsican National Liberation Front. See Crenshaw, How Terrorism Declines, Terrorism and Political Violence, Vol. 3, No. 1 (Spring 1991), pp. 6987. 22. On the characteristics of modern religious terrorist groups, see Bruce Hoffman, Inside Terrorism (New York: Columbia University Press, 1998), especially pp. 9495; and Bruce Hoffman, Terrorism Trends and Prospects, in Ian O. Lesser, Bruce Hoffman, John Arguilla, Michelle Zanini, and David Ronfeldt, eds., Countering the New Terrorism (Santa Monica, Calif.: RAND, 1999), especially pp. 1920. On the peculiar twists of one apocalyptic vision, see Robert Jay Lifton, Destroying the

Contending with Terrorism 40

Fifth, religious terrorism is especially worrisome because of its dispersed popular support in civil society. On the one hand, for example, groups such as al-Qaeda are able to nd support from some Muslim nongovernmental foundations throughout the world,23 making it truly a global network. On the other hand, in the process of trying to distinguish between the relatively few providers of serious support from the majority of genuinely philanthropic groups, there is the real risk of igniting the very holy war that the terrorists may be seeking in the rst instance. In sum, there are both enduring and new aspects to modern terrorism. The enduring features center on the common political struggles that have characterized major acts of international terrorism. The newest and perhaps most alarming aspect is the increasingly religious nature of modern terrorist groups. Against this historical background, the unique elements in the patterns of terrorist activity surrounding September 11 appear starkly.

Key Trends in Modern Terrorism

By the late 1990s, four trends in modern terrorism were becoming apparent: an increase in the incidence of religiously motivated attacks, a decrease in the overall number of attacks, an increase in the lethality per attack, and the growing targeting of Americans. Statistics show that, even before the September 11 attacks, religiously motivated terrorist organizations were becoming more common. The acceleration of this trend has been dramatic: According to the RANDSt. Andrews University Chronology of International Terrorism,24 in 1968 none of the identied international terrorist organizations could be classied as religious; in 1980, in the aftermath of the Iranian Revolution, there were 2 (out of 64), and that number had expanded to 25 (out of 58) by 1995.25

World to Save It: Aum Shinrikyo, Apocalyptic Violence, and the New Global Terrorism (New York: Henry Holt, 1999). 23. There is a long list of people and organizations sanctioned under Executive Order 13224, signed on September 23, 2001. Designated charitable organizations include the Benevolence International Foundation and the Global Relief Foundation. The list is available at http:/ / www.treas.gov/ofces/enforcement/ofac/sanctions/t11ter.pdf (accessed November 26, 2002). 24. The RANDSt. Andrews University Chronology of International Terrorism is a databank of terrorist incidents that begins in 1968 and has been maintained since 1972 at St. Andrews University, Scotland, and the RAND Corporation, Santa Monica, California. 25. Hoffman, Inside Terrorism, pp. 9091; and Nadine Gurr and Benjamin Cole, The New Face of Terrorism: Threats from Weapons of Mass Destruction (London: I.B. Tauris, 2000), pp. 2829.

Behind the Curve 41

Careful analysis of terrorism data compiled by the U.S. Department of State reveals other important trends regarding the frequency and lethality of terrorist attacks. The good news was that there were fewer such attacks in the 1990s than in the 1980s: Internationally, the number of terrorist attacks in the 1990s averaged 382 per year, whereas in the 1980s the number per year averaged 543.26 But even before September 11, the absolute number of casualties of international terrorism had increased, from a low of 344 in 1991 to a high of 6,693 in 1998.27 The jump in deaths and injuries can be partly explained by a few highprole incidents, including the bombing of the U.S. embassies in Nairobi and Dar-es-Salaam in 1998;28 but it is signicant that more people became victims of terrorism as the decade proceeded. More worrisome, the number of people killed per incident rose signicantly, from 102 killed in 565 incidents in 1991 to 741 killed in 274 incidents in 1998.29 Thus, even though the number of terrorist attacks declined in the 1990s, the number of people killed in each one increased. Another important trend relates to terrorist attacks involving U.S. targets. The number of such attacks increased in the 1990s, from a low of 66 in 1994 to a high of 200 in the year 2000.30 This is a long-established problem: U.S. nationals consistently have been the most targeted since 1968.31 But the percentage of international attacks against U.S. targets or U.S. citizens rose dramatically over the 1990s, from about 20 percent in 199395 to almost 50 percent in 2000.32 This is perhaps a consequence of the increased role and prole of the United States in the world, but the degree of increase is nonetheless troubling. The increasing lethality of terrorist attacks was already being noticed in the late 1990s, with many terrorism experts arguing that the tendency toward more casualties per incident had important implications. First it meant that, as had been feared, religious or sacred terrorism was apparently more dangerous than the types of terrorism that had predominated earlier in the twentieth
26. Statistics compiled from data in U.S. Department of State, Patterns of Global Terrorism, published annually by the Ofce of the Coordinator for Counterterrorism, U.S. Department of State. 27. Ibid. For a graphical depiction of this information, created on the basis of annual data from Patterns of Global Terrorism, see Cronin, Rethinking Sovereignty, p. 126. 28. In the 1998 embassy bombings alone, for example, 224 people were killed (with 12 Americans among them), and 4,574 were injured (including 15 Americans). U.S. Department of State, Patterns of Global Terrorism, 1998. 29. Ibid. For a graphical depiction of deaths per incident, created on the basis of annual data from Patterns of Global Terrorism, see Cronin, Rethinking Sovereignty, p. 128. 30. Ibid. 31. Hoffman, Terrorist Targeting, p. 24. 32. U.S. Department of State, Patterns of Global Terrorism, various years.

Contending with Terrorism 42

century. The world was facing the resurgence of a far more malignant type of terrorism, whose lethality was borne out in the larger death toll from incidents that increasingly involved a religious motivation.33 Second, with an apparent premium now apparently placed on causing more casualties per incident, the incentives for terrorist organizations to use chemical, biological, nuclear, or radiological (CBNR) weapons would multiply. The breakup of the Soviet Union and the resulting increased availability of Soviet chemical, biological, and nuclear weapons caused experts to argue that terrorist groups, seeking more dramatic and deadly results, would be more drawn to these weapons.34 The 1995 sarin gas attack by the Japanese cult Aum Shinrikyo in the Tokyo subway system seemed to conrm that worry. More recently, an examination of evidence taken from Afghanistan and Pakistan reveals al-Qaedas interest in chemical, biological, and nuclear weapons.35 In addition to the evolving motivation and character of terrorist attacks, there has been a notable dispersal in the geography of terrorist actsa trend that is likely to continue. Although the Middle East continues to be the locus of most terrorist activity, Central and South Asia, the Balkans, and the Transcaucasus have been growing in signicance over the past decade. International connections themselves are not new: International terrorist organizations inspired by common revolutionary principles date to the early nineteenth century; clandestine state use of foreign terrorist organizations occurred as early as the 1920s (e.g., the Mussolini government in Italy aided the Croat Ustasha); and complex mazes of funding, arms, and other state support for international terrorist organizations were in place especially in the 1970s and 1980s.36 During the Cold War, terrorism was seen as a form of surrogate warfare and
33. Examples include Bruce Hoffman, Holy Terror: The Implications of Terrorism Motivated by a Religious Imperative, RAND Paper P-7834 (Santa Monica, Calif.: RAND, 1993); and Mark Juergensmeyer, Terror Mandated by God, Terrorism and Political Violence, Vol. 9, No. 2 (Summer 1997), pp. 1623. 34. See, for example, Steven Simon and Daniel Benjamin, America and the New Terrorism, Survival, Vol. 42, No. 1 (Spring 2000), pp. 5975, as well as the responses in the subsequent issue, America and the New Terrorism: An Exchange, Survival, Vol. 42, No. 2 (Summer 2000), pp. 156 172; and Hoffman, Terrorism Trends and Prospects, pp. 738. 35. See Peter Finn and Sarah Delaney, Al-Qaedas Tracks Deepen in Europe, Washington Post, October 22, 2001, p. A1; Kamran Khan and Molly Moore, 2 Nuclear Experts Briefed Bin Laden, Pakistanis Say, Washington Post, December, 12, 2001, p. A1; James Risen and Judith Miller, A Nation Challenged: Chemical WeaponsAl Qaeda Sites Point to Tests of Chemicals, New York Times, November 11, 2001, p. B1; Douglas Frantz and David Rohde, A Nation Challenged: Biological Terror2 Pakistanis Linked to Papers on Anthrax Weapons, New York Times, November 28, 2001; and David Rohde, A Nation Challenged: The EvidenceGerm Weapons Plans Found at a Scientists House in Kabul, New York Times, December 1, 2001. 36. Laqueur, Terrorism, pp. 112116.

Behind the Curve 43

seemed almost palatable to some, at least compared to the potential prospect of major war or nuclear cataclysm.37 What has changed is the self-generating nature of international terrorism, with its diverse economic means of support allowing terrorists to carry out attacks sometimes far from the organizations base. As a result, there is an important and growing distinction between where a terrorist organization is spawned and where an attack is launched, making the attacks difcult to trace to their source. Reecting all of these trends, al-Qaeda and its associated groups38 (and individuals) are harbingers of a new type of terrorist organization. Even if alQaeda ceases to exist (which is unlikely), the dramatic attacks of September 2001, and their political and economic effects, will continue to inspire similarly motivated groupsparticularly if the United States and its allies fail to develop broad-based, effective counterterrorist policies over the long term. Moreover, there is signicant evidence that the global links and activities that alQaeda and its associated groups perpetuated are not short term or anomalous. Indeed they are changing the nature of the terrorist threat as we move further into the twenty-rst century. The resulting intersection between the United States, globalization, and international terrorism will dene the major challenges to international security.

The United States, Globalization, and International Terrorism

Whether deliberately intending to or not, the United States is projecting uncoordinated economic, social, and political power even more sweepingly than it is in military terms. Globalization,39 in forms including Westernization, secularization, democratization, consumerism, and the growth of market capitalism, represents an onslaught to less privileged people in conservative cultures repelled by the fundamental changes that these forces are bringingor angered by the distortions and uneven distributions of benets that result.40 This
37. Ibid., pp. 115116. 38. Groups with known or alleged connections to al-Qaeda include Jemaah Islamiyah (Indonesia, Malaysia, and Singapore), the Abu Sayyaf group (Philippines), al-Gamaa al-Islamiyya (Egypt), Harakat ul-Mujahidin (Pakistan), the Islamic Movement of Uzbekistan (Central Asia), Jaish-eMohammed (India and Pakistan), and al-Jihad (Egypt). 39. For the purposes of this article, globalization is a gradually expanding process of interpenetration in the economic, political, social, and security realms, uncontrolled by (or apart from) traditional notions of state sovereignty. Victor D. Cha, Globalization and the Study of International Security, Journal of Peace Research, Vol. 37, No. 3 (March 2000), pp. 391393. 40. With respect to the Islamic world, there are numerous books and articles that point to the phenomenon of antipathy with the Western world, either because of broad cultural incompatibility or

Contending with Terrorism 44

is especially true of the Arab world. Yet the current U.S. approach to this growing repulsion is colored by a kind of cultural navet, an unwillingness to recognizelet alone appreciate or take responsibility forthe inuence of U.S. power except in its military dimension. Even doing nothing in the economic, social, and political policy realms is still doing something, because the United States is blamed by disadvantaged and alienated populations for the powerful Western-led forces of globalization that are proceeding apace, despite the absence of a focused, coordinated U.S. policy. And those penetrating mechanisms of globalization, such as the internet, the media, and the increasing ows of goods and peoples, are exploited in return. Both the means and ends of terrorism are being reformulated in the current environment. the means Important changes in terrorist methods are apparent in the use of new technologies, the movement of terrorist groups across international boundaries, and changes in sources of support. Like globalization itself, these phenomena are all intertwined and overlapping but, for ease of argument, they are dealt with consecutively here. First, the use of information technologies such as the internet, mobile phones, and instant messaging has extended the global reach of many terrorist groups. Increased access to these technologies has so far not resulted in their widely feared use in a major cyberterrorist attack: In Dorothy Dennings words, terrorists still prefer bombs to bytes.41 Activists and terrorist groups have increasingly turned to hacktivismattacks on internet sites, including

a specic conict between Western consumerism and religious fundamentalism. Among the earliest and most notable are Samuel P. Huntington, The Clash of Civilizations? Foreign Affairs, Vol. 72, No. 3 (Summer 1993); Benjamin R. Barber, Jihad vs. McWorld: Terrorisms Challenge to Democracy (New York: Random House, 1995); and Samuel P. Huntington, The Clash of Civilizations and the Remaking of World Order (New York: Simon and Schuster, 1996). 41. For more on cyberterrorism, see Dorothy Denning, Activism, Hacktivism, and Cyberterrorism: The Internet as a Tool for Inuencing Foreign Policy, paper presented at Internet and International Systems: Information Technology and American Foreign Policy Decision-making Workshop at Georgetown University, http:/ /www.nautilus.org/info-policy/workshop/papers/ denning.html (accessed January 5, 2003); Dorothy Denning, Cyberterrorism, testimony before the U.S. House Committee on Armed Services, Special Oversight Panel on Terrorism, 107th Cong., 1st sess., May 23, 2001, available on the Terrorism Research Center website, http:/ / www.cs.georgetown.edu/?denning/infosec/cyberterror.html (accessed January 5, 2003); Jerold Post, Kevin Ruby, and Eric Shaw, From Car Bombs to Logic Bombs: The Growing Threat of Information Terrorism, Terrorism and Political Violence, Vol. 12, No. 2 (Summer 2000), pp. 97122; and Tom Regan, When Terrorists Turn to the Internet, Christian Science Monitor, July 1, 1999, http:/ / www.csmonitor.com (accessed January 5, 2003).

Behind the Curve 45

web defacements, hijackings of websites, web sit-ins, denial-of-service attacks, and automated email bombingsattacks that may not kill anyone but do attract media attention, provide a means of operating anonymously, and are easy to coordinate internationally.42 So far, however, these types of attacks are more an expense and a nuisance than an existential threat. Instead the tools of the global information age have led to enhanced efciency in many terrorist-related activities, including administrative tasks, coordination of operations, recruitment of potential members, communication among adherents, and attraction of sympathizers.43 Before the September 11 attacks, for example, members of al-Qaeda communicated through Yahoo email; Mohammed Atta, the presumed leader of the attacks, made his reservations online; and cell members went online to do research on subjects such as the chemical-dispersing powers of crop dusters. Although not as dramatic as shutting down a power grid or taking over an air trafc control system, this practical use of technology has signicantly contributed to the effectiveness of terrorist groups and the expansion of their range.44 Consider, for example, the lethal impact of the synchronized attacks on the U.S. embassies in 1998 and on New York and Washington in 2001, neither of which would have been possible without the revolution in information technology. When he was arrested in 1995, Ramzi Yousef, mastermind of the 1993 World Trade Center attack, was planning the simultaneous destruction of eleven airliners.45 The internet has become an important tool for perpetuating terrorist groups, both openly and clandestinely. Many of them employ elaborate list serves, collect money from witting or unwitting donors, and distribute savvy political messages to a broad audience online.46 Groups as diverse as Aum Shinrikyo, Israels Kahane Chai, the Popular Front for the Liberation of Palestine, the Kurdistan Workers Party, and Perus Shining Path maintain user-friendly

42. Ibid. Dorothy Denning cites numerous examples, among them: In 1989, hackers released a computer worm into the NASA Space Physics Analysis Network in an attempt to stop a shuttle launch; during Palestinian riots in October 2000, pro-Israeli hackers defaced the Hezbollah website; and in 1999, following the mistaken U.S. bombing of the Chinese embassy in Belgrade during the war in Kosovo, Chinese hackers attacked the websites of the U.S. Department of the Interior, showing images of the three journalists killed during the bombing. 43. Paul R. Pillar, Terrorism and U.S. Foreign Policy (Washington, D.C.: Brookings, 2001), p. 47. 44. Ibid. 45. Simon Reeve, The New Jackals: Ramzi Yousef, Osama bin Laden, and the Future of Terrorism (Boston: Northeastern University Press, 1999), p. 260. 46. Dorothy Denning, Cyberwarriors: Activists and Terrorists Turn to Cyberspace, Harvard International Review, Vol. 23, No. 2 (Summer 2001), pp. 7075. See also Brian J. Miller, Terror.org: An Assessment of Terrorist Internet Sites, Georgetown University, December 6, 2000.

Contending with Terrorism 46

ofcial or unofcial websites, and almost all are accessible in English.47 Clandestine methods include passing encrypted messages, embedding invisible graphic codes using steganography,48 employing the internet to send death threats, and hiring hackers to collect intelligence such as the names and addresses of law enforcement ofcers from online databases.49 All of these measures help to expand and perpetuate trends in terrorism that have already been observed: For example, higher casualties are brought about by simultaneous attacks, a diffusion in terrorist locations is made possible by internet communications, and extremist religious ideologies are spread through websites and videotapes accessible throughout the world. More ominous, globalization makes CBNR weapons increasingly available to terrorist groups.50 Information needed to build these weapons has become ubiquitous, especially through the internet. Among the groups interested in acquiring CBNR (besides al-Qaeda) are the PLO, the Red Army Faction, Hezbollah, the Kurdistan Workers Party, German neo-Nazis, and the Chechens.51 Second, globalization has enabled terrorist organizations to reach across international borders, in the same way (and often through the same channels) that commerce and business interests are linked. The dropping of barriers through the North American Free Trade Area and the European Union, for instance, has facilitated the smooth ow of many things, good and bad, among countries. This has allowed terrorist organizations as diverse as Hezbollah, alQaeda, and the Egyptian al-Gamaat al-Islamiyya to move about freely and establish cells around the world.52 Movement across borders can obviously en47. Miller, Terror.org, pp. 9, 12. 48. Steganography is the embedding of messages usually in pictures, where the messages are disguised so that they cannot be seen with the naked eye. See Denning, Cyberwarriors. 49. I am indebted to Dorothy Denning for all of this information. The Provisional IRA hired contract hackers to nd the addresses of British intelligence and law enforcement ofcers. See Denning, Cyberterrorism; and Denning, Cyberwarriors. 50. There are many recent sources on CBNR. Among the best are Jonathan B. Tucker, ed., Toxic Terror: Assessing Terrorist Use of Chemical and Biological Weapons (Cambridge, Mass.: MIT Press, 2000); Joshua Lederberg, Biological Weapons: Limiting the Threat (Cambridge, Mass.: MIT Press, 1999); Richard A. Falkenrath, Robert D. Newman, and Bradley A. Thayer, Americas Achilles Heel: Nuclear, Biological, and Chemical Terrorism and Covert Attack (Cambridge, Mass.: MIT Press, 1998); Gurr and Cole, The New Face of Terrorism; Jessica Stern, The Ultimate Terrorists (Cambridge, Mass.: Harvard University Press, 1999); and Brad Roberts, ed., Terrorism with Chemical and Biological Weapons: Calibrating Risks and Responses (Alexandria, Va.: Chemical and Biological Arms Control Institute, 1997). 51. See Falkenrath, Newman, and Thayer, Americas Achilles Heel, pp. 3146. 52. A clear example of this phenomenon was the uncovering in December 2001 of a multinational plot in Singapore by the international terrorist group Jemaah Islamiyah to blow up several Western targets, including the U.S. embassy. A videotape of the intended targets (including a description of

Behind the Curve 47

able terrorists to carry out attacks and potentially evade capture, but it also complicates prosecution if they are apprehended, with a complex maze of extradition laws varying greatly from state to state. The increased permeability of the international system has also enhanced the ability of nonstate terrorist organizations to collect intelligence (not to mention evade it); states are not the only actors interested in collecting, disseminating, and/or acting on such information. In a sense, then, terrorism is in many ways becoming like any other international enterprisean ominous development indeed. Third, terrorist organizations are broadening their reach in gathering nancial resources to fund their operations. This is not just an al-Qaeda phenomenon, although bin Ladens organizationespecially its numerous business interestsgures prominently among the most innovative and wealthy pseudocorporations in the international terrorist network. The list of groups with global nancing networks is long and includes most of the groups identied by the U.S. government as foreign terrorist organizations, notably Aum Shinrikyo, Hamas, Hezbollah, and the Tamil Tigers. Sources of nancing include legal enterprises such as nonprot organizations and charities (whose illicit activities may be a small or large proportion of overall nances, known or unknown to donors); legitimate companies that divert prots to illegal activities (such as bin Ladens large network of construction companies); and illegal enterprises such as drug smuggling and production (e.g., the Revolutionary Armed Forces of ColombiaFARC), bank robbery, fraud, extortion, and kidnapping (e.g., the Abu Sayyaf group, Colombias National Liberation Army, and FARC).53 Websites are also important vehicles for raising funds. Although no comprehensive data are publicly available on how lucrative this avenue is, the proliferation of terrorist websites with links or addresses for contributions is at least circumstantial evidence of their usefulness. The uid movement of terrorists nancial resources demonstrates the growing informal connections that are countering the local fragmentation caused elsewhere by globalization. The transit of bars of gold and bundles of dollars
the plans in Arabic) was discovered in Afghanistan after al-Qaeda members ed. Thus there are clear connections between these organizations, as well as evidence of cooperation and coordination of attacks. See, for example, Dan Murphy, Activated Asian Terror Web Busted, Christian Science Monitor, January 23, 2002, http:/ /www.csmonitor.com (accessed January 23, 2002); and Rajiv Changrasekaran, Al Qaedas Southeast Asian Reach, Washington Post, February 3, 2002, p. A1. 53. Rensselaer Lee and Raphael Perl, Terrorism, the Future, and U.S. Foreign Policy, issue brief for Congress, received through the Congressional Research Service website, order code IB95112, Congressional Research Service, Library of Congress, July 10, 2002, p. CRS-6.

Contending with Terrorism 48

across the border between Afghanistan and Pakistan as U.S. and allied forces were closing in on the Talibans major strongholds is a perfect example. Collected by shopkeepers and small businessmen, the money was moved by operatives across the border to Karachi, where it was transferred in the millions of dollars through the informal hawala or hundi banking system to the United Arab Emirates.54 There it was converted into gold bullion and scattered around the world before any government could intervene. In this way, al-Qaeda preserved and dispersed a proportion of its nancial resources.55 In addition to gold, money was transferred into other commoditiessuch as diamonds in Sierra Leone and the Democratic Republic of Congo, and tanzanite from Tanzaniaall while hiding the assets and often making a prot,56 and all without interference from the sovereign governments that at the time were at war with al-Qaeda and the Taliban.57 As this example illustrates, globalization does not necessarily require the use of high technology: It often takes the form of traditional practices used in innovative ways across increasingly permeable physical and commercial borders. Terrorist groups, whose assets comparatively represent only a small fraction of the amount of money that is moved by organized crime groups and are thus much more difcult to track, use everything from direct currency transport (by couriers) to reliance on traditional banks, Islamic banks, money changers (using accounts at legitimate institutions), and informal exchange (the hawala or hundi system). This is by no means a comprehensive presentation of global interpenetration of terrorist means, and some of the connections described above have existed for some time and in other contexts. The broad strategic picture, however, is of

54. Roger G. Weiner, The Financing of International Terrorism, Terrorism and Violence Crime Section, Criminal Division, U.S. Department of Justice, October 2001, p. 3. According to Weiner, the hawala (or hundi) system relies entirely on trust that currency left with a particular service provider or merchant will be paid from bank accounts he controls overseas to the recipient specied by the party originating the transfer. Ibid. See also Douglas Frantz, Ancient Secret System Moves Money Globally, New York Times, October 3, 2001, http:/ /www.nytimes.com (accessed October 3, 2001). 55. International efforts to freeze bank accounts and block transactions between suspected terrorists have hindered, at least to some degree, al-Qaedas ability to nance attacks; however, a proportion remains unaccounted for. Cash Moves a Sign Al-Qaeda Is Regrouping, Straits Times, March 18, 2002, http:/ /www.straitstimes.asia1.com.sg (accessed March 18, 2002). 56. U.S. Department of State, Patterns of Global Terrorism, 2001. According to the U.S. Department of State, Hezbollah also may have transferred resources by selling millions of dollars worth of Congolese diamonds to nance operations in the Middle East. 57. Douglas Farah, Al Qaedas Road Paved with Gold, Washington Post, February 17, 2002, pp. A1, A32.

Behind the Curve 49

an increasing ability of terrorist organizations to exploit the same avenues of communication, coordination, and cooperation as other international actors, including states, multinational corporations, nongovernmental organizations, and even individuals. It would be nave to assume that what is good for international commerce and international communication is not also good for international terrorists58who are increasingly becoming opportunistic entrepreneurs whose product (often quite consciously sold) is violence against innocent targets for a political end. the ends The objectives of international terrorism have also changed as a result of globalization. Foreign intrusions and growing awareness of shrinking global space have created incentives to use the ideal asymmetrical weapon, terrorism, for more ambitious purposes. The political incentives to attack major targets such as the United States with powerful weapons have greatly increased. The perceived corruption of indigenous customs, religions, languages, economies, and so on are blamed on an international system often unconsciously molded by American behavior. The accompanying distortions in local communities as a result of exposure to the global marketplace of goods and ideas are increasingly blamed on U.S.sponsored modernization and those who support it. The advancement of technology, however, is not the driving force behind the terrorist threat to the United States and its allies, despite what some have assumed.59 Instead, at the heart of this threat are frustrated populations and international movements that are increasingly inclined to lash out against U.S.-led globalization. As Christopher Coker observes, globalization is reducing tendencies toward instrumental violence (i.e., violence between states and even between communities), but it is enhancing incentives for expressive violence (or violence that is ritualistic, symbolic, and communicative).60 The new international terrorism is

58. Pillar, Terrorism and U.S. Foreign Policy, p. 48. 59. Many in the United States focus on the technologies of terrorism, with a much less developed interest in the motivations of terrorists. Brian M. Jenkins, Understanding the Link between Motives and Methods, in Roberts, Terrorism with Chemical and Biological Weapons, pp. 4351. An example of a study that focuses on weapons and not motives is Sidney D. Drell, Abraham D. Sofaer, and George W. Wilson, eds., The New Terror: Facing the Threat of Biological and Chemical Weapons (Stanford, Calif.: Hoover Institution, 1999). 60. Christopher Coker, Globalisation and Insecurity in the Twenty-rst Century: NATO and the Management of Risk, Adelphi Paper 345 (London: International Institute for Strategic Studies, June 2002), p. 40.

Contending with Terrorism 50

increasingly engendered by a need to assert identity or meaning against forces of homogeneity, especially on the part of cultures that are threatened by, or left behind by, the secular future that Western-led globalization brings. According to a report recently published by the United Nations Development Programme, the region of greatest decit in measures of human developmentthe Arab worldis also the heart of the most threatening religiously inspired terrorism.61 Much more work needs to be done on the signicance of this correlation, but increasingly sources of political discontent are arising from disenfranchised areas in the Arab world that feel left behind by the promise of globalization and its assurances of broader freedom, prosperity, and access to knowledge. The results are dashed expectations, heightened resentment of the perceived U.S.-led hegemonic system, and a shift of focus away from more proximate targets within the region. Of course, the motivations behind this threat should not be oversimplied: Anti-American terrorism is spurred in part by a desire to change U.S. policy in the Middle East and Persian Gulf regions as well as by growing antipathy in the developing world vis--vis the forces of globalization. It is also crucial to distinguish between the motivations of leaders such as Osama bin Laden and their followers. The former seem to be more driven by calculated strategic decisions to shift the locus of attack away from repressive indigenous governments to the more attractive and media-rich target of the United States. The latter appear to be more driven by religious concepts cleverly distorted to arouse anger and passion in societies full of pent-up frustration. To some degree, terrorism is directed against the United States because of its engagement and policies in various regions.62 Anti-Americanism is closely related to antiglobalization, because (intentionally or not) the primary driver of the powerful forces resulting in globalization is the United States. Analyzing terrorism as something separate from globalization is misleading and potentially dangerous. Indeed globalization and terrorism are intricately intertwined forces characterizing international security in the twenty-rst century. The main question is whether terrorism will succeed in disrupting the

61. The indicators studied included respect for human rights and human freedoms, the empowerment of women, and broad access to and utilization of knowledge. See United Nations Development Programme, Arab Fund for Economic and Social Development, Arab Human Development Report, 2002: Creating Opportunities for Future Generations (New York: United Nations Development Programme, 2002). 62. Martha Crenshaw, Why America? The Globalization of Civil War, Current History, December 2001, pp. 425432.

Behind the Curve 51

promise of improved livelihoods for millions of people on Earth. Globalization is not an inevitable, linear development, and it can be disrupted by such unconventional means as international terrorism. Conversely, modern international terrorism is especially dangerous because of the power that it potentially derives from globalizationwhether through access to CBNR weapons, global media outreach, or a diverse network of nancial and information resources.

Prospects for the Future

Long after the focus on Osama bin Laden has receded and U.S. troops have quit their mission in Afghanistan, terrorism will be a serious threat to the world community and especially to the United States. The relative preponderance of U.S. military power virtually guarantees an impulse to respond asymmetrically. The lagging of the Arab region behind the rest of the world is impelling a violent redirection of antiglobalization and antimodernization forces toward available targets, particularly the United States, whose scope and policies are engendering rage. Al-Qaeda will eventually be replaced or redened, but its successors reach may continue to grow via the same globalized channels and to direct their attacks against U.S. and Western targets. The current trajectory is discouraging, because as things currently stand, the wellspring of terrorisms means and ends is likely to be renewed: Arab governments will probably not reform peacefully, and existing Western governments and their supporting academic and professional institutions are disinclined to understand or analyze in depth the sources, patterns, and history of terrorism. Terrorism is a by-product of broader historical shifts in the international distribution of power in all of its formspolitical, economic, military, ideological, and cultural. These are the same forms of power that characterize the forces of Western-led globalization. At times of dramatic international change, human beings (especially those not beneting from the changeor not beneting as much or as rapidly from the change) grasp for alternative means to control and understand their environments. If current trends continue, widening global disparities, coupled with burgeoning information and connectivity, are likely to accelerateunless the terrorist backlash, which is increasingly taking its inspiration from misoneistic religious or pseudoreligious concepts, successfully counters these trends. Because of globalization, terrorists have access to more powerful technologies, more targets, more territory, more means of recruitment, and more exploitable sources of rage than ever before. The Wests

Contending with Terrorism 52

twentieth-century approach to terrorism is highly unlikely to mitigate any of these long-term trends. From a Manichean perspective, the ad hoc and purportedly benign intentions of the preponderant, secular West do not seem benign at all to those ill served by globalization. To frustrated people in the Arab and Muslim world, adherence to radical religious philosophies and practices may seem a rational response to the perceived assault, especially when no feasible alternative for progress is offered by their own governments. This is not to suggest that terrorists should be excused because of environmental factors or conditions. Instead, Western governments must recognize that the tiny proportion of the population that ends up in terrorist cells cannot exist without the availability of broader sources of active or passive sympathy, resources, and support. Those avenues of sustenance are where the center of gravity for an effective response to the terrorist threat must reside. The response to transnational terrorism must deal with the question of whether the broader enabling environment will increase or decrease over time, and the answer will be strongly inuenced by the policy choices that the United States and its allies make in the near future.

Conclusions and Policy Prescriptions

The characteristics and causes of the current threat can only be analyzed within the context of the deadly collision occurring between U.S. power, globalization, and the evolution of international terrorism. The U.S. government is still thinking in outdated terms, little changed since the end of the Cold War. It continues to look at terrorism as a peripheral threat, with the focus remaining on states that in many cases are not the greatest threat. The means and the ends of terrorism are changing in fundamental, important ways; but the means and the ends of the strategy being crafted in response are not. Terrorism that threatens international stability, and particularly U.S. global leadership, is centered on power-based political causes that are enduring: the weak against the strong, the disenfranchised against the establishment, and the revolutionary against the status quo. Oversimplied generalizations about poverty and terrorism, or any other single variable, are caricatures of a serious argument.63 The rise in political and material expectations as a result of the in63. A number of recent arguments have been put forth about the relationship between poverty and terrorism. See, for example, Anatol Lieven, The Roots of Terrorism, and a Strategy against It,

Behind the Curve 53

formation revolution is not necessarily helpful to stability, in the same way that rising expectations led terrorists to take up arms against the czar in Russia a century ago. Indeed the fact that so many people in so many nations are being left behind has given new ammunition to terrorist groups; produced more sympathy for those willing to take on the United States; and spurred Islamic radical movements to recruit, propagandize, and support terrorism throughout many parts of the Muslim world. The al-Qaeda network is an extremist religious terrorist organization, its Taliban puppet regime was lled with religious zealots, and its suicide recruits were convinced that they were waging a just holy war. But the driving forces of twenty-rst-century terrorism are power and frustration, not the pursuit of religious principle. To dismiss the broad enabling environment would be to focus more on the symptoms than the causes of modern terrorism. The prescriptions for countering and preventing terrorism should be twofold: First, the United States and other members of the international community concerned about this threat need to use a balanced assortment of instruments to address the immediate challenges of the terrorists themselves. Terrorism is a complex phenomenon; it must be met with short-term military action, informed by in-depth, long-term, sophisticated analysis. Thus far, the response has been virtually all the former and little of the latter. Second, the United States and its counterterrorist allies must employ a much broader array of longer-term policy tools to reshape the international environment, which enables terrorist networks to breed and become robust. The mechanisms of globalization need to be exploited to thwart the globalization of terrorism. In the short term, the United States must continue to rely on capable military forces that can sustain punishing air strikes against terrorists and those who harbor them with an even greater capacity for special operations on the ground. This requires not only improved stealthy, long-range power projection capabilities but also agile, highly trained, and lethal ground forces, backed up
Prospect (London), October 2001, http:/ /www.ceip.org/les/Publications/lieventerrorism.asp? from pubdate (accessed November 17, 2002); and Daniel Pipes, God and Mammon: Does Poverty Cause Militant Islam? National Interest, No. 66 (Winter 2001/02), pp. 1421. This is an extremely complex question, however, and much work remains to be done. On the origins of the new religious terrorism, see Hoffman, Inside Terrorism; and Mark Juergensmeyer, Terror in the Mind of God: The Global Rise of Religious Violence (Berkeley: University of California Press, 2000). Important earlier studies on the sources of terrorism include Martha Crenshaw, The Causes of Terrorism, Comparative Politics, July 1981, pp. 379399; Martha Crenshaw, Terrorism in Context (University Park: Pennsylvania State University Press, 1995); and Walter Reich, ed., Origins of Terrorism: Psychologies, Ideologies, Theologies, States of Mind, 2d ed. (Washington, D.C.: Woodrow Wilson Center for International Scholars, 1998).

Contending with Terrorism 54

with greater intelligence, including human intelligence supported by individuals with language skills and cultural training. The use of military force continues to be important as one means of responding to terrorist violence against the West, and there is no question that it effectively preempts and disrupts some international terrorist activity, especially in the short term.64 Over time, however, the more effective instruments of policy are likely to remain the nonmilitary ones. Indeed the United States needs to expand and deepen its nonmilitary instruments of power such as intelligence, public diplomacy, cooperation with allies, international legal instruments, and economic assistance and sanctions. George Kennan, in his 1947 description of containment, put forth the same fundamental argument, albeit against an extremely different enemy.65 The strongest response that the United States can muster to a serious threat has to include political, economic, and military capabilitiesin that order; yet, the U.S. government consistently structures its policies and devotes its resources in the reverse sequence. The economic and political roots of terrorism are complex, increasingly worrisome, and demanding of as much breadth and subtlety in response as they display in their genesis. The United States must therefore be strategic in its response: An effective grand strategy against terrorism involves planning a global campaign with the most effective means available, not just the most measurable, obvious, or gratifying. It must also include plans for shaping the global environment after the so-called war on terrorism has endedor after the current political momentum has subsided. The United States, working with other major donor nations, needs to create an effective incentive structure that rewards good performersthose countries with good governance, inclusive education programs, and adequate social programsand works around bad performers and intervenes to assist so-called failed states. Also for the longer term, the United States and its allies need to project a vision of sustainable developmentof economic growth, equal access to basic social needs such as education and health, and good governancefor the developing world. This is particularly true in mostly Muslim countries whose populations are angry with the United States over a perceived double standard regarding its long-standing support for Israel at the expense
64. For more discussion on the traditional elements of U.S. grand strategy, especially military strategy, see Barry R. Posen, The Struggle against Terrorism: Grand Strategy, Strategy, and Tactics, International Security, Vol. 26, No. 3 (Winter 2001/02), pp. 3955. 65. George F. Kennan, The Sources of Soviet Conduct, Foreign Affairs, Vol. 25, No. 4 (July 1947), pp. 575576.

Behind the Curve 55

of Palestinians, policies against the regime of Saddam Hussein at the expense of some Iraqi people, and a general abundance of American power, including the U.S. military presence throughout the Middle East. Whether these policies are right or wrong is irrelevant here; the point is that just as the denition of terrorism can be subjective and value laden, so too can the response to terrorism take into account perceptions of reality. In an attempt to craft an immediate military response, the U.S. government is failing to put into place an effective long-term grand strategy. This is not just a problem for the U.S. government. The inability to develop a strategy with a deep-rooted, intellectually grounded understanding of the history, patterns, motivations, and types of terrorism is reective of the paucity of understanding of the terrorist phenomenon in the academic community. Terrorism is considered too policy-oriented an area of research in political science,66 and it operates in an uncomfortable intersection between disciplines unaccustomed to working together, including psychology, sociology, theology, economics, anthropology, history, law, political science, and international relations. In political science, terrorism does not t neatly into either the realist or liberal paradigms, so it has been largely ignored.67 There are a few outstanding, well-established senior scholars in the terrorism studies community people such as Martha Crenshaw, David Rapoport, and Paul Wilkinsonbut in the United States, most of the publicly available work is being done in policy-oriented research institutes or think tanks that are sometimes limited by the narrow interests and short time frames of the government contracts on which they depend. Some of that research is quite good,68 but it is not widely known within the academy. The situation for graduate students who wish to study terrorism is worse: A principal interest in terrorism virtually guarantees exclusion from consideration for most academic positions. This would not necessarily be a problem if the bureaucracy were more exible and creative than the academy is, but as we know from the analysis of the behavior of U.S. agencies shortly before September 11, it is not. In the United States, academe is no more strategic in its understanding of terrorism than is the U.S. government.
66. See the extremely insightful article by Bruce W. Jentleson, The Need for Praxis: Bringing Policy Relevance Back In, International Security, Vol. 26, No. 4 (Spring 2002), pp. 169183. 67. I am indebted to Fiona Adamson for this observation. 68. Important terrorism scholars in the think tank community include Walter Laqueur (Center for Strategic and International Studies), Brian Jenkins (RAND), Bruce Hoffman (RAND) and, from the intelligence community, Paul Pillar. This list is illustrative, not comprehensive.

Contending with Terrorism 56

The globalization of terrorism is perhaps the leading threat to long-term stability in the twenty-rst century. But the benet of globalization is that the international response to terrorist networks has also begun to be increasingly global, with international cooperation on law enforcement, intelligence, and especially nancial controls being areas of notable recent innovation.69 If globalization is to continueand there is nothing foreordained that it willthen the tools of globalization, including especially international norms, the rule of law, and international economic power, must be fully employed against the terrorist backlash. There must be a deliberate effort to move beyond the current episodic interest in this phenomenon: Supercial arguments and short attention spans will continue to result in event-driven policies and ultimately more attacks. Terrorism is an unprecedented, powerful nonstate threat to the international system that no single state, regardless of how powerful it may be in traditional terms, can defeat alone, especially in the absence of long-term, serious scholarship engaged in by its most creative minds.

69. On these issues, see Cronin and Ludes, The Campaign against International Terrorism.

Motives for Martyrdom

Motives for Martyrdom

Al-Qaida, Sala Jihad, and the Spread of Suicide Attacks

Assaf Moghadam

uicide missionsor attacks whose success is dependent on the death of their perpetrator/sare one of the most lethal tactics employed by terrorist and insurgent groups today. Moreover, they have demonstrated great potential to create turbulence in international affairs.1 The four suicide attacks of September 11, 2001, and the war in Iraqwhere suicide operations have become the signature mode of attackhave highlighted how this tactic can lead to considerable losses of human life and physical infrastructure while inuencing the course of global events in their wake. During the 1980s and 1990s, suicide missions wreaked considerable havoc on their targets; yet these targets were relatively few in number. The vast majority of attacks took place in only a handful of countries, namely, Israel, Lebanon, Sri Lanka, and Turkey. More than thirty-ve countries on every continent save for Antarctica and Australia have experienced the wanton violence brought on by suicide attacks. In the past decade, suicide bombings have not only occurred in a growing number of countries, but these attacks have been planned and executed by an even greater number of organizations and have killed larger numbers of people every year. The targets of these attacks have also undergone some shifts. More suicide bombings have occurred in Iraq since 2003 than in all other countries in the twenty-ve years preceding the U.S.-led invasion of Iraq. Suicide missions have increasingly targeted Muslims and have been adopted as part of
Assaf Moghadam is Assistant Professor and Senior Associate at the Combating Terrorism Center at West Point, and Research Fellow at the Initiative on Religion in International Affairs at Harvard Universitys Belfer Center for Science and International Affairs. This article is drawn from his book The Globalization of Martyrdom: Al Qaeda, Sala Jihad, and the Diffusion of Suicide Attacks (Baltimore, Md.: Johns Hopkins University Press, 2008), but it analyzes a larger, more up-to-date set of data. The author would like to thank participants at Ph.D. colloquia and research seminars at the Fletcher School at Tufts University, the John M. Olin Institute for Strategic Studies, and the Belfer Center for Science and International Affairs, both at Harvard University. He is especially grateful to the anonymous reviewers for their valuable insights. 1. I use the terms suicide missions, suicide attacks, and suicide operations interchangeably. The term suicide missions is drawn from Diego Gambetta, ed., Making Sense of Suicide Missions (Oxford: Oxford University Press, 2005).
International Security, Vol. 33, No. 3 (Winter 2008/09), pp. 4678 2009 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

57

Contending with Terrorism 58

a strategy not only to gain a national homeland, but also to depose regimes regarded as un-Islamic. In recent years, suicide missions have been launched in countries with little or no prior history of such attacks, including Afghanistan, Pakistan, Somalia, Uzbekistan, and Yemen.2 Perhaps most important for the West, suicide attacks are no longer a distant threat, having targeted cities such as London, Madrid, and New York.3 This article examines the causes and characteristics of the phenomenon of the globalization of martyrdom.4 Two interrelated factors explain the proliferation of suicide missions: the evolution of al-Qaida into a global terrorist actor and the growing appeal of Sala jihad, the guiding ideology of al-Qaida and its associated movements. Discussion of these factors is largely missing from scholarly work on suicide attacks.5 Although a few scholars have claimed that most contemporary suicide attacks can be attributed to jihadist groups,6 this article is the rst to test this argument empirically. In the rst section, I present my data set on suicide missions from December

2. Only a small number of suicide attacks occurred in some of these countries before the turn of the millennium. In Pakistan there were just three such attacks. On November 19, 1995, a truck bomb rocked the Egyptian embassy in Islamabad. On December 21, 1995, a car bomb exploded in a crowded street in Peshawar in northwestern Pakistan, killing at least 30 and wounding more than 100; and on April 29, 1996, a suicide bomber blew himself up inside a passenger-lled bus returning from a religious festival in Punjab, killing 52. A handful of attacks occurred in 2000 and 2002, but suicide attacks in Pakistan were widely adopted only after 2003. 3. Although the Madrid train bombings of March 11, 2004, were not a suicide mission, seven of the suspects detonated themselves weeks later in the Madrid suburb of Leganes, as Spanish special forces were about to storm their apartment. See Rogelio Alonso and Fernando Reinares, Maghreb Immigrants Becoming Suicide Terrorists: A Case Study on Religious Radicalization Processes in Spain, in Amy Pedahzur, ed., Root Causes of Suicide Terrorism: The Globalization of Martyrdom (New York: Routledge, 2006), pp. 179198. 4. See Assaf Moghadam, The New Martyrs Go Global, Boston Globe, November 18, 2005. 5. Some exceptions exist. Yoram Schweitzer and Sari Goldstein Ferber highlight the role of alQaida in internationalizing suicide attacks, but they do not emphasize the role of Sala jihadist ideology. Schweitzer and Goldstein Ferber, Al-Qaida and the Internationalization of Suicide Terrorism, Jaffee Center Memorandum, No. 78 (Tel Aviv: Jaffee Center for Strategic Studies, Tel Aviv University, November 2005). Mohammed M. Hafez discusses the appeal of Sala jihadist ideology, but he limits his analysis to the case of Iraq. Hafez, Suicide Bombers in Iraq: The Strategy and Ideology of Martyrdom (Washington, D.C.: United States Institute of Peace Press, 2007). 6. Bruce Hoffman, citing data from the RAND Terrorism Incident Database, said that 78 percent of all suicide attacks since 1968 occurred after September 11, 2001, adding that 31 out of the 35 groups employing this tactic were Islamic. He did not, however, specically refer to jihadist groups, nor did he argue that these groups were ideologically motivated. Hoffman, email correspondence with author, December 27, 2005. Scott Atran points out that most suicide terrorists today are inspired by a global Jihadism, but he does not provide empirical support for his claim. See Atran, The Moral Logic and Growth of Suicide Terrorism, Washington Quarterly, Vol. 29, No. 2 (Spring 2006), p. 139. In an earlier study, I, too, suggested that contemporary patterns of suicide attacks are dominated by Sala jihadists without offering any empirical data. See Assaf Moghadam, Suicide Terrorism, Occupation, and the Globalization of Martyrdom: A Critique of Dying to Win, Studies in Conict and Terrorism, Vol. 29, No. 8 (December 2006), pp. 707729.

Motives for Martyrdom 59

1981 through March 2008. In the second section, I review existing studies of suicide attacks and discuss their limitations. In the third section, I present the main argument of this study. In the fourth section, I discuss the theoretical implications of my ndings. The conclusion offers practical implications of these ndings for efforts to confront the challenges posed by suicide operations.

The Global Rise of Suicide Attacks

According to most indicators, suicide missions have been on the rise since 1981, but they have grown at an unprecedented pace since the turn of the millennium.7 According to my data set, 1,857 suicide attacks were perpetrated from December 1981 through March 2008.8 Throughout the 1980s and early 1990s, the number of suicide missions remained relatively small, not exceeding 7 attacks per year, with the exception of 1985, when 22 such attacks were carried out. Beginning in 1994, the number of suicide missions started to increase, peaking temporarily in 1995, when 27 attacks were launched. The number dropped slightly in the second half of the 1990s. The year 2000 witnessed 37 attacksa record number. It also signaled the beginning of an upward trend in the number of suicide missions that would span most of the rst decade of this century. Thus, between 2000 and 2007, the number of attacks rose steadily each year, from 54 in 2001 to 71 in 2002, 81 in 2003, 104 in 2004, 348 in 2005, 353 in 2006, and 535 in 2007 (see gure 1).9 The global proliferation of suicide missions is reected in the rise in the number of organizations that employed them. From 1981 to 1990, an average of 1.6 organizations perpetrated suicide attacks every year. From 1991 to 2000,

7. Suicide attacks and their precursors have existed since biblical times. Prior to 1981, however, these attacks were not generally considered acts of terrorism. The modern phenomenon of suicide terrorism began in Lebanon with the December 1981 bombing of the Iraqi embassy. For an overview of modern-day suicide attacks and their historical precursors, see Introduction, in Assaf Moghadam, The Globalization of Martyrdom: Al-Qaida, Sala Jihad, and the Diffusion of Suicide Attacks (Baltimore, Md.: Johns Hopkins University Press, 2008). 8. The data set relies heavily on two sources. The rst is the Suicide Terrorism Database collected by the National Security Studies Center at the University of Haifa in Israel. For a version of the database that includes 1,165 suicide attacks conducted until April 17, 2006, see http:/ / www.laits.utexas.edu/tiger/terrorism_data/suicide_attacks_worldwide/. For data since April 18, 2006, as well as for all data related to suicide attacks in Iraq and Afghanistan, I used the National Counterterrorism Centers (NCTC) Worldwide Incidents Tracking System. The NCTC database is available at www.nctc.gov. After combining these two databases, eliminating duplicates, and updating the resulting database with additional information, I arrived at a data set with a total of 1,857 suicide attacks recorded from December 1981 to March 2008. For a copy of the data set, please contact me at [email protected]. 9. One hundred nine attacks were recorded in the rst quarter of 2008.

Contending with Terrorism 60

Figure 1. Number of Suicide Missions, 19812007

the average increased to 4.8. And from 2001 to 2007, it rose to an average of 14.3. Suicide attacks have exacted an enormous human toll. The 1,857 suicide attacks recorded in my data set claimed 20,603 lives and left at least 48,209 wounded. Of these, more than 87 percent were killed and more than 80 percent injured in this decade alone (see gure 2). The overall trend in the number of injured in suicide attacks is comparable to the numbers of people killed, although in two years1996 and 1998the numbers of wounded (2,082 and 4,666) were particularly high.10 The sharp increase in dead and wounded beginning in 2003 stemmed largely from the war in Iraq. Based on my data set, the average suicide mission in the period under review killed 11 people and injured 26. These numbers should be approached with caution, however, because the range was exceedingly wide, and because information about casualty rates (especially number of wounded) is often incomplete or missing altogether. The median of the numbers of people killed was 3, and that of people wounded was 9. From 1981 to 2007, the number of countries in which suicide missions were
10. The 1996 gure includes a suicide car bombing on January 31, 1996, by a member of the Liberation Tigers of Tamil Eelam (LTTE), in which 19 people died and an estimated 1,400 people were wounded. The 1998 gure includes the August 7, 1998, suicide bombing of the U.S. embassy in Nairobi, Kenya, which killed 213 people and wounded an estimated 4,000.

Motives for Martyrdom 61

Figure 2. Number of Casualties from Suicide Missions, 19812007

launched generally increased. The average number of countries that experienced a suicide attack from 1981 to 1994 was a relatively low 1.7 per year. Lebanon and Sri Lanka were the most frequently targeted countries. From 1995 to 2007, an average of 8.6 countries per year experienced a suicide attack. In both 2005 and 2006, suicide missions were executed in 15 countriesthe highest number of countries recorded. According to my data, 1,020 suicide missions took place in Iraq (54.9 percent of all suicide missions worldwide); 235 (12.7 percent) in Afghanistan; 188 (10.1 percent) in Israel (including the West Bank and Gaza Strip); 107 (5.8 percent) in Sri Lanka; 88 (4.7 percent) in Pakistan; 41 (2.2 percent) in Lebanon; 37 (2.0 percent) in Russia; and 141 (7.5 percent) in 29 other countries.11

Existing Theories on Suicide Attacks

Studies dedicated to explaining the causes of suicide attacksrare until the terrorist attacks of September 11, 2001can be divided into at least four gen-

11. These other countries are Algeria, Argentina, Bangladesh, China, Croatia, Egypt, Finland, India, Indonesia, Italy, Jordan, Kenya, Kuwait, Laos, Moldova, Morocco, Panama, Qatar, Saudi Arabia, Serbia, Somalia, Syria, Tanzania, Tunisia, Turkey, the United Kingdom, the United States, Uzbekistan, and Yemen.

Contending with Terrorism 62

eral categories: those that focus on the level of the individual bomber;12 those stressing the group or organizational factors;13 those emphasizing sociostructural causes; and those suggesting the need to integrate multiple levels of analysis.14 At the individual level of analysis, scholars from a variety of disciplines have concluded that suicide bombers often believe that they are acting for altruistic reasons.15 Most analysts reject the notion that suicide bombers act irrationally, arguing that these martyrs believe that the benets of perpetrating suicide attacks outweigh the costs.16 In addition, they agree that proling suicide bombers is virtually impossible given their diverse backgrounds. Moreover, they claim that suicide bombersmuch like terrorists in generaldo not suffer from a salient psychopathology,17 and thus dismiss mental illness as a reason for their actions.18 Scholars focused on individual motivations suggest the following: a strong commitment to a group or cause,19 a desire for revenge,20 an expectation of

12. See, for example, Joan Lachkar, The Psychological Make-up of a Suicide Bomber, Journal of Psychohistory, Vol. 29, No. 4 (Spring 2002), pp. 349367; David Lester, Bijou Yang, and Mark Lindsay, Suicide Bombers: Are Psychological Proles Possible? Studies in Conict and Terrorism, Vol. 27, No. 4 (JulyAugust 2004), pp. 283295; Eyad Sarraj and Linda Butler, Suicide Bombers: Dignity, Despair, and the Need of Hope, Journal of Palestine Studies, Vol. 31, No. 4 (Summer 2002), pp. 7176; and Anat Berko, The Path to Paradise: The Inner World of Suicide Bombers and Their Dispatchers (Westport, Conn.: Praeger Security International, 2007). 13. Ehud Sprinzak, Rational Fanatics, Foreign Policy, No. 120 (September/October 2000), pp. 67 73; Mia Bloom, Dying to Kill: The Allure of Suicide Terror (New York: Columbia University Press, 2005); Robert A. Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (New York: Random House, 2005); and Ami Pedahzur, Suicide Terrorism (Cambridge: Polity, 2005). 14. Assaf Moghadam, The Roots of Suicide Terrorism: A Multi-Causal Approach, in Pedahzur, Root Causes of Suicide Terrorism, pp. 81107; and Hafez, Suicide Bombers in Iraq. 15. See, for example, Ami Pedahzur, Arie Perliger, and Leonard Weinberg, Altruism and Fatalism: The Characteristics of Palestinian Suicide Terrorists, Deviant Behavior, Vol. 24, No. 4 (July 2003), pp. 405423. 16. Sprinzak, Rational Fanatics; Assaf Moghadam, Palestinian Suicide Terrorism in the Second Intifada: Motivations and Organizational Aspects, Studies in Conict and Terrorism, Vol. 26, No. 2 (March 2003), pp. 6592; Bloom, Dying to Kill; Pedahzur, Suicide Terrorism; and Bruce Hoffman and Gordon H. McCormick, Terrorism, Signaling, and Suicide Attack, Studies in Conict and Terrorism, Vol. 27, No. 4 (July 2004), pp. 243281. 17. Clark R. McCauley and M.E. Segal, Social Psychology of Terrorist Groups, in Clyde A. Hendrick, ed., Group Processes and Intergroup Relations: Review of Personality and Social Psychology (Newbury Park, Calif.: Sage, 1987). 18. Jeff Victoroff, The Mind of the Terrorist: A Review and Critique of Psychological Approaches, Journal of Conict Resolution, Vol. 49, No. 1 (February 2005), pp. 342. 19. According to some researchers, suicide attackers tend to act out of a deep sense of commitment to a larger cause, to their social network, or to a terrorist organization. See, for example, Pedahzur, Suicide Terrorism, pp. 126134. 20. A suicide attacker may acquire a thirst for revenge after the death of a family member or friend, which can be reinforced by a perceived sense of humiliation. For an application of humilia-

Motives for Martyrdom 63

benets after death,21 and personal crisis.22 No study, however, has identied either necessary or sufcient conditions for an individuals resort to suicide terrorism, that is, why some highly committed individuals become suicide bombers while others do not, or why revenge leads to suicide terrorism in some cases and not in others. Nor has any study explained the globalization of suicide terrorism. A second category of studies focuses on the organizational-strategic level of analysis. Building on the theories advanced by Martha Crenshaw, who argues that terrorist organizations believe that violence is the best means to advance their political goals,23 several scholars have suggested that terrorist organizations engage in suicide attacks to fulll rational objectives, ranging from basic survival to sophisticated strategic and tactical plans for success. Suicide terrorism, the argument goes, can weaken an external opponent while strengthening the organization itself. Internally, suicide attacks can strengthen the group because they enhance its perceived need to survive.24 They may also broaden support among the domestic population.25 Externally, suicide attacks are a proven strategy to weaken the groups opponent. Robert Pape, for example, argues that their high degree of lethality makes suicide attacks a rational or logical choice for organizations and states under certain circumstances, asserting that the main reason that suicide terrorism is growing is that terrorists have learned that it works.26 Pape, however, appears to have exaggerated the success rate he ascribes to suicide terrorism.27 Moreover, as Robert Brym and Bader Araj have noted, characterizing suicide missions as strategically ration-revenge theory, see Mark Juergensmeyer, Terror in the Mind of God: The Global Rise of Religious Violence (Berkeley: University of California Press, 2001). 21. Several authors have stressed the expectation of posthumous benets as a motive for suicide attackers, particularly when the perpetrators of the attacks are Muslims. Such benets can include the suicide attackers elevated social status after death, rewards for the family, as well as the attainment of heavenly pleasures in the afterlife. 22. Personal crisis appears to be a particularly common motivation among women suicide bombers such as the Chechen Black Widows. Anat Berko and Edna Erez, Ordinary People and Death Work: Palestinian Suicide Bombers as Victimizers and Victims, Violence and Victims, Vol. 20, No. 6 (December 2005), pp. 603623. 23. Martha Crenshaw, An Organizational Approach to the Analysis of Political Terrorism, Orbis, Vol. 29, No. 3 (Fall 1985), pp. 465489; and Martha Crenshaw, Theories of Terrorism: Instrumental and Organizational Approaches, in David C. Rapoport, ed., Inside Terrorist Organizations (London: Frank Cass, 1988). 24. On the organizational goals of survival and maintenance, see James Q. Wilson, Political Organizations (New York: Basic Books, 1973); and Crenshaw, An Organizational Approach to the Analysis of Political Terrorism. 25. Mia M. Bloom, Palestinian Suicide Bombing: Public Support, Market Share, and Outbidding, Political Science Quarterly, Vol. 119, No. 1 (Spring 2004), pp. 6188; and Bloom, Dying to Kill. 26. Pape, Dying to Win, p. 61. 27. See Moghadam, Suicide Terrorism, Occupation, and the Globalization of Martyrdom.

Contending with Terrorism 64

tional oversimplies the complexity of motivations at the organizational level. They nd that suicide bombings involve mixed rationales, including the urge for retaliation or the mere existence of opportunities to strike, which can prevail over purely strategic considerations.28 A consensus among scholars does exist, however, with regard to the tactical utility of suicide missions. Scholars agree that terrorist groups use suicide attacks because of the disproportionate amount of fear they create in the target population;29 their ability to boost the groups morale;30 and operational benets, such as their cost efciency and high precision, as well as the low security risks they pose to the organization at large.31 Organizational-strategic explanations have several limitations. They do not explain why, if the benets of suicide missions are so numerous, many organizations avoid their use;32 when a terrorist group is likely to engage in suicide attacks; or their dramatic rise since the turn of the millennium. The third major category of studies of suicide attacks argues that individuals and organizations will employ suicide terrorism if they are likely to enjoy social support for this tactic. This explanation appears to account for the widespread use of suicide attacks in places such as Israel and Lebanon, where a cult of martyrdom has manifested itself in the veneration of suicide bombers; in the prominent use of heroic and euphemistic labels for suicide attacks and their perpetrators; and in the penetration of the suicide bomber into popular culture, including movies, comics, and plays. Some researchers claim that sustained levels of suicide terrorism depend entirely on strong support among the
28. Robert J. Brym and Bader Araj, Suicide Bombing as Strategy and Interaction: The Case of the Second Intifada, Social Forces, Vol. 84, No. 4 (June 2006), pp. 19691986. 29. Although this is a key feature of all terrorist attacks, suicide attacks further demonstrate the inefcacy of the targeted government, given in part the demoralization of the public and of law enforcement agencies. In addition, the effect of a suicide attack can be particularly traumatizing and long-lasting. See, for example, Keith B. Richburg, Suicide Bomb Survivors Face Worlds Blown Apart, Washington Post, January 31, 2004; and Amos Harel, Suicide Attacks Frighten Israelis More Than Scuds, Haaretz, February 13, 2003. 30. Suicide attacks often lead to a sense of moral superiority of the groups members over their adversaries, which may result in a groups perception that it will eventually prevail over its enemies. See Adam Dolnik, Die and Let Die: Exploring Links between Suicide Terrorism and Terrorist Use of Chemical, Biological, Radiological, and Nuclear Weapons, Studies in Conict and Terrorism, Vol. 26, No. 1 (January 2003), pp. 1735. 31. See, for example, Sprinzak, Rational Fanatics; and Boaz Ganor, Suicide Attacks in Israel, in International Policy Institute for Counter-Terrorism (ICT), ed., Countering Suicide Terrorism (Herzliyya, Israel: ICT, 2001), pp. 140154. 32. A notable exception to these limitations is found in recent research conducted by Michael Horowitz, who employs adoption capacity theory to argue that groups with higher levels of organizational capital are more likely to adopt suicide terrorism than groups with lower levels of capital. See Michael Horowitz, Non-State Actors and the Diffusion of Innovations: The Case of Suicide Bombing, unpublished manuscript, Harvard University, July 2008.

Motives for Martyrdom 65

attackers domestic population.33 In recent years, however, an increasing number of suicide attacks have been carried out in countries where such domestic support appears to be lacking: examples include Afghanistan, Pakistan, and even Iraq.34 A culture of martyrdom may inuence the suicide bombers in these countries, but increasingly that culture seems to be found in cyberspace rather than in the streets.35 Another argument put forward by some scholars is that societies are more inclined to produce suicide bombers when they are subjected to foreign occupation. In the next section I examine the applicability of this occupation thesis as well as that of another explanation, the outbidding thesis, to the globalization of suicide attacks. occupation In his book Dying to Win, Pape argues that the bottom line is that suicide terrorism is mainly a response to foreign occupation.36 He denes an occupation as one in which a foreign power has the ability to control the local government independent of the wishes of the local community.37 There are three reasons, however, why foreign occupation does not explain many contemporary suicide missions.38 First, these attacks increasingly occur in countries where there is no discernible occupation, including Bangladesh, Indonesia, Jordan, Morocco, Pakistan, Saudi Arabia, the United Kingdom, the United States, Uzbekistan, and Yemen. Second, in foreign-occupied countries such as Iraq, the attacks are often not directed at the occupiers themselves who, according to the logic of the occupation thesis, should be the most obvious targets. Many suicide bombings in Iraq, for example, have targeted Kurds, Shiites, and Sus, in an effort to stir ethnic tensions in the country and delegitimize the Iraqi government in the eyes of Iraqis. Third, even if they do target the occupation forces, many suicide attacks are not carried out by those individuals most directly affected by the occupation. In Iraq, for instance, most attacks against occupation forces are carried out by foreign jihadis from places
33. Bloom, Dying to Kill. 34. See Moghadam, Suicide Terrorism, Occupation, and the Globalization of Martyrdom, pp. 707729. 35. See, for example, Gabriel Weimann, Terrorism on the Internet: The New Arena, The New Challenges (Washington, D.C.: United States Institute of Peace Press, 2006), pp. 6475; Marc Sageman, Leaderless Jihad: Terror Networks in the Twenty-First Century (Philadelphia: University of Pennsylvania Press, 2008), pp. 109123; and Moghadam, The Globalization of Martyrdom, chap. 4. 36. Pape, Dying to Win, p. 23. 37. Ibid., p. 46. 38. For a more extensive critique of Papes book Dying to Win, see Moghadam, Suicide Terrorism, Occupation, and the Globalization of Martyrdom.

Contending with Terrorism 66

such as Kuwait, Libya, Saudi Arabia, and Syria.39 The perpetrators of the September 11 attacks were from Egypt, Saudi Arabia, and the United Arab Emirates. Papes focus on occupation and his dismissal of religion or ideology as important variables are most striking in his discussion of al-Qaida. For alQaeda, religion matters, Pape writes, but mainly in the context of national resistance to foreign occupation.40 The evidence, however, does not support Papes argument. A closer reading of statements issued by al-Qaida leaders suggests that religion plays a more central role in the organizations ideology and mission than Pape would ascribe to it. Osama bin Laden and al-Qaida are engaged in a defensive jihad against what they portray as the CrusaderZionist alliance because they believe that the United States has made a clear declaration of war on God, his messenger, and Muslims.41 In a statement from November 2001, bin Laden declared, This war is fundamentally religious. . . . Under no circumstances should we forget this enmity between us and the indels. For, the enmity is based on creed.42 In another message, bin Laden went as far as urging Americans to convert to Islam: A message to the American people: Peace be upon those who follow the right path. . . . I urge you to become Muslims, for Islam calls for the principle of there is no God but Allah.43 Al-Qaidas understanding of occupation differs from that of Pape. Whereas Pape suggests that foreign occupation consists of boots on the groundor, as he puts it, the ability of a foreign power to control the local government independent of the wishes of the local community44al-Qaidas understanding

39. In February 2006, for example, Director of National Intelligence John Negroponte conrmed that extreme Sunni jihadist elements, a subset of which are foreign ghters, constitute a small minority of the overall insurgency, but their use of high-prole suicide attacks gives them a disproportionate impact. See Negroponte, Statement by the Director of National Intelligence to the Senate Select Committee on Intelligence, 109th Cong., 2d sess., February 2, 2006. As recently as May 2007, for instance, Gen. David Petraeus stated that 80 to 90 percent of the suicide bombers come from outside Iraq. Quoted in Joshua Partlow, An Uphill Battle to Stop Fighters at Border, Washington Post, May 5, 2007. 40. Pape, Dying to Win, p. 104. 41. Osama bin Laden, Text of Fatwa Urging Jihad against Americans, Al-Quds al-Arabi (London), February 23, 1998, quoted in Christopher M. Blanchard, Al-Qaida: Statements and Evolving Ideology, CRS Report for Congress (Washington, D.C.: Congressional Research Service, Library of Congress, November 16, 2004), Order Code RS21973, p. 3. 42. Osama bin Laden, speech broadcast on Al-Jazeera satellite channel television, November 3, 2001. Quoted in Bin Laden Rails against Crusaders and UN, BBC News, November 3, 2001. 43. Statement by Usama bin Ladin, Waqiaah, October 26, 2002, quoted in Anonymous [Michael Scheuer], Imperial Hubris: Why the West Is Losing the War on Terror (Washington, D.C.: Brasseys, 2004), p. 154. 44. Pape, Dying to Win, p. 46.

Motives for Martyrdom 67

of occupation is much broader. It includes a long history of injustices manifested today in the military, religious, political, economic, and cultural humiliation of the larger Muslim world by the Crusader-Zionist alliance. It is this ideologically inspired denition of occupation that matters most for al-Qaida but that is absent from Papes analysis. Pape also does not explain why most suicide missions are perpetrated by groups claiming to act in the name of religion, while attacks by secular organizations have declined in recent years. Although he correctly notes that modern suicide terrorism is not limited to Islamic fundamentalism,45 he does not acknowledge that most such attacks are perpetrated by radical Islamist groups. Overall, Pape calculates, Islamic fundamentalism is associated with about half of the suicide terrorist attacks that have occurred from 1980 to 2003.46 Not included in Papes count, however, is the high tally of suicide attacks that have occurred in Iraq since the 2003 U.S. invasiona total of 1,020 such attacks by March 2008. According to my data set, Sala-jihadist groups were the most dominant perpetrators of suicide missions in Iraq in the ve years since the U.S.-led invasion.47 outbidding Mia Bloom posits that terrorist groups may engage in suicide missions because they are trying to compete against other groups for the support of the local population. This tactic, known as outbidding, is designed to increase the groups market share among that community. The thesis thus assumes that suicide bombing campaigns depend on the support of the local population. In the war for public support, Bloom writes, when the bombings resonate positively with the population that insurgent groups purport to represent, they help the organization mobilize support. If suicide bombing does not resonate among the larger population, the tactic will fail.48 The outbidding thesis appears plausible as an explanation for the adoption of suicide missions by several organizations, including the Popular Front for

45. Ibid., p. 16. 46. Ibid., p. 17. 47. Of the 1,020 suicide attacks in Iraq recorded in the data set, 208 were claimed by Sala jihadist groups. The next most popular were nationalist-separatist groups, with 11 claimed attacks. Although the perpetrators of 794 attacks in Iraq are still unknown, anecdotal accounts suggest that the overwhelming number of all suicide attacks in Iraq are conducted by Sala jihadist groups. See, for example, Hafez, Suicide Bombers in Iraq; and International Crisis Group, In Their Own Words: Reading the Iraqi Insurgency, Middle East Report, No. 50 (Brussels: International Crisis Group, February 15, 2006). 48. Bloom, Dying to Kill, p. 78.

Contending with Terrorism 68

the Liberation of Palestine (PFLP),49 the Fatah al-Aqsa Martyrs Brigades,50 and Amal in Lebanon.51 The outbidding thesis, however, falls short of providing a satisfactory explanation for the adoption of this tactic in many other cases, including some noted by Bloom in Dying to Kill. For instance, it cannot explain why Sri Lankas Liberation Tigers of Tamil Eelam (LTTE) adopted this tactic only in 1987, at a time when the internal rivalry between radical Tamil organizations had reached its pinnacle with the May 1986 massacre of the Tamil Eelam Liberation Organization (TELO) and the killing of its leader.52 By that time, the LTTE had eliminated not only the TELO but all of its other rivals as well. Moreover, Blooms own ndings appear to contradict one of the central assumptions of the outbidding thesis, namely, that groups are vying for the support of the local population. Her own survey data suggest that suicide missions in Sri Lanka are not condoned among the Tamil population. Based on her interviews and polls of hundreds of Tamils all over Sri Lanka, she found that there was virtually no support for attacking civilians, regardless of whether they were in Sinhalese territory or in the Tamil regions.53 Despite this lack of support, the LTTE continued its relentless suicide bombing campaign (including attacks against civilians), apparently undeterred. The outbidding thesis ts within the traditional paradigm that sees suicide attacks as occurring in the context of long-standing historical conicts in which a large segment of the population supports the actions of suicide attackers as a legitimate form of resistance designed to achieve self-determination, or at least some degree of autonomy. That paradigm, however, is incompatible with the global jihad being waged by transnational groups such as al-Qaida that are seeking to achieve less dened goals and that are unwilling to compromise. It is for that reason that the notion that suicide attackers are vying for domestic popular support is most problematic with regard to al-Qaida and the global jihad movement. The London bombers of July 2005, for example, targeted their own fellow citizens. Similarly, a growing number of suicide missions being conducted in Iraq target Iraqis rather than the occupying forces. It is hard to argue that Iraqi
49. Bloom, Palestinian Suicide Bombing. 50. See assessment by Ely Karmon, quoted in Christopher Dickey et al., Inside Suicide, Inc., Newsweek, April 15, 2002, p. 26. 51. Martin Kramer, Sacrice and Fratricide in Shiite Lebanon, Terrorism and Political Violence, Vol. 3, No. 3 (Autumn 1991), pp. 3047. 52. Bloom apparently recognizes this problem and does not argue that outbidding has been a factor in the adoption of suicide attacks by the LTTE. See Bloom, Dying to Kill, p. 71. 53. Ibid., p. 67.

Motives for Martyrdom 69

suicide attackers are trying to gain the sympathy of the very people in whose midst they are blowing themselves up.

Al-Qaida, Sala Jihad, and Suicide Attacks

The main reason for the global spread of suicide missions lies in two related and mutually reinforcing phenomena: al-Qaidas transition into a global terrorist actor and the growing appeal of its guiding ideology, Sala jihad. This argument requires an explanation of two separate issues: rst, why and how al-Qaida became a global entity in both outlook and practice; and second, alQaidas emphasis on suicide missions as the primary method of terrorist operations. al-qaidas global outlook Three key factors inuenced al-Qaidas decision to globalize its operations, the rst being the groups core doctrine. As envisioned by Abdullah Azzam, bin Ladens mentor, al-Qaida was designed as the vanguard of an Islamic army similar to an international rapid reaction force that would come to the rescue of Muslims wherever and whenever they were in need. This Muslim legion would be self-perpetuating, generating new waves of Islamic warriors who would ght and defeat indel and apostate countries the world over. The second reason for the globalization of al-Qaida was the spread of the Afghan Arabs, the foreign ghters who ocked to Afghanistan after the Soviet invasion, to other countries beginning in 1988. After the Red Armys withdrawal from Afghanistan, many Afghan Arabs returned to their home countries, where they participated in local jihads against entrenched regimes in countries such as Egypt, Jordan, and Saudi Arabia. Others moved to third countries, including some in Western Europe. Realizing Azzams dream, many of these Afghan Arabs radicalized and mobilized Muslims in their countries. They regarded themselves as the vanguard that Azzam had foreseen, and many chose violence as their preferred tactic.54 Third, al-Qaida based its decision to globalize on a deliberate shift in strategy. Between 1995 and 1996, after heated internal discussions, al-Qaida decided not to attack the near enemy (i.e., the local Arab regimes it regarded as apostate), but the far enemy (i.e., Western indel countries, above all the United States).55 This shift in strategy was epitomized when, in 1996, al-Qaida
54. Jason Burke, Al-Qaeda: The True Story of Radical Islam (London: I.B. Tauris, 2004), p. 290. 55. Marc Sageman, Global Sala Jihad, statement to the National Commission on Terrorist At-

Contending with Terrorism 70

declared war on the United States, and again, two years later, when it announced the formation of a global alliance to defeat the Crusader-Zionist enemy. Al-Qaidas rst major suicide attack, the August 1998 bombings of the U.S. embassies in Nairobi, Kenya, and Dar-es-Salaam, Tanzania, embodied that strategic shift. al-qaida and the primacy of suicide attacks In al-Qaidas tactical arsenal, suicide attacks play a pivotal role. No other tactic symbolizes al-Qaidas tenaciousness and ability to inspire a large number of Muslims worldwide as much as martyrdom operations, to use the groups euphemistic labeling. Al-Qaida has all but perfected this tactic and institutionalized it to an extent not seen in other terrorist groups. It instilled the spirit of self-sacrice in the collective psyche of virtually all of its ghters, thus creating a cult of martyrdom that far exceeds the Palestinian and Lebanese cult of death in both scope and depth. Abdullah Azzam was the rst theoretician to succeed in turning martyrdom and self-sacrice into a formative ethos of future al-Qaida members. It is largely because of him that self-sacrice has become a moral code that alQaida has used to justify suicide missions against its enemies.56 More than any other individual, Azzam persuaded jihadis in Afghanistan and beyond that those who die for the sake of God ( sabil Allah) will be rewarded in paradise. Ironically, Azzam understood martyrdom not as involving suicide missions per se, but as the death of any true Muslim waging jihad. Such martyrdom would wash away the jihadis sins and bestow glory upon him. Death-obsessed Afghan Arabs were so deeply affected by Azzams thinking that they became a curious sideshow to the real ghting in Afghanistan, Lawrence Wright observed. When a ghter fell, his comrades would congratulate him and weep because they were not also slain in battle. These scenes struck other Muslims as bizarre. The Afghans were ghting for their country, not for Paradise or an idealized Islamic community.57 Al-Qaidas decision to engage in suicide attacks was also inuenced by the Egyptian group al-Jihad and its leader, Ayman al-Zawahiri. Years before Zawahiri and al-Jihad formally joined al-Qaida in 2001, the Egyptian organization had employed suicide missions as a terrorist tactic. In August 1993 a suitacks upon the United States, July 9, 2003; and Fawaz A. Gerges, The Far Enemy: Why Jihad Went Global (Cambridge: Cambridge University Press, 2005). 56. Reuven Paz, interview by author, Washington, D.C., July 17, 2006. 57. Lawrence Wright, The Looming Tower: Al-Qaeda and the Road to 9/11 (New York: Vintage, 2007), p. 125.

Motives for Martyrdom 71

cide bomber smashed his explosives-laden motorcycle into the car of Egypts interior minister, Hassan al-Al, who nevertheless survived the attack.58 On November 19, 1995, al-Jihad staged another attack, this one at the Egyptian embassy in Islamabad, Pakistan, involving two assailants, including one suicide bomber. Sixteen people were killed. In interviews with Fawaz Gerges, former jihadis conrmed that Zawahiris advocacy of suicide bombings fundamentally inuenced bin Ladens adoption of this tactic. The spectacular nature of al-Qaidas suicide attacks, they told him, were adopted from al-Jihad, which had always used extremely lethal and psychologically damaging attacks to differentiate itself from its jihadist rival in Egypt, the al-Gamaa al-Islamiyya.59 When pressed to explain the use of suicide bombersa tactic still considered taboo, especially when used against fellow MuslimsZawahiri stated that these martyrs represented a generation of mujahideen that has decided to sacrice itself and its property in the cause of God. That is because the way of death and martyrdom is a weapon that tyrants and their helpers, who worship their salaries instead of God, do not have.60 Zawahiri made a claim that many other supporters of suicide attacks would repeat: the suicide attacker does not kill himself for personal reasons, but sacrices himself for God. He is therefore not committing suicide, but achieving martyrdom. It was a game of words, but it provided justication for hundreds of future suicide bombers to emulate these early shuhada. In August 1996, bin Laden formally declared war against the United States, imploring Muslim youths to sacrice themselves. Seeking religious justication, bin Laden ties the longing for martyrdom to verses from the Quran, hadith, and poems. According to bin Laden, Our youths believe in paradise after death. They believe that taking part in ghting will not bring their day nearer, and staying behind will not postpone their day either. Exalted be to Allah who said: And a soul will not die but with the permission of Allah, the term is xed (Aal Imraan: 3:145). . . . Our youths took note of the meaning of the poetic verse: If death is a predetermined must, then it is a shame to die cowardly. Bin Laden then highlights a number of sayings that together describe the rewards of the martyr in Paradise: Allah, the Exalted, also said: And do not speak of those who are slain in Allahs way as dead; nay, they are alive, but you do not perceive (Bagarah;
58. The group al-Jihad is also known as Egyptian Islamic Jihad. Some authors dispute that alJihad was responsible for the attempt on al-Als life and blame the al-Gamaa al-Islamiyya instead. 59. Gerges, The Far Enemy, pp. 142143. 60. Quoted in Wright, The Looming Tower, pp. 248249.

Contending with Terrorism 72

2:154). . . . And: A martyr will not feel the pain of death except like how you feel when you are pinched (Saheeh Al-Jame As-Sagheer). He also said: A martyrs privileges are guaranteed by Allah; forgiveness with the rst gush of his blood, he will be shown his seat in paradise, he will be decorated with the jewels of belief (Imaan), married off to the beautiful ones, protected from the test in the grave, assured security in the day of judgement, crowned with the crown of dignity, a ruby of which is better than this whole world (Duniah) and its entire content, wedded to seventy-two of the pure Houries (beautiful ones of Paradise) and his intercession on the behalf of seventy of his relatives will be accepted. Bin Laden goes on to praise the courage of youths willing to sacrice themselves, suggesting that through death, young Muslims will prevail in the struggle against the Crusaders: Those youths know that their rewards in ghting you, the USA, is double than their rewards in ghting someone else not from the people of the book. They have no intention except to enter paradise by killing you.61 Al-Qaidas emphasis on suicide missions was on display in its Afghan training camps. A document found in an al-Qaida safe house in Afghanistan titled Goals and Objectives of Jihad, for example, ranked the goal of attaining martyrdom in the cause of God second only to establishing the rule of God on earth. Another document listed two illegitimate excuses for leaving jihad as love of the world and hatred of death.62 Bin Laden sought to spread the virtues of martyrdom through videotapes and statements on the internet. In 2004, for instance, he urged his followers to become diligent in carrying out martyrdom operations; these operations, praise be to God, have become a great source of terror for the enemy. . . . These are the most important operations.63 The use of suicide attacks is also a logical outcome of al-Qaidas desire to maximize the pain and suffering of its enemies in a protracted struggle. In his 2001 book, Knights under the Prophets Banner, Zawahiri writes, If our goal is comprehensive change and if our path, as the Koran and our history have shown us, is a long road of jihad and sacrices, we must not despair of repeated strikes and recurring calamities.64 He adds that there is a need within
61. Bin Ladens Fatwa, PBS Online Newshour, August 8, 1996, http:/ /www.pbs.org/newshour/ terrorism/international/fatwa_1996.html. 62. C.J. Chivers and David Rhode, Turning Out Guerrillas and Terrorists to Wage a Holy War, New York Times, March 18, 2002. 63. Quoted in Blanchard, Al-Qaida: Statements and Evolving Ideology, p. 10. 64. Ayman al-Zawahiri, Knights under the Prophets Banner (London: Al-Sharq al-Awsat, 2001), part 11. The book was serialized in the London-based magazine Al-Sharq al-Awsat between December 2 and December 10, 2001, and translated by the Foreign Broadcast Information Service (FBIS), FBISNES-2001-1202.

Motives for Martyrdom 73

the jihadist movement to offset the power of the Muslims enemies, whose numbers and capabilities have risen tremendously, as did the quality of their weapons, their destructive powers, their disregard for all taboos, and disrespect for the customs of wars and conict.65 To address this asymmetry, Zawahiri suggests a number of steps, including concentrat[ing] on the method of martyrdom operations as the most successful way of inicting damage against the opponent and the least costly to the mujahidin in terms of casualties.66 sala jihad and suicide attacks The Sala jihad is a radical offshoot movement with roots in a broader Islamist trend known as Salasm, as well as in Wahhabism and Qutbist factions of the Muslim Brotherhood.67 Salas have adopted a strict interpretation of Islamic religious law, and their doctrine centers around a more literal understanding of the concept of tawhid (the unity of God) than does that of ordinary Muslims. For Salas, the unity of Goda concept adhered to by all Muslimsextends to the belief that all man-made laws must be rejected because they interfere with the word and will of God. Salas reject the division of religion and state and believe that only the salafthe Prophet himself and his companionsled lives in accordance with Gods will. Only by emulating that lifestyle can Muslims reverse the decline of Islam. Whereas ordinary Salas believe that Gods word should be spread by dawa alonethe nonviolent call to Islam by proselytizingSala jihadists advocate waging violent jihad. This advocacy of violence leads to four main points of contention between the two groups: unlike Salas, Sala jihadists elevate jihad to the same level as the ve pillars of Islam; they engage in takr, the process of labeling fellow Muslims as indels (kufr), thus justifying violence against them; they condone the targeting of civilians; and they support the use of suicide operations.68 Sala jihadists believe that suicide operations against indels and apostates (i.e., non-Muslim heretics and nominally Muslim traitors) represent the ultimate form of devotion to God and the optimal way to wage jihad. They present jihad and self-sacrice as the antithesis to everything the West stands

65. Ibid. 66. Ibid. 67. Wahhabism is a puritanical strand of Islam closely related to Salasm, which is common in Saudi Arabia. 68. Quintan Wiktorowicz, Anatomy of the Sala Movement, Studies in Conict and Terrorism, Vol. 29, No. 3 (May 2006), pp. 207239.

Contending with Terrorism 74

forhence the mantra, The West loves life, while true Muslims love death. In the words of Abu Ayman al-Hilali, a key interpreter of bin Ladens ideas on Sala jihadist websites, First we have to acknowledge a basic fact, proved by experience and reality, already acknowledged by the enemy, which is that the vital contradiction to the Zionist and American enemy is the doctrine of Jihad and Martyrdom (Istishhad).69 In certain mosques, Sala jihadist preachers such as Abu Hamza al-Mazri and Omar Bakri Muhammed led thousands of Muslim youths to develop a cult-like fascination with martyrdom. Other preachers are active mainly on the internet, providing legitimation for martyrdom operations. Because Islam forbids the taking of ones own life, Sala jihadists draw a conceptual distinction between suicide and martyrdom, arguing that those committing ordinary suicide do so for personal reasons, such as distress or depression; in contrast, martyrs die primarily for the sake of God, but also for the greater good of the Muslim community. Although statistical evidence for the growth of Sala jihad is scant, there is ample anecdotal evidence of its increasing popularity among both men and women70 in general,71 and in Europe,72 the Middle East,73 Central Asia,74 Southeast Asia,75 and Africa76 in particular. In this study, the rise of the Sala jihad and its growing inuence on suicide missions are examined through a coding of the fty groups that employed them from December 1981 through March 2008 as part of their guiding ideol69. Cited in Reuven Paz, Qaidat Al-Jihad: A New Name on the Road to Palestine (International Policy Institute for Counter-Terrorism, May 7, 2002), http:/ /www.global-report.co.il/prism/ ?l en&a 4959. 70. On women, see, for example, Sebastian Rotella, European Women Join Ranks of Jihadis, Los Angeles Times, January 10, 2006; and Katharina von Knop, The Female Jihad: Al-Qaidas Women, Studies in Conict and Terrorism, Vol. 30, No. 5 (May 2007), pp. 397414. 71. See, for example, Gilles Kepel, Jihad: The Trail of Political Islam (Cambridge, Mass.: Harvard University Press, 2002); and Olivier Roy, Globalized Islam: The Search for a New Ummah (New York: Columbia University Press, 2004). 72. Kathryn Haahr, Emerging Terrorist Trends in Spains Moroccan Communities, Terrorism Monitor, Vol. 4, No. 9 (May 4, 2006), pp. 12; Pascale Combelles Siegel, Radical Islam and the French Muslim Prison Population, Terrorism Monitor, Vol. 4, No. 15 (July 27, 2006), pp. 12; and Lorenzo Vidino, The Danger of Homegrown Terrorism to Scandinavia, Terrorism Monitor, Vol. 4, No. 20 (October 19, 2006), pp. 56. 73. On the Middle East, see, for example, Marc Sageman, Understanding Terror Networks (Philadelphia: University of Pennsylvania Press, 2004). 74. See, for example, Anar Valiyev, The Rise of Sala Islam in Azerbaijan, Terrorism Monitor, Vol. 3, No. 13 (July 1, 2005), pp. 67. 75. See International Crisis Group, Indonesia Backgrounder: Why Salasm and Terrorism Mostly Dont Mix, Asia Report, No. 83 (Brussels: International Crisis Group, September 13, 2004). 76. Douglas Farah, Salasts, China, and West Africas Growing Anarchy (Alexandria, Va.: International Assessment and Strategy Center, December 7, 2004).

Motives for Martyrdom 75

ogy or doctrine (see table 1). In all, 1,857 such attacks took place. To be coded as Sala jihadist, a group must be a Sunni Islamic group to which at least one of the following characteristics must also apply: (1) afliation with and/or adherence to al-Qaida was reected in the groups name;77 (2) the group had internalized the worldview of al-Qaida and global jihad;78 (3) the group engaged in violence to overthrow an Islamic regime and create a transnational caliphate in its stead;79 or (4) the group engaged in the labeling of some other Muslims as heretics.80 My analysis of the data yields the following ndings: of the 788 suicide attacks from December 1981 to March 2008 in which the identity of the group could be identied, Sala jihadist groups carried out 37.7 percentmore than any other group. They were followed by nationalist-separatist groups with 18.5 percent and hybrid groups with 17.8 percent. One thousand sixty-nine attacks (57.6 percent) were perpetrated by organizations whose identities remain unknown. Of these, however, 795 (74.4 percent) occurred in Iraq, where the vast majority of organizations conducting suicide bombings are known to be Sala jihadist.81 The bulk of suicide missions in the unknown category were therefore likely carried out by Sala jihadist groups, too. More important, according to the following criteria, Sala jihadist groups have assumed the leadership among groups that employ this modus operandi: number of suicide attacks, number of organizations engaged in suicide attacks, total number of fatalities, and average number killed per attack. In 1997, for example, none of the groups that undertook suicide missions were Sala jihadist. In 1998 a quarter of the groups that employed this tactic adhered to Sala jihadist ideology. After 2004, at least half of all groups conducting suicide missions adhered to Sala jihadist ideology in every given year. The growing ascendancy of Sala jihadist groups among groups employing suicide attacks is paralleled by the relative decline in the importance of groups

77. An example would be the group al-Qaida in Iraq. 78. This assessment is based on the best judgment of a group of eight terrorism experts at the RAND Corporation. See Angel Rabasa, Peter Chalk, Kim Cragin, Sara A. Daly, Heather Gregg, Theodore W. Krasik, Kevin A. OBrien, and William Rosenau, Beyond Al-Qaeda, Part 1: The Global Jihadist Movement, and Part 2: The Outer Rings of the Terrorist Universe (Santa Monica, Calif.: RAND, 2006). 79. This excludes groups such as Hamas, which engages primarily in violence against Israel, a non-Muslim state, but has generally avoided systematic attacks against the Palestinian Authority (prior to Hamass electoral victory in 2006) for fear of sparking a civil war. 80. Takr is not generally practiced by mainstream Islamist groups and not even by all Sala jihadists. Those groups and individuals who do practice this form of excommunication, however, are exclusively Sala jihadist. 81. See International Crisis Group, In Their Own Words; and Hafez, Suicide Bombers in Iraq.

Contending with Terrorism 76

Table 1. Ideological Affiliation of Groups Conducting Suicide Attacks, December 1981 through March 2008 Organizations Al Al Jihad Al-Gamaa al-Islamiya3 Al-Qaida Al-Qaida in Iraq4 Al-Qaida in the Islamic Magreb5 Amal Ansar al-Islam6 Ansar Allah Ansar al-Sunnah7 Armed Islamic Group8 As-Sirat al-Moustaquim9 Chechen Separatists Chechen SeparatistsArbi Barayev10 Chechen SeparatistsKarachaev Jamaat11 Chechen SeparatistsRamzan Akhmadov12 Fatah Hamas13 Hezballah Hizb-i-Islami14 Hizb-ul Mujahideen15 Hizb-ut-Tahrir16 Islamic Army in Iraq17 Islamic Courts Union18 Islamic Jihad of Uzbekistan19 Islamic Movement of Uzbekistan20 Islamic State of Iraq21 Jaish-e-Muhammad22 Jamatul Mujahedin Bangladesh23 Dawa1 2 Ideology SH SJ SJ SJ SJ SJ SH SJ SH SJ SJ SJ H H U H NS MI/NS SH SJ SJ SJ MI/NS H SJ SJ SJ SJ SJ Organizations Jemaah Islamiyya Jund al-Sham (Army of the Levant)25 Kashmir Separatists26 Kurdistan Workers Party Lashkar-e-Jhangvi27 Lashkar-e-Taibeh28 Lebanese Liberation Organization29 Liberation Tigers of Tamil Eelam Mujahideen Army30 Mujahideen Shura Council31 Mujahideen Youth Movement32 Palestinian Islamic Jihad Partisans of the Sunni Popular Front for the Liberation of Palestine Popular Resistance Committees Qari Zafar Group33 Revolutionary Peoples Liberation Party Riyad us-Saliheyn Martyrs Brigade34 Saddam Loyalists Shields of Islam Soldiers of the Prophets Companion Syrian Baath Organization Syrian Social Nationalist Party Taliban Tawhid wal Jihad35 Tehrik-i-Taliban36 Victory and Jihad in Greater Syria37
24

Ideology SJ SJ H M/NS SJ SJ U NS H SJ SJ MI/NS U M/NS MI/NS SJ M H NS U SJ NS M/NS H SJ H U

SOURCES: The following sources were used to ascertain the ideological or doctrinal orientation

of groups that employ suicide terrorism: The Terrorism Knowledge Base (TKB) of the Memorial Institute for the Prevention of Terrorism (MIPT); the 2005 Country Reports on Terrorism, published by the U.S. Department of States Office of the Coordinator for Counterterrorism; Angel Rabasa, Peter Chalk, Kim Cragin, Sara A. Daly, Heather Gregg, Theodore W. Krasik, Kevin A. OBrien, and William Rosenau, Beyond Al-Qaeda, Part 1: The Global Jihadist Movement, and Part 2: The Outer Rings of the Terrorist Universe (Santa Monica, Calif.: RAND, 2006); and anecdotal information, when the first three sources did not provide sufficient information to establish ideological identity. The TKB integrates data from the RAND Terrorism Chronology and RAND-MIPT Terrorism Incident databases; the Terrorism Indictment database; and DFI Internationals research on terrorist organizations. On March 21, 2008, the MIPT announced via email to subscribers of the TKB newsletter that the TKB would cease operations on March 31, 2008, and elements of the system will be merged with the Global Terrorism Database, managed by the National Consortium for the Study of Terrorism and Responses to Terrorism (START) at the University of Mary-

Motives for Martyrdom 77

Table 1. (Continued)

land. The TKB website (http://www.tkb.org) is not longer accessible. The RAND studies detail information on the ideological affiliation of jihadist groups. The assessments made in the two RAND studies are based on the consensus agreement of eight RAND terrorism experts. NOTE: Groups are coded as hybrid (H), Marxist (M), mainstream Islamist (MI), nationalistseparatist (NS), Shiite (SH), Salafi jihadist (SJ), unknown (U) groups, as well as in combinations (e.g., MI/NS or M/NS). Hybrid organizations comprise members who have adopted a Salafi jihadist ideology as well as those who seem to be motivated primarily by ethnonationalist and separatist concerns. Groups coded as mainstream Islamist groups, such as Hamas and Palestinian Islamic Jihad, differ from Salafi jihadist groups in that they participate in the political processsomething that Salafi jihadist groups consider heretical, given that all power must derive from God, not from the electorate. In addition, mainstream Islamist groups do not engage in takfirthe labeling of other Muslims as kufr, or heretics. Salafi jihadists, all of whom are Sunnis, consider the Shiite stream of Islam to be heretical. 1 Al-Dawa conducted at least three suicide attacks in Kuwait between 1983 and 1985. According to the Terrorism Knowledge Base (TKB), al-Dawa is a Shiite organization. 2 Al-Jihad was led by Ayman al-Zawahiri. The group formally merged with al-Qaida in June 2001. 3 Although the al-Gamaa al-Islamiyya has recently distanced itself from al-Qaida, it was one of the prominent Salafi jihadist groups during the 1990s and a major influence on al-Qaida. 4 Al-Qaida in Iraq is a franchise of al-Qaida. The group was formerly known as Tawhid wal Jihad. 5 Al-Qaida in the Islamic Maghreb is a franchise of al-Qaida. The group was formerly known as the Salafist Group for Call and Combat. 6 Ansar al-Islam seeks the violent overthrow of Iraqi Kurdistan and conversion of the province into an Islamist state. A faction of Ansar al-Islam split off and established Tawhid wal Jihad. According to GlobalSecurity.org, Ansar al-Islam has close links to and support from al-Qaida. Al-Qaida and Usama Bin Laden participated in the formation and funding of the group, which has provided safe haven to al-Qaida in northeastern Iraq. See Ansar al Islam (Supporters of Islam), GlobalSecurity.org, http://www.globalsecurity.org/military/world/ para/ansar_al_islam.htm. 7 According to the TKB, Ansar al-Sunnah (Followers of the Tradition) is an Iraqi jihadist group dedicated to the establishment of an Islamic state in Iraq based on sharia law. It seeks to achieve this objective by defeating coalition forces and the foreign occupation. Taken from Group Profile: Ansar Al-Sunnah Army, Terrorism Knowledge Base. On Ansar al-Sunnahs Salafi roots, see also Mohammed M. Hafez, Suicide Terrorism in Iraq: A Preliminary Assessment of the Quantitative Data and Documentary Evidence, Studies in Conflict and Terrorism, Vol. 29, No. 6 (September 2006), pp. 591619; International Crisis Group, In Their Own Words: Reading the Iraqi Insurgency, Middle East Report, No. 50 (Brussels: International Crisis Group, February 15, 2006); and Michael Eisenstadt and Jeffrey White, Assessing Iraqs Sunni Arab Insurgency, Policy Focus, No. 50 (Washington, D.C.: Washington Institute for Near East Policy, December 2005), pp. 139. 8 The Armed Islamic Group engages in takfir. See Rabasa et al., Beyond Al-Qaeda, Part 2, p. 28. 9 As-Sirat al-Moustaquim is the cell responsible for the 2003 Casablanca bombings. It is sometimes referred to as Salafia Jihadiya. A panel of eight experts at the RAND Corporation concluded that Salafia Jihadiya has internalized the Al-Qaida worldview of Global Jihad. See Rabasa et al., Beyond Al-Qaeda, Part 1, pp. xxii, 2, 79. Information also taken from Group Profile: Salafia Jihadia, Terrorism Knowledge Base. There are persistent reports that As-Sirat al-Moustaquim/Salafia Jihadiya does not exist as a group, but is a term

Contending with Terrorism 78

Table 1. (Continued) invented by Moroccan authorities. There is little doubt, however, that the cell responsible for the bombings was driven by Salafi jihadist ideology. For a discussion of the controversy over the existence of Salafia Jihadiya, see Thomas Renard, Moroccan Crackdown on Salafiya Jihadiya Recruitment of Foreign Fighters for Iraq, Terrorism Focus, Vol. 5, No. 27 (July 23, 2008), p. 5. 10 According to the Haifa database, on June 7, 2000, a truck bomb driven by two Chechen suicide bombers (one male and one female) exploded in Alkhan Yurt, targeting an OMON (Special Forces Police) unit. Chechen separatists under the leadership of Arbi Barayev, who are also known as the Special Purpose Islamic Regiment (SPIR), claimed responsibility. According to the Terrorism Knowledge Base, the primary objective of SPIR is the liberation of Chechnya and the formation of an independent Chechen state. However, the Islamic fighters also promoted a more radical strain of Islam and a desire to install a fundamentalist Islamic republic governed by Sharia law in Chechnya. Taken from Group Profile: Special Purpose Islamic Regiment (SPIR), Terrorism Knowledge Base. 11 Chechen separatists under the leadership of Karachaev sent three suicide bombers to detonate themselves in coordinated attacks near the Russian border with Chechnya. They killed 20 and wounded about 140. The purpose of the attacks remains unknown. No additional information was available about the group. 12 The Chechen separatist group led by Ramzan Akhmadov was responsible for six suicide attacks in June and July 2000. Although information is to determine whether the group is Salafi jihadist in character is insufficient, Akhmadov was known as a radical Islamist and was likely influenced by Wahhabism. Like other Chechen terrorist groups, this group likely comprised both Salafi jihadists as well as more nationalist elements. 13 Although Hamass origins extend to the Muslim Brotherhood, its primary goal is the elimination of Israel. Hamas has resisted the adoption of al-Qaidas doctrine of global jihad, and it does not engage in takfir. Its unwillingness to adopt al-Qaidas worldview of global jihad has elicited several heated exchanges between al-Qaidas deputy leader, Ayman alZawahiri, and the Hamas leadership. Zawahiri has appealed to Hamasand to the Muslim Brotherhood at largenot to participate in the democratic process because he and other Salafi jihadists believe that power derived from the electorate rather than from God is heretical. For additional information, see Reuven Paz, The Islamic Debate over Democracy: Jihadi-Salafi Responses to Hamas Victory in the Palestinian Elections, Project for the Research of Islamist Movements (PRISM), Occasional Papers, Vol. 4, No. 1 (Herzliya, Israel: PRISM, January 2006); and Stephen Ulph, Al Zawahiri Takes Hamas to Task, Terrorism Focus, Vol. 3, No. 9 (March 7, 2006), p. 1. 14 Hizb-i-Islami is led by Gulbuddin Hekmatyar, who is believed to be allied with Osama bin Laden. Its goal is the end of occupation in Afghanistan and the establishment of an Islamist state there. See Hizb-i Islami Gulbuddin, GlobalSecurity.org, http://www.globalsecurity .org/security/profiles/hizb-i_islami_gulbuddin.htm/. 15 Although Hizb-ul Mujahideen has ties with Salafi jihadist groups, including Lashkar-eTaibeh, its primary focus is the liberation of Kashmir and its accession to Pakistan. Taken from Group Profile: Hizbul Mujahideen (HM), Terrorism Knowledge Base. It is also tied to Jamaat-i-Islami, the mainstream Islamist party, and the equivalent of the Muslim Brotherhood in Pakistan. 16 Hizb ut-Tahrir was responsible for a suicide attack at the entrance to a childrens clothing store in the local market in Tashkent, Uzbekistan, on March, 29, 2004. According to GlobalSecurity.org, Hizb ut-Tahrir al-Islami (Islamic Party of Liberation) is a radical Islamic political movement that seeks implementation of pure Islamic doctrine and the creation of an Islamic caliphate in Central Asia. . . . Its basic aim was struggle with infidels and the organization of a universal caliphate embracing all Islamic countries. . . . The political struggle is manifested in the struggle against the disbelieving imperialists, to deliver the Ummah from their domination and to liberate her from their influence by uprooting their intellectual,

Motives for Martyrdom 79

Table 1. (Continued) cultural, political, economic and military roots from all of the Muslim countries. See Hizb ut-Tahrir Al-Islami (Islamic Party of Liberation), GlobalSecurity.org, http://www .globalsecurity.org/military/world/para/hizb-ut-tahrir.htm. 17 The TKB does not provide sufficient information to determine whether the Islamic Army in Iraq is a Salafi jihadist organization. According to Mohammed Hafez, the Islamic Army in Iraq is both nationalist and Islamist. Hafez, Suicide Bombers in Iraq: The Strategy and Ideology of Martyrdom (Washington, D.C.: United States Institute of Peace Press, 2007). 18 The Somali Islamic Courts Union consists of members with a mainstream Islamist orientation as well as Salafi jihadists with suspected ties to al-Qaida. The Supreme Islamic Courts Union/al-Ittihad Mahakem al-Islamiya (ICU), GlobalSecurity.org, http://www.globalsecurity .org/military/world/para/icu.htm/. 19 The Islamic Jihad of Uzbekistan openly declared its participation in the global jihad and is a splinter group of the Salafi jihadist Islamic Movement of Uzbekistan. See Islamic Jihad Group of Uzbekistan, GlobalSecurity.org, http://www.globalsecurity.org/security/profiles/ islamic_jihad_group_of_uzbekistan.htm/. 20 The Islamist Movement of Uzbekistan has internalized the Al-Qaida worldview of Global Jihad. Rabasa et al., Beyond Al-Qaeda, Part 1, pp. xxii, 2, 79. 21 The Islamic State of Iraq is the successor organization to the Salafi jihadist Mujahideen Shura Council and is dominated by al-Qaida in Iraq. 22 Jaish-e-Muhammad has internalized the al-Qaida worldview of Global Jihad. Rabasa et al., Beyond Al-Qaeda, Part 1, pp. xxii, 2, 79. 23 According to the TKB, Jamatul Mujahedin Bangladesh (JMB) is a terrorist group dedicated to removing the countrys secular government and imposing a Taliban-inspired Islamic theocracy in its place. In addition to calling for an Islamic state based on Sharia law, JMB has denounced the U.S.-led invasion of Iraq, warning President [George W.] Bush and British Prime Minister [Tony] Blair to leave all Muslim countries. Taken from Group Profile: Jamatul Mujahedin Bangladesh (JMB), Terrorism Knowledge Base. 24 Jemaah Islamiyya has internalized the Al-Qaida worldview of Global Jihad. Rabasa et al., Beyond Al-Qaeda, Part 1, pp. xxii, 2, 79. 25 A group named the Army of the Levant (Jund al-Sham) took responsibility for the March 19, 2005, bombing of a theater in Doha, Qatar, near a British school and popular among Westerners. The name has been claimed by several Sunni Islamic extremist entities, all or none of which may be linked. According to TKB, all of the Jund al-Sham entities desire to achieve the unified purpose of replacing what they view as misguided forms of Islam and governmental rule with their vision of a traditional Islamic caliphate extending across the Levant. . . . Like many second- and third-tier Islamic extremist entities, the Jund al-Sham organizations are believed to be incorporated, however loosely, under the greater al-Qaeda umbrella. Taken from Group Profile: Jund Al-Sham, Terrorism Knowledge Base. 26 Lashkar-e-Taibeh, Jaish-e-Muhammad, and Harkat-ul-Mujahideen, the three major Kashmir separatist groups, have internalized the Al-Qaida worldview of Global Jihad. Rabasa et al, Beyond Al-Qaeda, Part 1, pp. xxii, 2, 79. 27 Lashkar-e-Jhangvi has internalized the Al-Qaida worldview of Global Jihad. Ibid. 28 Lashkar-e-Taibeh has internalized the Al-Qaida worldview of Global Jihad. Ibid. 29 On November 11, 1987, a female suicide bomber detonated 12 pounds of explosives packed in a briefcase at Beirut airport. On November 14, 1987, a female suicide bomber detonated 2 pounds of explosives connected to a nail-filled grenade concealed in a box of chocolates in the lobby of the American University Hospital in West Beirut. The Lebanese Liberation Army claimed responsibility, but no solid information on the ideology of this group is available. Given the use of female suicide bombers, however, it is unlikely that this was a Salafi jihadist group.

Contending with Terrorism 80

Table 1. (Continued)
30

The Mujahideen Army, or Jaish-e Mujahideen, is associated with Al-Qaida but has recently distanced itself from al-Qaidas indiscriminatory killing. See Jaysh al-Mujahideen Terrorist Lieutenant and Propaganda Chief Captured, GlobalSecurity.org, http://www .globalsecurity.org/military/library/news/2005/11/mil-051124-mnfi02.htm/. According to Evan F. Kohlmann, the Mujahideen Army has wavered back and forth in its stated jihadist political platform. See Evan F. Kohlmann, State of the Sunni Insurgency in Iraq: August 2007 (Washington, D.C.: NEFA Foundation, July 20, 2008), http://www.nefafoundation .org/miscellaneous/iraqreport0807.pdf/. 31 The Mujahideen Shura Council was the primary Salafi jihadist grouping in Iraq until it was renamed the Islamic State of Iraq on October 15, 2006. Its stated goal was to manage the struggle in the battle of confrontation to ward off the invading infidels and their apostate stooges. Taken from Group Profile: Mujahideen Shura Council, Terrorism Knowledge Base. 32 There is a high likelihood that the Mujahideen Youth Movement is directly influenced by alQaida. See Andrew Black, Somalias Mujahideen Youth Movement, Terrorism Focus, Vol. 4, No. 19 (June 19, 2007), p. 2. The groups founder and leader, Adan Hashi Ayro, has close al-Qaida connections. See Alisha Ryu, Youth Mujahideen Group Leading Attacks in Somalia, GlobalSecurity.org, June 19, 2007, http://www.globalsecurity.org/military/library/news/2007/06/mil-070619-voa03.htm/. 33 Qari Zafar is a splinter group of the Salafi jihadist Lashkar-e Jhangvi and affiliated with alQaida. See Abbas Naqvi, They Were Targeting Pakistan and Its Leaders, Daily Times, February 18, 2007. 34 According to the TKB, The Riyad us-Saliheyn Martyrs Brigade is a relatively young terrorist organization, dedicated to the creation of an independent Islamic republic in Chechnya (and other primarily Muslim parts of Russia such as Dagestan, Kabardino-Balkaria, Ingushetia, Ossetia, and Tataria). The group, whose name translates to Requirements for Getting into Paradise, espouses radical Islamic doctrine (Wahabbism) and is believed to have strong ties to al-Qaida. Most experts, however, agree that the primary inspiration behind Riyads activities is a desire for the independence of Chechen lands, rather than religious zealotry. Taken from Group Profile: Riyad Us-Saliheyn Martyrs Brigade, Terrorism Knowledge Base. 35 Tawhid wal Jihad is the forerunner of al-Qaida in Iraq. 36 Led by Beitullah Mehsud, the Tehrik-i-Taliban, also known as the Pakistani Taliban, is dedicated to enforcing sharia law, fighting NATO, and conducting defensive jihad against the Pakistani government. Its leader is said to have close ties to al-Qaida. See Hassan Abbas, A Profile of Tehrik-i Taliban, CTC Sentinel, Vol. 1, No. 2 (January 2008), pp. 14. 37 This previously unknown group claimed responsibility for the killing of Lebanese Prime Minister Rafiq Hariri on February 14, 2005. At the time of this writing, no additional information about this group was available. The investigation into the killing of Hariri was ongoing. For information on the attack being most likely a suicide attack, see UN Probe into Murder of Former Lebanese Leader Nears Sensitive StageInquiry Chief, United Nations News Centre, December 18, 2006.

Motives for Martyrdom 81

Figure 3. Number of Attacks by Ideology, December 1981March 2008

*Includes Marxist/nationalist-separatist groups. **Includes hybrid groups.

adhering to other ideologies. Figure 3 illustrates the rise in the number of attacks by Sala jihadist groups and the concomitant decline of attacks by groups guided by other ideologies.82 During the 1980s, Shiite groups were the main perpetrators of suicide missions,83 followed by groups with a nationalistseparatist agenda. During the 1990s, nationalist-separatist groups were the most frequent users of suicide attacks, followed by groups with a combination of a mainstream Islamist and nationalist-separatist agenda. Since the turn of the millennium, suicide attacks by Sala jihadist groups have become more common than attacks by all other groups. The steep rise in this number is par-

82. Attacks by unknown groups, whose ideology could not be ascertained, are omitted from this table. 83. According to other databases, Shiite groups such as Hezbollah and Amal conducted fewer attacks in Lebanon than secular or Sunni groups. See, for example, Pape, Dying to Win. Indeed, the identity of many groups that conducted suicide attacks in Lebanon during the 1980s were not identied in the data set used here. It is therefore possible that Shiite groups may not have been the dominant perpetrators of suicide attacks during the 1980s.

Contending with Terrorism 82

Figure 4. Lethality of Suicide Missions by Ideology, December 1981March 2008

ticularly worrisome given that their attacks were much more lethal than those by non-Sala jihadist groups (see gure 4).

From Localized to Globalized Patterns of Suicide Attacks

Al-Qaida and its Sala jihadist ideology have produced an altogether new pattern of suicide attacks, namely, globalized suicide missions, which can be distinguished from localized suicide missions, the more traditional pattern of suicide attacks. Localized and globalized patterns of suicide missions differ in ve key areas: the types of conicts in which these attacks are used; group ideology; the geographic scope of these actors; their target denition; and their goals (see table 2). localized suicide attacks The overwhelming majority of suicide missions during the 1980s and 1990s occurred in relatively localized settings. conict type. Suicide missions have traditionally occurred in the context of relatively localized conicts between two belligerents. Examples include conicts between Israel and Hezbollah, Israel and the Palestinians, Tamils and Sinhalese, and Turks and Kurds. These conicts have generally endured for many years, and often decades, between ethnic/religious groups.

Motives for Martyrdom 83

Table 2. Patterns of Suicide Missions Localized Conflict Ideology Actors Target definition Goals Examples Identifiable; long-standing Religious; ethno-nationalist; secular Subnational Narrow Limited Hezbollah, Liberation Tigers of Tamil Eelam, Kurdistan Workers Party, Hamas, Palestinian Islamic Jihad, Popular Front for the Liberation of Palestine, Fatah Globalized Less identifiable; short-term Salafi jihadist Transnational Broad Unlimited Al-Qaida and associated movements

ideology. Suicide attacks that fall into the traditional, localized pattern have been planned and executed by religious, secular, Marxist, ethnonationalist, and nationalist groups. Examples of religious groups include Hamas, Palestinian Islamic Jihad (PIJ), and Hezbollah. Secular or nationalist groups include the LTTE, the PFLP, Fatahs al-Aqsa Martyrs Brigades, the Kurdistan Workers Party (PKK), and the Syrian Socialist Nationalist Party. geographic scope of actors. Traditionally, suicide missions have been planned and executed by subnational terrorist or insurgent actors such as Hezbollah, the LTTE, Hamas, PIJ, and the PKK. Palestinian organizations employing suicide missions, for example, have largely conducted the operational planning of these missions locally, although they may have been receptive to the strategic message and direction of an exile leadership.84 The subnational nature of these groups suggests that they recruited and trained suicide bombers mostly in or near the conict area, and have rarely sought them from abroad. The majority of PKK recruits, for example, come from large, poor families residing in Turkey.85 Among Palestinian organizations, more than 99 percent of the bombers between 1993 and 2008 were residents of the area of conictthe West Bank, Gaza, and Israel proper. The only exceptions were two Britons involved in the attacks on the bar Mikes Place in Tel Aviv on April 30, 2003.86 As for the LTTE, experts believe that it is unlikely to have drawn its recruits from outside Sri Lanka because they are chosen from within the ranks
84. Yoram Schweitzer, email communication with author, November 19, 2006. 85. Pedahzur, Suicide Terrorism, pp. 167168. 86. They were Asif Mohammed Hanif, a 21-year-old student from West London who blew himself up at the Mikes Place bar in Tel Aviv on April 30, 2003; and 27-year-old Omar Khan Sharif, a married resident of Derby, England. Sharif also intended to perpetrate a suicide bombing at Mikes Place along with Hanif, but Sharifs explosive device failed to detonate. He ed the scene and later

Contending with Terrorism 84

of the regular LTTE army, where the motivation to serve is high.87 According to Stephen Hopgood, for example, The emphasis on commitment to the cause both for regular cadres and Black Tigers makes non-Sri Lankan or Indian Tamil recruits highly unlikely. The LTTE seems to have no recruitment problems for Black Tigers, so looking outside would only be necessary if some ethnic or linguistic feature of the operatives identity was necessary to accomplish the mission.88 targets. Groups that conducted localized suicide missions mostly targeted people and assets of the enemy state near or in the conict area, while largely refraining from targeting assets of their foes in other locations. The PKK, for instance, conducted all of its 16 suicide attacks in Turkey. Hamas and other Palestinian organizations did not execute a suicide mission against Israeli or Jewish targets outside of Israel, the West Bank, and Gaza. Hezbollahs suicide operations against Israel were staged mostly against Israel Defense Forces troops inside Lebanonwith the exception of two suicide attacks on Israeli and Jewish targets in Argentina, for which the group declined to assume responsibility.89 The LTTE staged nearly all of its attacks in Sri Lanka propera notable exception being the killing of Indian Prime Minister Rajiv Gandhi in the Indian city of Madras in May 1991. Two experts on the LTTE whom I consulted were unaware of additional suicide missions carried out by the LTTE outside of Sri Lanka.90 To quote Hopgood, The LTTE is very careful to make it clear its target is the Sri Lankan state and its collaborators, rather than all Sri Lankans. It is conscious of its public image, and escalating to attack on foreign soil would be counterproductive both to legitimacy and diaspora fundraising.91 goals. The subnational terrorist or insurgent movements that followed a localized pattern of suicide attacks generally aim to advance limited and welldened political goals for the community they purport to represent. These political goals may include an end to foreign occupation or military presence, increased regional autonomy, and self-determination. The struggle for an indedrowned in the Mediterranean. His body washed ashore on the Tel Aviv beach front on May 12, 2003. 87. Michael Roberts, email communication with author, November 20, 2006; and Stephen Hopgood, email communication with author, November 24, 2006. 88. Hopgood, email communication with author. 89. Hezbollah is believed to have staged two attacks in Argentina: the March 17, 1992, suicide car bombing of the Israeli embassy in Buenos Aires, which killed 29 people and injured more than 250; and the July 18, 1994, suicide car bombing of the Jewish Community Center building in Buenos Aires, which killed more than 80 people and wounded some 300. 90. Roberts, email communication with author; and Hopgood, email communication with author. 91. Hopgood, email communication with author.

Motives for Martyrdom 85

pendent homeland, whether it is Tamil Eelam, Kurdistan, or Palestine, lies at the center of the conicts in which suicide missions have traditionally been employed. globalized suicide attacks The localized pattern of suicide missions contrasts sharply with the new globalized pattern displayed by al-Qaida and its Sala jihadist associated movements. Although the localized pattern continues to exist, the globalized pattern has become increasingly dominant since the millennium. conict type. Globalized suicide missions may occur in the context of clearly identiable conicts such as Iraq, but those conicts need not have a long history. Suicide attacks in Iraq, for example, occurred less than a week after the start of the U.S.-led invasion in March 200392hardly long enough to produce the types of deep-seated grievances that inuenced Palestinian, Tamil, or Kurdish suicide bombers in localized contexts. Nor are the targets of many globalized suicide bombers aware that they are involved in a conict with a bitter enemy who seeks their death along with its own. Unlike traditional suicide attacks, globalized suicide operations frequently occur in areas thatby any objective standardare not identied by all parties as zones of conict. The September 11 attacks, for instance, did not take place in a region where a large ethnic group was vying for an independent state while battling an occupation army. The same is true for the U.S. embassy bombings in Kenya and Tanzania in August 1998 and the attack on the USS Cole in October 2000. Other examples include Djerba (April 2002), Bali (October 2002 and October 2005), Mombasa (November 2002), Casablanca (May 2003), Istanbul (November 2003), London (July 2005), and Amman (November 2005). ideology. Sala jihadist groups have overwhelmingly planned and executed the new globalized suicide missions. geographic scope of actors. Globalized suicide missions tend to be planned and executed by cells and groups that are connected to a transnational terrorist or insurgent network or movement. This transnational network suggests that the planning of suicide missions and their execution may occur in different places. Examples include the September 11 attacks and the July 2005 London bombings. Additionally, organizations conducting globalized suicide missions no longer recruit and train suicide bombers exclusively in the country where the
92. Steven Lee Myers, A Nation at War: Suicide Strike; With Bombing, Iraqis Escalate Guerrilla Tactics and Show New Danger on Front Lines, New York Times, March 30, 2003.

Contending with Terrorism 86

attacks are to take place. This is true, again, in the case of the September 11 attacks and the 2005 bombings in Amman, and it is especially evident in the preponderance of foreigners who volunteer for suicide attacks in Iraq and Afghanistan.93 The November 2005 bombings in Amman, for example, were executed by three Iraqis.94 targets. Organizations and cells that stage globalized suicide missions do not limit their attacks to an identiable zone of conict. Due in large part to the expansive nature of Sala jihad, many of todays suicide attackers regard much of the world as a legitimate target. Hence, even though al-Qaida has declared the United States its main enemy, it does not limit its suicide attacks to the U.S. homeland. Instead, it will strike U.S. interests wherever an opportunity may arise. In addition, it may strike targets of real or perceived allies of the United States. goals. Suicide missions that fall within the globalized category are generally perpetrated by organizations whose goals are more elusive than those in the localized category. It is unclear, for instance, whether the suicide bombings in Amman in November 2005 were intended to punish the Hashemite monarchy for its pro-Western stance, including its relations with Israel; to target foreign diplomats; to hurt Israeli and Jewish interests in the kingdom; to create instability and spark an anti-Hashemite backlash; or to extend the jihad in Iraq to the broader Middle East. Similarly, Western analysts often argue over al-Qaidas goals and motivations, although few would disagree that its demands are maximalist. implications of ndings Distinguishing between localized and globalized patterns of suicide attacks has several theoretical implications for the study of terrorism. It allows re93. According to a Department of Defense news brieng with Col. Sean MacFarland, commander of the First Brigade Combat Team, First Armored Division stationed in Ramadi, Iraq, [Foreign ghters] are very few in number, although as far as we can tell, they constitute about 100 percent of the suicide bombers. Quoted in Michael OHanlon and Nina Kamp, eds., Iraq Index: Tracking Variables of Reconstruction and Security in Post-Saddam Iraq (Washington, D.C.: Brookings Institution, November 13, 2006, updated October 1, 2007), p. 18. See also comments by Maj. Gen. Rick Lynch, who stated on December 1, 2005, that at least 96 percent of suicide bombers [in Iraq] are not Iraqis. Quoted in Chris Tomlinson, U.S. General: Suicide and Car Bomb Attacks Down in Iraq, Associated Press, December 1, 2005. On Afghanistan, a recent report by the United Nations Assistance Mission in Afghanistan, for instance, suggests that there is little doubt that some perpetrators in Afghanistan crossed the border from Pakistan, although at least some are Afghan refugees. United Nations Assistance Mission in Afghanistan (UNAMA), Suicide Attacks in Afghanistan (20012007) (New York: UNAMA, 2007). 94. Craig Whitlock, Amman Bombings Reect Zarqawis Growing Reach, Washington Post, November 13, 2005.

Motives for Martyrdom 87

searchers to place existing explanations into their proper context, to recognize their limitations, and to dene important new avenues for research. For instance, scholars can better assess the role of occupation, which appears to be a signicant factor in countries whose suicide attacks have a more traditional, localized pattern. It plays a different role in the globalized pattern associated with Sala jihadist ideology, which adopts an extremely loose denition of occupation, rendering virtually any perceived offense an example of Western occupation.95

Conclusion
Most suicide attacks today are perpetrated by terrorist groups that adhere to a radical Sala jihadist ideology. Although ideology thus plays an important role in explaining the global proliferation of suicide attacks, there is no evidence that it is the cause of suicide attacks per se. The causes of suicide attacks are complex: they can be found in the interplay of personal motivations, the strategic and tactical objectives of the sponsoring groups, societal and structural factors, as well as intergroup dynamics at the level of the terrorist cell.96 In addition, individuals acquire ideology for reasons having to do with emotions and beliefsa complex process whose examination exceeds the scope of this article. Ideology plays an important role, however, in helping reduce the suicide attackers reservations to perpetrate the acts of killing and dying. It helps the suicide bomber justify his or her actions and to disengage morally from his act and his victims. Because ideology is an importantand often neglectedfactor in the genesis and spread of suicide attacks, challenging the appeal of this ideology is a crucial component of an overall counterterrorism strategy. The task for the United States in challenging the appeal of Sala jihad will be particularly difcult because of widespread antipathies toward U.S. policies in parts of the Arab and Muslim world. According to a forty-seven-nation Pew Global Atti-

95. A theoretical division of suicide attacks into two patterns also helps contextualize the outbidding thesis. That explanation may account for the adoption of suicide attacks in some cases, but is less capable of accounting for cases of globalized suicide missions. The perpetrators of the London bombings of July 2005, for example, hardly vied for the sympathies of the domestic population on the contrary, they detested the local population to such an extent that they blew themselves up in its midst. The outbidding thesis is therefore less relevant to our understanding of globalized suicide missions because the importance of killing indels seems to supersede organizational rivalries. 96. On the importance of small group dynamics, see Sageman, Understanding Terror Networks; and Hafez, Suicide Bombers in Iraq.

Contending with Terrorism 88

tudes Survey released in the summer of 2007, for example, the U.S. image remains abysmal in most Muslim countries in the Middle East and Asia, and continues to decline among the publics of many of Americas oldest allies.97 U.S. practices in detention centers such as Guantnamo Bay and Abu Ghraib and the rush to war in Iraq have aggravated the negative views of the United States in Arab and Muslim countries and beyond. Partly because of the poor standing of the United States, and more important, because of the grave danger Sala jihad poses to Muslims, nonviolent Salasts, Islamists, and moderate Muslims must begin to challenge this ideology. The United States and its allies can do little to inuence what must primarily be an internal Muslim debate over the future of the Muslim community. They can, however, discreetly convey to moderate Muslims and nonviolent Salasts why waging this internal battle is so important, thus quietly supporting these communities without running the risk of exposing them as subservient to the West. As Muslims prepare for this debate, Western states can underscore what most Muslims already know: the credibility of Sala jihad suffers from a fundamental contradiction. On the one hand, Sala jihadists claim to act for the benet of Muslims. On the other hand, Muslims suffer the consequences of Sala jihadist ideology and terrorism more than any other group. Moderate Muslims can marshal the following three arguments to undermine Sala jihadists. First, Muslims are the primary victims of Sala jihadist terrorism, including suicide attacks. More Muslims than non-Muslims have died or been maimed by Sala jihadist terror in the last three decades. In Algeria alone, 100,000 or more Muslims have lost their lives to acts of violence largely committed by the Sala jihadist Armed Islamic Group. In Iraq, where more than half of all suicide attacks since 1981 have taken place, suicide missions have killed more Iraqi civilians than foreign military or foreign civilian personnel. In Afghanistan, civilians have been the prime victims of the growing number of suicide attacks, even if these attacks were aimed at members of the International Security Assistance Force. In Pakistan, too, an increasing number of suicide attacks have targeted the indigenous population. Second, Sala jihadists defend the killing of Muslims by claiming that the ends justify the means. Innocent Muslims not only die as a by-product of war and insurgency waged by Sala jihadists, but Sala jihadists also seem to believe that Muslims are expendable. As Abu Musab al-Zarqawi noted, Admittedly, the killing of a number of Muslims whom it is forbidden to kill
97. Pew Global Attitudes Project, Global Unease with Major World Powers: Rising Environmental Concern in 47-Nation Survey (Washington, D.C.: Pew Research Center, June 27, 2007), p. 3.

Motives for Martyrdom 89

is undoubtedly a grave evil; however, it is permissible to commit this evil indeed, it is even requiredin order to ward off a greater evil, namely, the evil of suspending jihad.98 Third, the use of takrthe labeling of some Muslims as indelsis dividing the Islamic community and runs the risk of creating a Muslim civil war. The Algerian civil war of the 1990s offers a devastating example of this practice. The use of the takr label has created serious tensions within the Islamic community, and it is used to justify scores of suicide bombings against Muslims in countries such as Afghanistan, Algeria, Iraq, Jordan, and Pakistan. Unless it is rejected by moderate Muslims, who form the majority of the Islamic community, such labeling will continue to lead the Islamic nation on a downward spiral of self-inicted violence. Moderate Muslims should remind their coreligionists that wrongly accusing another Muslim of being an indel is a major sin in Islam. The battle against suicide attacks will not be won by exposing the inconsistencies of Sala jihad alone. Like terrorism more generally, suicide missions are a tactic, and as such cannot be defeated entirely. Like war, there are countless reasons why terrorism occursand like war, it is unlikely that terrorism and suicide attacks will disappear. Governments struggling against terrorism should therefore conceive their battle not as a war whose goal is victory, but as a long-term effort that requires commitment, endurance, and ingenuity.

98. Quoted in Middle East Media Research Institute (MEMRI), Abu Musab Zarqawi: Collateral Killing of Muslims is Legitimate, Special Dispatch Series, No. 917 (Washington, D.C.: Jihad and Terrorism Project, MEMRI, June 7, 2005).

Part II: Terrorist Strategies: Why Are They Chosen? Do They Work?

The Strategies of Terrorism

The Strategies of Terrorism

Andrew H. Kydd and Barbara F. Walter

errorism often works. Extremist organizations such as al-Qaida, Hamas, and the Tamil Tigers engage in terrorism because it frequently delivers the desired response. The October 1983 suicide attack against the U.S. Marine barracks in Beirut, for example, convinced the United States to withdraw its soldiers from Lebanon.1 The United States pulled its soldiers out of Saudi Arabia two years after the terrorist attacks of September 11, 2001, even though the U.S. military had been building up its forces in that country for more than a decade.2 The Philippines recalled its troops from Iraq nearly a month early after a Filipino truck driver was kidnapped by Iraqi extremists.3 In fact, terrorism has been so successful that between 1980 and 2003, half of all suicide terrorist campaigns were closely followed by substantial concessions by the target governments.4 Hijacking planes, blowing up buses, and kidnapping individuals may seem irrational and incoherent to outside observers, but these tactics can be surprisingly effective in achieving a terrorist groups political aims. Despite the salience of terrorism today, scholars and policymakers are only beginning to understand how and why it works. Much has been written on the origins of terror, the motivations of terrorists, and counterterror responses, but little has appeared on the strategies terrorist organizations employ and the conditions under which these strategies succeed or fail. Alan Krueger, David Laitin, Jitka Maleckova, and Alberto Abadie, for example, have traced the effects of poverty, education, and political freedom on terrorist recruitment.5
Andrew H. Kydd is Associate Professor of Political Science at the University of Pennsylvania. Barbara F. Walter is Associate Professor at the Graduate School of International Relations and Pacic Studies at the University of California, San Diego. This article is the second installment in a collaborative project, and the order of the authors names was determined by alphabetical order. The authors would like to thank the participants at the Project on International Affairs seminar at the University of California, San Diego, for helpful comments on an earlier draft. 1. Thomas L. Friedman, Marines Complete Beirut Pullback: Moslems Move In, New York Times, February 27, 2004. 2. Don Van Natta Jr., The Struggle for Iraq: Last American Combat Troops Quit Saudi Arabia, New York Times, September 22, 2003. 3. James Glanz, Hostage Is Freed after Philippine Troops Are Withdrawn from Iraq, New York Times, July 21, 2004. 4. Robert A. Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (New York: Random House, 2005), p. 65. 5. Alan B. Krueger and David D. Laitin, Kto Kogo? A Cross-Country Study of the Origins and
International Security, Vol. 31, No. 1 (Summer 2006), pp. 4980 2006 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

93

Contending with Terrorism 94

Jessica Stern has examined the grievances that give rise to terrorism and the networks, money, and operations that allow terrorist organizations to thrive.6 What is lacking, however, is a clear understanding of the larger strategic games terrorists are playing and the ways in which state responses help or hinder them. Effective counterstrategies cannot be designed without rst understanding the strategic logic that drives terrorist violence. Terrorism works not simply because it instills fear in target populations, but because it causes governments and individuals to respond in ways that aid the terrorists cause. The Irish Republican Army (IRA) bombed pubs, parks, and shopping districts in London because its leadership believed that such acts would convince Britain to relinquish Northern Ireland. In targeting the World Trade Center and the Pentagon on September 11, al-Qaida hoped to raise the costs for the United States of supporting Israel, Saudi Arabia, and other Arab regimes, and to provoke the United States into a military response designed to mobilize Muslims around the world. That so many targeted governments respond in the way that terrorist organizations intend underscores the need for understanding the reasoning behind this type of violence. In this article we seek answers to four questions. First, what types of goals do terrorists seek to achieve? Second, what strategies do they pursue to achieve these goals? Third, why do these strategies work in some cases but not in others? And fourth, given these strategies, what are the targeted governments best responses to prevent terrorism and protect their countries from future attacks? The core of our argument is that terrorist violence is a form of costly signaling. Terrorists are too weak to impose their will directly by force of arms. They are sometimes strong enough, however, to persuade audiences to do as they wish by altering the audiences beliefs about such matters as the terrorists ability to impose costs and their degree of commitment to their cause. Given the conict of interest between terrorists and their targets, ordinary communication or cheap talk is insufcient to change minds or inuence behavior. If al-Qaida had informed the United States on September 10, 2001, that it would

Targets of Terrorism, Princeton University and Stanford University, 2003; Alan B. Krueger and Jitka Maleckova, Education, Poverty, and Terrorism: Is There a Causal Connection? Journal of Economic Perspectives, Vol. 17, No. 4 (November 2003), pp. 119144; and Alberto Abadie, Poverty, Political Freedom, and the Roots of Terrorism, Faculty Research Working Papers Series, RWP04043 (Cambridge, Mass.: John F. Kennedy School of Government, Harvard University, 2004). 6. Jessica Stern, Terror in the Name of God: Why Religious Militants Kill (New York: EccoHarperCollins, 2003).

The Strategies of Terrorism 95

kill 3,000 Americans unless the United States withdrew from Saudi Arabia, the threat might have sparked concern, but it would not have had the same impact as the attacks that followed. Because it is hard for weak actors to make credible threats, terrorists are forced to display publicly just how far they are willing to go to obtain their desired results. There are ve principal strategic logics of costly signaling at work in terrorist campaigns: (1) attrition, (2) intimidation, (3) provocation, (4) spoiling, and (5) outbidding. In an attrition strategy, terrorists seek to persuade the enemy that the terrorists are strong enough to impose considerable costs if the enemy continues a particular policy. Terrorists using intimidation try to convince the population that the terrorists are strong enough to punish disobedience and that the government is too weak to stop them, so that people behave as the terrorists wish. A provocation strategy is an attempt to induce the enemy to respond to terrorism with indiscriminate violence, which radicalizes the population and moves them to support the terrorists. Spoilers attack in an effort to persuade the enemy that moderates on the terrorists side are weak and untrustworthy, thus undermining attempts to reach a peace settlement. Groups engaged in outbidding use violence to convince the public that the terrorists have greater resolve to ght the enemy than rival groups, and therefore are worthy of support. Understanding these ve distinct strategic logics is crucial not only for understanding terrorism but also for designing effective antiterror policies.7 The article is divided into two main sections. The rst discusses the goals terrorists pursue and examines the forty-two groups currently on the U.S. State Departments list of foreign terrorist organizations (FTOs).8 The second section develops the costly signaling approach to terrorism, analyzes the ve strategies that terrorists use to achieve their goals, discusses the conditions in which each of these strategies is likely to be successful, and draws out the implications for the best counterterror responses.

The Goals of Terrorism

For years the press has portrayed terrorists as crazy extremists who commit indiscriminate acts of violence, without any larger goal beyond revenge or a de7. Of course, terrorists will also be seeking best responses to government responses. A pair of strategies that are best responses to each other constitutes a Nash equilibrium, the fundamental prediction tool of game theory. 8. Ofce of Counterterrorism, U.S. Department of State, Foreign Terrorist Organizations, fact sheet, October 11, 2005, http:/ /www.state.gov/s/ct/rls/fs/3719.htm.

Contending with Terrorism 96

sire to produce fear in an enemy population. This characterization derives some support from statements made by terrorists themselves. For example, a young Hamas suicide bomber whose bomb failed to detonate said, I know that there are other ways to do jihad. But this one is sweetthe sweetest. All martyrdom operations, if done for Allahs sake, hurt less than a gnats bite!9 Volunteers for a suicide mission may have a variety of motivesobtaining rewards in the afterlife, avenging a family member killed by the enemy, or simply collecting nancial rewards for their descendants. By contrast, the goals driving terrorist organizations are usually political objectives, and it is these goals that determine whether and how terrorist campaigns will be launched. We dene terrorism as the use of violence against civilians by nonstate actors to attain political goals.10 These goals can be conceptualized in a variety of ways. Individuals and groups often have hierarchies of objectives, where broader goals lead to more proximate objectives, which then become specic goals in more tactical analyses.11 For the sake of simplicity, we adopt the common distinction between goals (or ultimate desires) and strategies (or plans of action to attain the goals). Although the ultimate goals of terrorists have varied over time, ve have had enduring importance: regime change, territorial change, policy change, social control, and status quo maintenance. Regime change is the overthrow of a government and its replacement with one led by the terrorists or at least one more to their liking.12 Most Marxist groups, including the Shining Path (Sendero Luminoso) in Peru have sought this goal. Territorial change is taking territory away from a state either to establish a new state (as the Tamil Tigers seek to do in Tamil areas of Sri Lanka) or to join another state (as Lashkar-e Tayyiba would like to do by incorporating Indian Kashmir into Pakistan).
9. Quoted in Nasra Hassan, An Arsenal of Believers: Talking to the Human Bombs, New Yorker, November 19, 2001, p. 37. 10. For discussion of differing denitions of terrorism, see Alex P. Schmid and Albert J. Jongman, Political Terrorism: A New Guide to Actors, Authors, Concepts, Data Bases, Theories, and Literature (New Brunswick, N.J.: Transaction, 1988), pp. 138. We do not focus on state terrorism because states face very different opportunities and constraints in their use of violence, and we do not believe the two cases are similar enough to be protably analyzed together. 11. For the distinction between goals and strategies, see David A. Lake and Robert Powell, eds., Strategic Choice and International Relations (Princeton, N.J.: Princeton University Press, 1999), especially chap. 1. 12. On revolutionary terrorism, see Martha Crenshaw Hutchinson, The Concept of Revolutionary Terrorism, Journal of Conict Resolution, Vol. 16, No. 3 (September 1972), pp. 383396; Martha Crenshaw Hutchinson, Revolutionary Terrorism: The FLN in Algeria, 19541962 (Stanford, Calif.: Hoover Institution Press, 1978); and H. Edward Price Jr., The Strategy and Tactics of Revolutionary Terrorism, Comparative Studies in Society and History, Vol. 19, No. 1 (January 1977), pp. 5266.

The Strategies of Terrorism 97

Policy change is a broader category of lesser demands, such as al-Qaidas demand that the United States drop its support for Israel and corrupt Arab regimes such as Saudi Arabia. Social control constrains the behavior of individuals, rather than the state. In the United States, the Ku Klux Klan sought the continued oppression of African Americans after the Civil War. More recently, antiabortion groups have sought to kill doctors who perform abortions to deter other doctors from providing this service. Finally, status quo maintenance is the support of an existing regime or a territorial arrangement against political groups that seek to change it. Many right-wing paramilitary organizations in Latin America, such as the United Self-Defense Force of Colombia, have sought this goal.13 Protestant paramilitary groups in Northern Ireland supported maintenance of the territorial status quo (Northern Ireland as British territory) against IRA demands that the territory be transferred to Ireland.14 Some organizations hold multiple goals and may view one as facilitating another. For instance, by seeking to weaken U.S. support for Arab regimes (which would represent a policy change by the United States), al-Qaida is working toward the overthrow of those regimes (or regime change). As another example, Hamas aims to drive Israel out of the occupied territories (territorial change) and then to overthrow it (regime change). A cross section of terrorist organizations listed in Table 1 illustrates the range of goals and their relative frequency. Of the forty-two groups currently designated as FTOs by the U.S. State Department, thirty-one seek regime change, nineteen seek territorial change, four seek policy change, and one seeks to maintain the status quo.15 The list is neither exhaustive nor representative of all terrorist groups, and it does not reect the frequency of goals in the universe of cases. None of the FTOs appear to pursue social control, but some domestic groups, which are by denition not on the list, are more interested in

13. This group has recently surrendered its weapons. 14. Some analysts argue that many terrorist organizations have degenerated into little more than self-perpetuating businesses that primarily seek to enhance their own power and wealth, and only articulate political goals for rhetorical purposes. See, for example, Stern, Terror in the Name of God, pp. 235236. This suggests that power and wealth should be considered goals in their own right. All organizations, however, seek power and wealth to further their political objectives, and these are better viewed as instrumental in nature. 15. A difcult coding issue arises in determining when a group is a nonstate actor engaged in status quo maintenance and when it is simply a covert agent of the state. Some death squads were linked to elements in the armed forces, yet were not necessarily responsive to the chief executive of the country. Others were tied to right-wing parties and are more clearly nonstate, unless that party is the party in power. See Bruce D. Campbell and Arthur D. Brenner, eds., Death Squads in Global Perspective: Murder with Deniability (New York: Palgrave Macmillan, 2002).

Table 1. Foreign Terrorist Organizations and Their Goals Ultimate Goals X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X Destroy Israel; establish Palestinian state Secede from Philippines Destroy Israel; establish Palestinian state Evict United States from Iraq; establish Islamic state Establish Islamic state in Algeria Establish Islamic state in Lebanon Seize power in Japan; hasten the Apocalypse Secede from Spain Establish Communist state in Philippines X X X RC TC PC SC SQM

Name

Abu Nidal Organization Abu Sayyaf Group Al-Aqsa Martyrs Brigade Ansar al-Islam Armed Islamic Group Asbat al-Ansar Aum Shinrikyo Basque Fatherland and Liberty (ETA) Communist Party of the Philippines/New Peoples Army Continuity Irish Republican Army Al-Gamaa al-Islamiyya (Islamic Group) Hamas (Islamic Resistance Movement) Harakat ul-Mujahidin Hezbollah (Party of God)

Contending with Terrorism 98

Islamic Jihad Group

Islamic Movement of Uzbekistan Jaish-e-Mohammed (Army of Mohammed) Jemaah Islamiya Al-Jihad (Egyptian Islamic Jihad) Kahane Chai (Kach) Kongra-Gel (formerly Kurdistan Workers Party) Lashkar-e Tayyiba (Army of the Righteous) Lashkar i Jhangvi Liberation Tigers of Tamil Eelam Libyan Islamic Fighting Group Moroccan Islamic Combatant Group Mujahedin-e Khalq Organization National Liberation Army Palestine Liberation Front Palestinian Islamic Jihad

Evict Britain from Northern Ireland; unite with Eire Establish Islamic state in Egypt Destroy Israel; establish Palestinian Islamic state Evict India from Kashmir; unite with Pakistan Originally: evict Israel from Lebanon; now: destroy Israel and establish Palestinian Islamic state Establish Islamic state in Uzbekistan; reduce U.S. influence Establish Islamic state in Uzbekistan Evict India from Kashmir; unite with Pakistan Establish Islamic state in Indonesia Establish Islamic state in Egypt Expand Israel Secede from Turkey Evict India from Kashmir; unite with Pakistan Establish Islamic state in Pakistan Secede from Sri Lanka Establish Islamic state in Libya Establish Islamic state in Morocco Overthrow Iranian government Establish Marxist government in Colombia Destroy Israel; establish Palestinian state Destroy Israel; establish Palestinian state X X

Table 1. continued Ultimate Goals Destroy Israel; establish Palestinian state Destroy Israel; establish Palestinian state X X X X X X X X X 31 19 4 0 X 1 X Establish Islamic states in Middle East; destroy Israel; reduce U.S. influence Evict United States from Iraq; establish Islamic state Evict Britain from Northern Ireland; unite with Eire Establish Marxist state in Colombia Establish Marxist state in Greece Establish Marxist state in Greece Establish Marxist state in Turkey Establish Islamic state in Algeria Establish Marxist state in Peru Preserve Colombian state X X X X X X RC TC PC SC SCM

Name

Popular Front for the Liberation of Palestine Popular Front for the Liberation of Palestine General Command Al-Qaida

Al-Qaida in Iraq (Zarqawi group) Real Irish Republican Army Revolutionary Armed Forces of Colombia Revolutionary Nuclei (formerly Revolutionary Peoples Struggle) Revolutionary Organization 7 November Revolutionary Peoples Liberation Party/Front

Salafist Group for Call and Combat Shining Path (Sendero Luminoso) United Self-Defense Forces of Colombia Total

SOURCE: Office of Counterterrorism, U.S. Department of State, Foreign Terrorist Organizations, fact sheet, October 11, 2005. NOTE: RC: regime change; TC: territorial change; PC: policy change; SC: social control; and SQM: status quo maintenance. Coding of

The Strategies of Terrorism 99

goals is the authors.

Contending with Terrorism 100

this goal.16 What Table 1 reveals, however, is the instrumental nature of terrorist violence and some of the more popular political objectives being sought.

The Strategies of Terrorist Violence

To achieve their long-term objectives, terrorists pursue a variety of strategies. Scholars have suggested a number of typologies of terrorist strategies and tactics over the years. In a pathbreaking early analysis of terrorism, Thomas Thornton offered ve proximate objectives: morale building, advertising, disorientation (of the target population), elimination of opposing forces, and provocation.17 Martha Crenshaw also identies advertising and provocation as proximate objectives, along with weakening the government, enforcing obedience in the population, and outbidding.18 David Fromkin argues that provocation is the strategy of terrorism.19 Edward Price writes that terrorists must delegitimize the regime and impose costs on occupying forces, and he identies kidnapping, assassination, advertising, and provocation as tactics.20 Although these analyses are helpful in identifying strategies of terrorism, they fail to derive them from a coherent framework, spell out their logic in detail, and consider best responses to them. A fruitful starting point for a theory of terrorist strategies is the literature on uncertainty, conict, and costly signaling. Uncertainty has long been understood to be a cause of conict. Geoffrey Blainey argued that wars begin when states disagree about their relative power, and they end when states agree again.21 James Fearon and other theorists built upon this insight and showed
16. The Taliban, which is not listed, does pursue social control; and the Israeli group Kach, which seeks to maintain the subordinate status of Palestinians in Israel and eventually to expel them, may also be considered to seek it. The Memorial Institute for the Prevention of Terrorism maintains a database of terrorist organizations that includes more than forty groups based in the United States. Some of these can be considered to seek social control, such as the Army of God, which targets doctors who provide abortions. See http:/ /www.tkb.org. 17. Thomas Perry Thornton, Terror as a Weapon of Political Agitation, in Harry Eckstein, ed., Internal War: Problems and Approaches (London: Free Press of Glencoe, 1964), p. 87. 18. Martha Crenshaw, The Causes of Terrorism, Comparative Politics, Vol. 13, No. 4 (July 1981), pp. 379399. 19. David Fromkin, The Strategy of Terrorism, Foreign Affairs, Vol. 53, No. 4 (July 1975), pp. 683 698. 20. Price, The Strategy and Tactics of Revolutionary Terrorism, pp. 5458. Other related discussions include Paul Wilkinson, The Strategic Implications of Terrorism, in M.L. Sondhi, ed., Terrorism and Political Violence: A Sourcebook (New Delhi: Har-anand Publications, 2000); Paul Wilkinson, Terrorism and the Liberal State (New York: New York University Press, 1986), pp. 110 118; and Schmid and Jongman, Political Terrorism, pp. 5059. 21. Geoffrey Blainey, The Causes of War, 3d ed. (New York: Free Press, 1988), p. 122.

The Strategies of Terrorism 101

that uncertainty about a states willingness to ght can cause conict.22 If states are unsure what other states will ght for, they may demand too much in negotiations and end up in conict. This uncertainty could reect a disagreement about power, as Blainey understood, or a disagreement over resolve, willpower, or the intensity of preferences over the issue. The United States and North Vietnam did not disagree over their relative power, but the United States fatally underestimated North Vietnamese determination to achieve victory. Uncertainty about trustworthiness or moderation of preferences can also cause conict. Thomas Hobbes argued that if individuals mistrust each other, they have an incentive to initiate an attack rather than risk being attacked by surprise.23 John Herz, Robert Jervis, and others have developed this concept in the international relations context under the heading of the security dilemma and the spiral model.24 States are often uncertain about each others ultimate ambitions, intentions, and preferences. Because of this, anything that increases one sides belief that the other is deceitful, expansionist, risk acceptant, or hostile increases incentives to ght rather than cooperate. If uncertainty about power, resolve, and trustworthiness can lead to violence, then communication on these topics is the key to preventing (or instigating) conict. The problem is that simple verbal statements are often not credible, because actors frequently have incentives to lie and bluff. If by saying Were resolved, the North Vietnamese could have persuaded the United States to abandon the South in 1965, then North Vietnam would have had every incentive to say so even if it was not that resolute. In reality, they had to ght a long and costly war to prove their point. Similarly, when Mikhail Gorbachev wanted to reassure the West and end the Cold War, verbal declarations of innocent intentions were insufcient, because previous Soviet leaders had made similar statements. Instead, real arms reductions, such as the 1987 Intermediate-Range Nuclear Forces Treaty, were necessary for Western opinion to change.
22. James D. Fearon, Rationalist Explanations for War, International Organization, Vol. 49, No. 3 (Summer 1995), pp. 379414; and Robert Powell, Bargaining Theory and International Conict, Annual Review of Political Science, Vol. 5 (June 2002), pp. 130. 23. Thomas Hobbes, Leviathan (New York: Penguin, [1651] 1968), pp. 184. 24. John H. Herz, Idealist Internationalism and the Security Dilemma, World Politics, Vol. 2, No. 2 (January 1950), pp. 157180; Robert Jervis, Perception and Misperception in International Politics (Princeton, N.J.: Princeton University Press, 1976); Robert Jervis, Cooperation under the Security Dilemma, World Politics, Vol. 30, No. 2 (January 1978), pp. 167214; and Charles L. Glaser, The Security Dilemma Revisited, World Politics, Vol. 50, No. 1 (October 1997), pp. 171202.

Contending with Terrorism 102

Because talk is cheap, states and terrorists who wish to inuence the behavior of an adversary must resort to costly signals.25 Costly signals are actions so costly that bluffers and liars are unwilling to take them.26 In international crises, mobilizing forces or drawing a very public line in the sand are examples of strategies that less resolved actors might nd too costly to take.27 War itself, or the willingness to endure it, can serve as a forceful signal of resolve and provide believable information about power and capabilities.28 Costly signals separate the wheat from the chaff and allow honest communication, although sometimes at a terrible price. To obtain their political goals, terrorists need to provide credible information to the audiences whose behavior they hope to inuence. Terrorists play to two key audiences: governments whose policies they wish to inuence and individuals on the terrorists own side whose support or obedience they seek to gain.29 The targeted governments are central because they can grant concessions over policy or territory that the terrorists are seeking. The terrorists domestic audience is also important, because they can provide resources to the terrorist group and must obey its edicts on social or political issues. Figure 1 shows how the three subjects of uncertainty (power, resolve, and trustworthiness) combine with the two targets of persuasion (the enemy government and the domestic population) to yield a family of ve signaling strategies. These strategies form a theoretically cohesive set that we believe represents most of the commonly used strategies in important terrorist campaigns around the world today.30 A terrorist organization can of course pursue
25. Andrew H. Kydd, Trust and Mistrust in International Relations (Princeton, N.J.: Princeton University Press, 2005). 26. John G. Riley, Silver Signals: Twenty-ve Years of Screening and Signaling, Journal of Economic Literature, Vol. 39, No. 2 (June 2001), pp. 432478. 27. James D. Fearon, Signaling Foreign Policy Interests: Tying Hands vs. Sunk Costs, Journal of Conict Resolution, Vol. 41, No. 1 (February 1977), pp. 6890. 28. Dan Reiter, Exploring the Bargaining Model of War, Perspectives on Politics, Vol. 1, No. 1 (March 2003), pp. 2743; and Robert Powell, Bargaining and Learning While Fighting, American Journal of Political Science, Vol. 48, No. 2 (April 2004), pp. 344361. 29. Rival terrorist or moderate groups are also important, but terrorism is not often used to signal such groups. Sometimes rival groups are targeted in an effort to eliminate them, but this violence is usually thought of as internecine warfare rather than terrorism. The targeted government may also be divided into multiple actors, but these divisions are not crucial for a broad understanding of terrorist strategies. 30. This list is not exhaustive. In particular, it omits two strategies that have received attention in the literature: advertising and retaliation. Advertising may play a role in the beginning of some conicts, but it does not sustain long-term campaigns of terrorist violence. Retaliation is a motivation for some terrorists, but terrorism would continue even if the state did not strike at terrorists, because terrorism is designed to achieve some goal, not just avenge counterterrorist attacks.

The Strategies of Terrorism 103

Figure 1. Strategies of Terrorist Violence

more than one strategy at a time. The September 11 terrorist attacks, for example, were probably part of both an attrition strategy and a provocation strategy. By targeting the heart of the United States nancial district, al-Qaida may have been attempting to increase the cost of the U.S. policy of stationing soldiers in Saudi Arabia. But by targeting prominent symbols of American economic and military power, al-Qaida may also have been trying to goad the United States into an extreme military response that would serve al-Qaidas larger goal of radicalizing the worlds Muslim population. The challenge for policymakers in targeted countries is to calibrate their responses in ways that do not further any of the terrorists goals. Below we analyze the ve terrorist strategies in greater detail, discuss the conditions under which each is likely to succeed, and relate these conditions to the appropriate counterterrorism strategies. attrition: a battle of wills The most important task for any terrorist group is to persuade the enemy that the group is strong and resolute enough to inict serious costs, so that the enemy yields to the terrorists demands.31 The attrition strategy is designed to accomplish this task.32 In an attrition campaign, the greater the costs a terrorist
31. Per Baltzer Overgaard, The Scale of Terrorist Attacks as a Signal of Resources, Journal of Conict Resolution, Vol. 38, No. 3 (September 1994), pp. 452478; and Harvey E. Lapan and Todd Sandler, Terrorism and Signaling, European Journal of Political Economy, Vol. 9, No. 3 (August 1993), pp. 383398. 32. J. Maynard Smith, The Theory of Games and Evolution in Animal Conicts, Journal of Theoretical Biology, Vol. 47 (1974), pp. 209211; John J. Mearsheimer, Conventional Deterrence (Ithaca,

Contending with Terrorism 104

organization is able to inict, the more credible its threat to inict future costs, and the more likely the target is to grant concessions. During the last years of the British Empire, the Greeks in Cyprus, Jews in Palestine, and Arabs in Aden used a war of attrition strategy against their colonizer. By targeting Britain with terrorist attacks, they eventually convinced the political leadership that maintaining control over these territories would not be worth the cost in British lives.33 Attacks by Hezbollah and Hamas against Israel, particularly during the second intifada, also appear to be guided by this strategy. In a letter written in the early 1990s to the leadership of Hamas, the organizations master bomb maker, Yahya Ayyash, said, We paid a high price when we used only sling-shots and stones. We need to exert more pressure, make the cost of the occupation that much more expensive in human lives, that much more unbearable.34 Robert Pape presents the most thorough exposition of terrorism as a war of attrition in his analysis of suicide bombing.35 Based on a data set of all suicide attacks from 1980 to 2003 (315 in total), Pape argues that suicide terrorism is employed by weak actors for whom peaceful tactics have failed and conventional military tactics are infeasible because of the imbalance of power. The strategy is to inict costs on the enemy until it withdraws its occupying forces: the greater the costs inicted, the more likely the enemy is to withdraw. Pape asserts that terrorists began to recognize the effectiveness of suicide terrorism with the 1983 Hezbollah attack against U.S. Marines in Beirut that killed 241 people. Since then, suicide terrorism has been employed in nationalist struggles around the world. conditions favorable to attrition. A war of attrition strategy is more effective against some targets than others. Three variables are likely to gure in the outcome: the states level of interest in the issue under dispute, the constraints on its ability to retaliate, and its sensitivity to the costs of violence. The rst variable, the states degree of interest in the disputed issue, is fundamental. States with only peripheral interests at stake often capitulate to terrorist demands; states with more important interests at stake rarely do. The

N.Y.: Cornell University Press, 1983), pp. 3335; and James D. Fearon, Bargaining, Enforcement, and International Cooperation, International Organization, Vol. 52, No. 2 (Spring 1998), pp. 269 305. 33. Bernard Lewis, The Revolt of Islam, New Yorker, November 19, 2001, p. 61. 34. Quoted in Hassan, An Arsenal of Believers, p. 38. 35. Robert A. Pape, The Strategic Logic of Suicide Terrorism, American Political Science Review, Vol. 97, No. 3 (August 2003), pp. 343361; and Pape, Dying to Win.

The Strategies of Terrorism 105

United States withdrew from Lebanon following the bombing of the marine barracks because it had only a marginal interest in maintaining stability and preventing Syrian domination of that country. In that case, the costs of the attack clearly outweighed the U.S. interests at stake. Similarly, Israel withdrew from southern Lebanon in 2000 because the costs of the occupation outstripped Israels desire to maintain a buffer zone in that region. In contrast, the United States responded to the September 11 attacks by launching offensive wars in Afghanistan and Iraq rather than withdrawing U.S. troops from the region, as al-Qaida demanded (though U.S. troops did ultimately leave Saudi Arabia for Iraq). Similarly, Israel is unlikely to withdraw from East Jerusalem, much less allow itself to become an Islamic state as Hamas has demanded. The second variable, constraints on retaliation, affects the costs paid by the terrorists for pursuing a war of attrition. Terrorist organizations almost always are weaker than the governments they target and, as a result, are vulnerable to government retaliation. The more constrained the government is in its use of force, the less costly an attrition strategy is, and the longer the terrorists can hold out in the hopes of achieving their goal. For instance, the Israelis have the military means to commit genocide against the Palestinian people or to expel them to surrounding Arab countries. Israel, however, depends for its longterm survival on close ties with Europe and the United States. Western support for Israel would plummet in response to an Israeli strategy designed to inict mass casualties, making such a strategy prohibitively costly. This constraint makes a war of attrition strategy less costly (and more attractive) for the Palestinians. Democracies may be more constrained in their ability to retaliate than authoritarian regimes. Pape nds that suicide bombers target democracies exclusively and argues that this is in part because of constraints on their ability to strike back.36 Capable authoritarian regimes are able to gather more information on their populations than democracies and can more easily round up suspected terrorists and target those sympathetic to them. They are also less constrained by human rights considerations in their interrogation and retaliation practices.37 The ease with which a terrorist organization can be targeted also inuences

36. Pape, Dying to Win, p. 44. Krueger and Laitin also nd that targets of terrorism tend to be democratic. See Krueger and Laitin, Kto Kogo? 37. The U.S. program of extraordinary rendition, for example, is an effort to evade the restrictions usually faced by democracies by outsourcing the dirty work.

Contending with Terrorism 106

a countrys ability to retaliate forcefully. Terrorist organizations such as alQaida that are widely dispersed, difcult to identify, or otherwise hard to target are at an advantage in a war of attrition because their enemies will have difculty delivering punishment. Israel has, through superior intelligence gathering, been able to assassinate top members of Hamass leadership at will, including its founder and spiritual leader, Sheik Ahmed Yassin, as well as his successor, Abdel Aziz Rantisi. The United States, by contrast, has been unable to locate Osama bin Laden and his top deputy, Ayman al-Zawahiri. The third variable is a targets cost tolerance. Governments that are able to absorb heavier costs and hold out longer are less inviting targets for an attrition strategy. Terrorist organizations are likely to gauge a targets cost tolerance based on at least two factors: the targets regime type and the targets past behavior toward other terrorists. Regime type is important because democracies may be less able to tolerate the painful effects of terrorism than nondemocracies. Citizens of democracies, their fears stoked by media reports and warnings of continued vulnerability, are more likely to demand an end to the attacks. In more authoritarian states, the government exerts more control over the media and can disregard public opinion to a greater extent. The Russian governments heavy-handed response to hostage situations, for example, suggests a higher tolerance for casualties than a more fully democratic government would have. Additionally, because terrorist organizations operate more freely in democracies and politicians must interact with the public to maintain political support, terrorists have an easier time targeting prominent individuals for assassination. Of four leaders assassinated by terrorists in the past quarter centuryIndira Gandhi, Rajiv Gandhi, Yitzak Rabin, and Anwar Sadatthree were leaders of democracies. Among democratic states, sensitivity to costs may vary with the party in power. When more dovish parties are in charge, the target may be perceived to have lower cost tolerances than if a more hawkish party were at the helm. The dove-hawk dimension may correlate with the left-right dimension in domestic politics, leading left-wing parties to be more likely to grant terrorist demands. This traditional divide between peace and security has characterized Israeli politics for years. Labor Party Prime Minister Ehud Barak was elected on a platform of withdrawing Israeli forces from Lebanon and making peace with the Palestinians; in contrast, Likud Party Prime Minister Ariel Sharon was elected on a platform of meeting terrorists with military force. Hoping for greater concessions, terrorists may preferentially attack dovish parties. The number of prior concessions made to other terrorists is also likely to

The Strategies of Terrorism 107

inuence perceptions of the targets cost tolerance. Governments that have already yielded to terrorist demands are more likely to experience additional terrorist attacks. Evidence abounds that terrorists explicitly consider the prior behavior of states and are encouraged by signs of weakness. Israels precipitous withdrawal from southern Lebanon in May 2000 convinced Hamas that the Israeli leaderships resolve was weakening and encouraged Hamas leaders to initiate the second intifada in September 2000.38 Israelis fear the same inference will be drawn from their withdrawal from Gaza. A Hamas leader interviewed in October 2005 declared, When we took up arms and launched [the second intifada], we succeeded in less than ve years to force the Israelis to withdraw from the Gaza Strip. This fullled everyones dream. I think we have to benet from this experience by applying it accordingly to the West Bank and other occupied areas.39 The past behavior of a targeted government, therefore, also provides important information to terrorist groups about its likely future behavior and the success of this particular strategy. Perhaps the most important example of a terrorist group pursuing an attrition strategy is al-Qaidas war with the United States. In a November 2004 broadcast, bin Laden boasted, We gained experience in guerilla and attritional warfare in our struggle against the great oppressive superpower, Russia, in which we and the mujahidin ground it down for ten years until it went bankrupt, and decided to withdraw in defeat. . . . We are continuing to make America bleed to the point of bankruptcy.40 Al Qaidas goalpolicy change is well suited to an attrition strategy. Bin Laden has frequently argued that the United States lacks the resolve to ght a long attritional war, as in his February 1996 declaration of jihad: Where was this false courage of yours when the explosion in Beirut took place in 1983 a.d.? You were transformed into scattered bits and pieces; 241 soldiers were killed, most of them Marines. And where was this courage of yours when two explosions made you leave Aden in less than twenty-four hours! But your most disgraceful case was in Somalia; where, after vigorous propaganda about the power of the U.S. and its postcold war leadership of the new world order, you moved tens of thousands of international forces, including twenty-eight thousand American soldiers, into Somalia. However, when tens of your soldiers were killed in minor battles and one American pilot was
38. Debate Goes On Over Lebanon Withdrawal, Haaretz, May 23, 2001; and Daoud Kuttab, The Lebanon Lesson, Jerusalem Post, May 25, 2000. 39. Interview with Mahmoud Khalid al-Zahar, Al Jazeera, October 22, 2005. 40. Osama bin Laden, Messages to the World: The Statements of Osama bin Laden, trans. James Howarth, ed. Bruce Lawrence (London: Verso, 2005), pp. 241242.

Contending with Terrorism 108

dragged in the streets of Mogadishu, you left the area in disappointment, humiliation, and defeat, carrying your dead with you. Clinton appeared in front of the whole world threatening and promising revenge, but these threats were merely a preparation for withdrawal. You had been disgraced by Allah and you withdrew; the extent of your impotence and weaknesses became very clear.41 Although difcult to prove, it also appears that bin Laden believed that he and his organization would be hard to target with counterattacks, making a war of attrition strategy even more appealing. In 2001 the Taliban was on the verge of eliminating armed resistance in northern Afghanistan; and, as a landlocked country, Afghanistan must have seemed relatively invulnerable to a U.S. invasion. The United States had bombed al-Qaida camps before to no effect. Even if the United States invaded, Afghanistan was both costly and difcult to conquer, as the Soviets discovered in the 1980s. In the end, of course, the Taliban would have been well advised to insist that the September 11 attacks be delayed until the Northern Alliance was defeated, but the latters dramatic success with U.S. help was perhaps difcult to anticipate. best responses to attrition. There are at least ve counterstrategies available to a state engaged in a war of attrition. First, the targeted government can concede inessential issues in exchange for peace, a strategy that we believe is frequently pursued though rarely admitted.42 In some cases, the terrorists will genuinely care more about the disputed issue and be willing to outlast the target. In such cases, concessions are likely to be the states best response. Other potential challengers, however, may perceive this response as a sign of weakness, which could lead them to launch their own attacks. To reduce the damage to its reputation, the target can vigorously ght other wars of attrition over issues it cares more deeply about, thus signaling a willingness to bear costs if the matter is of sufcient consequence. Second, where the issue under dispute is important enough to the targeted state that it does not want to grant any concessions, the government may engage in targeted retaliation. Retaliation can target the leadership of the terrorist group, its followers, their assets, and other objects of value. Care must be taken, however, that the retaliation is precisely targeted, because the terrorist

41. Osama bin Laden, Declaration of War against the Americans Occupying the Land of the Two Holy Places, Al-Quds Al-Arabi, August 1996, http:/ /www.pbs.org/newshour/terrorism/ international/fatwa_1996.html. 42. Peter C. Sederberg, Conciliation as Counter-terrorist Strategy, Journal of Peace Research, Vol. 32, No. 3 (August 1995), pp. 295312.

The Strategies of Terrorism 109

organization could simultaneously be pursuing a strategy of provocation. A harsh, indiscriminate response might make a war of attrition more costly for the terrorists, but it would also harm innocent civilians who might then serve as willing recruits for the terrorists. The Israeli policy of assassination of terrorist leaders is shaped by this concern. Third, a state can harden likely targets to minimize the costs the terrorist organization can inict. If targeted governments can prevent most attacks from being executed, a war of attrition strategy will not be able to inict the costs necessary to convince the target to concede. The wall separating Israel from the West Bank and Gaza is a large-scale example of this counterstrategy. The United States has been less successful in hardening its own valuable targets, such as nuclear and chemical plants and the container shipping system, despite the creation of the Department of Homeland Security.43 Protecting these types of targets is essential if one seeks to deter additional attacks and discourage the use of attrition. Fourth, states should seek to deny terrorists access to the most destructive weapons, especially nuclear and biological ones. Any weapon that can inict enormous costs will be particularly attractive to terrorists pursuing a war of attrition. The greater the destruction, the higher the likelihood that the target will concede increasingly consequential issues. Particular attention should be placed on securing Russian stockpiles of ssile material and on halting the spread of uranium enrichment technology to Iran and North Korea. No other country has as much material under so little government control as Russia, and Iran and North Korea are vital because of the links both countries have to terrorist organizations.44 Finally, states can strive to minimize the psychological costs of terrorism and the tendency people have to overreact. John Mueller has noted that the risks associated with terrorism are actually quite small; for the average U.S. citizen, the likelihood of being a victim of a terrorist attack is about the same as that of being struck by lighting.45 Government public education programs should therefore be careful not to overstate the threat, for this plays into the hands of
43. Stephen Flynn, America the Vulnerable: How Our Government Is Failing to Protect Us from Terrorism (New York: HarperCollins, 2004). 44. Graham T. Allison, Owen R. Cot Jr., Richard A. Falkenrath, and Steven E. Miller, Avoiding Nuclear Anarchy: Containing the Threat of Loose Russian Nuclear Weapons and Fissile Material (Cambridge, Mass.: MIT Press, 1996); and Graham Allison, Nuclear Terrorism: The Ultimate Preventable Catastrophe (New York: Times Books, 2004). 45. John Mueller, Six Rather Unusual Propositions about Terrorism, Terrorism and Political Violence, Vol. 17, No. 4 (Winter 2005), pp. 487505.

Contending with Terrorism 110

the terrorists. If Americans become convinced that terrorism, while a deadly problem, is no more of a health risk than drunk driving, smoking, or obesity, then al-Qaidas attrition strategy will be undercut. What the United States should seek to avoid are any unnecessary costs associated with wasteful and misguided counterterror programs. The more costs the United States inicts on itself in the name of counterterrorism policies of dubious utility, the more likely a war of attrition strategy is to succeed. intimidation: the reign of terror Intimidation is akin to the strategy of deterrence, preventing some undesired behavior by means of threats and costly signals.46 It is most frequently used when terrorist organizations wish to overthrow a government in power or gain social control over a given population. It works by demonstrating that the terrorists have the power to punish whoever disobeys them, and that the government is powerless to stop them. Terrorists are often in competition with the government for the support of the population. Terrorists who wish to bring down a government must somehow convince the governments defenders that continued backing of the government will be costly. One way to do this is to provide clear evidence that the terrorist organization can kill those individuals who continue to sustain the regime. By targeting the governments more visible agents and supporters, such as mayors, police, prosecutors, and pro-regime citizens, terrorist organizations demonstrate that they have the ability to hurt their opponents and that the government is too weak to punish the terrorists or protect future victims. Terrorists can also use an intimidation strategy to gain greater social control over a population. Terrorists may turn to this strategy in situations where a government has consistently refused to implement a policy a terrorist group favors and where efforts to change the states policy appear futile. In this case, terrorists use intimidation to impose the desired policy directly on the population, gaining compliance through selective violence and the threat of future reprisals. In the United States, antiabortion activists have bombed clinics to prevent individuals from performing or seeking abortions, and in the 1960s racist groups burned churches to deter African Americans from claiming their

46. The literature on deterrence is vast. See, for example, Thomas C. Schelling, Arms and Inuence (New Haven, Conn.: Yale University Press, 1966); and Christopher H. Achen and Duncan Snidal, Rational Deterrence Theory and Comparative Case Studies, World Politics, Vol. 41, No. 2 (January 1989), pp. 143169.

The Strategies of Terrorism 111

civil rights. In Afghanistan, the Taliban beheaded the principal of a girls school to deter others from providing education for girls.47 An intimidation strategy can encompass a range of actionsfrom assassinations of individuals in positions of power to car bombings of police recruits, such as those carried out by the Zarqawi group in Iraq. It can also include massacres of civilians who have cooperated with the government or rival groups, such as the 1957 massacre at Melouza by the National Liberation Front during the Algerian war for independence.48 This strategy was taken to an extreme by the Armed Islamic Group in Algerias civil war of the 1990s. In that war, Islamist guerrillas massacred thousands of people suspected of switching their allegiance to the government. Massacres were especially common in villages that had once been under rm rebel control but that the army was attempting to retake and clear of rebels. Stathis Kalyvas argues that these conditions pose extreme dilemmas for the local inhabitants, who usually wish to support whoever will provide security, but are often left exposed when the government begins to retake an area but has not established effective control.49 conditions favorable to intimidation. When the goal is regime change, weak states and rough terrain are two factors that facilitate intimidation. James Fearon and David Laitin argue that civil wars are likely to erupt and continue where the government is weak and the territory is large and difcult to traverse. These conditions allow small insurgent groups to carve out portions of a country as a base for challenging the central government.50 Intimidation is likely to be used against civilians on the fault lines between rebel and government control to deter individuals from supporting the government. When the goal is social control, weak states again facilitate intimidation. When the justice system is too feeble to effectively prosecute crimes associated with intimidation, people will either live in fear or seek protection from nonstate actors such as local militias or gangs. Penetration of the justice system by sympathizers of a terrorist group also facilitates an intimidation strategy, because police and courts will be reluctant to prosecute crimes and may even be complicit in them.

47. Noor Khan, Militants Behead Afghan Principal for Educating Girls, Boston Globe, January 5, 2006. 48. Crenshaw Hutchinson, The Concept of Revolutionary Terrorism, p. 390. 49. Stathis N. Kalyvas, Wanton and Senseless? The Logic of Massacres in Algeria, Rationality and Society, Vol. 11, No. 3 (August 1999), pp. 243285. 50. James D. Fearon and David D. Laitin, Ethnicity, Insurgency, and Civil War, American Political Science Review, Vol. 97, No. 1 (February 2003), pp. 7590.

Contending with Terrorism 112

best responses to intimidation. When the terrorist goal is regime change, the best response to intimidation is to retake territory from the rebels in discrete chunks and in a decisive fashion. Ambiguity about who is in charge should be minimized, even if this means temporarily ceding some areas to the rebels to concentrate resources on selected sections of territory. This response is embodied in the clear-and-hold strategy that U.S. forces are employing in Iraq. The 2005 National Strategy for Victory in Iraq specically identies intimidation as the strategy of our enemies.51 The proper response, as Secretary of State Condoleezza Rice stated in October 2005, is to clear, hold, and build: clear areas from insurgent control, hold them securely, and build durable national Iraqi institutions.52 If rebels control their own zone and have no access to the government zone, they will have no incentive to kill the civilians they control and no ability to kill the civilians the government controls. In this situation, there is no uncertainty about who is in control; the information that would be provided by intimidation is already known. The U.S. military developed the clear-and-hold strategy during the nal years of U.S. involvement in Vietnam. A principal strategy of the Vietcong was intimidationto prevent collaboration with the government and build up control in the countryside. In the early years of the war, the United States responded with search and destroy missions, essentially an attrition strategy. Given that the insurgents were not pursuing an attrition strategy, and were not particularly vulnerable to one, this initial counterstrategy was a mistake. Clear-and-hold was the more appropriate response because it limited the Vietcongs access to potential targets and thus undercut its strategy.53 Clear-and-hold has its limitations. It is usually impossible to completely deny terrorists entry into the government-controlled zones. In 2002 Chechen terrorists were able to hold a theater audience of 912 people hostage in the heart of Moscow, and 130 were killed in the operation to retake the building. The Shining Path frequently struck in Lima, far from its mountain strongholds. In such situations, a more effective counterstrategy would be to invest in protecting the targets of attacks. In most states, most of the time, the majority of state agents do not need to worry about their physical security, because no one
51. United States National Security Council, National Strategy for Victory in Iraq (Washington, D.C.: White House, November 2005), p. 7. 52. Secretary of State Condoleezza Rice, Iraq and U.S. Policy, testimony before the U.S. Senate Committee on Foreign Relations, October 19, 2005, 109th Cong., 1st sess., http:/ /www.foreign .senate.gov/testimony/2005/RiceTestimony051019.pdf. 53. See Lewis Sorley, A Better War: The Unexamined Victories and the Final Tragedy of Americas Last Years in Vietnam (New York: Harcourt, 1999). This thesis is not without controversy. See Matt Steinglass, Vietnam and Victory, Boston Globe, December 18, 2005.

The Strategies of Terrorism 113

wants to harm them. However, certain state agents, such as prosecutors of organized crime, are more accustomed to danger, and procedures have been developed to protect them. These procedures should be applied to election workers, rural ofcials and police, community activists, and any individual who plays a visible role in the support and functioning of the embattled government. When the terrorist goal is social control, the best response is strengthening law enforcement. This may require more resources to enable the government to effectively investigate and prosecute crimes. More controversial, it may mean using national agencies such as the Federal Bureau of Investigation to bypass local ofcials who are sympathetic to the terrorist group and investigating law enforcement agencies to purge such sympathizers if they obstruct justice. The state can also offer additional protection to potential targets and increase penalties for violence against them. For instance, the 1994 federal Freedom of Access to Clinic Entrances Act, passed in the wake of the 1993 killing of a doctor at an abortion clinic in Florida, prohibits any violence designed to prevent people from entering such clinics. provocation: lighting the fuse A provocation strategy is often used in pursuit of regime change and territorial change, the most popular goals of the FTOs listed by the State Department. It is designed to persuade the domestic audience that the target of attacks is evil and untrustworthy and must be vigorously resisted. Terrorist organizations seeking to replace a regime face a signicant challenge: they are usually much more hostile to the regime than a majority of the states citizens. Al-Qaida may wish to topple the House of Saud, but if a majority of citizens do not support this goal, al-Qaida is unlikely to achieve it. Similarly, if most Tamils are satised living in a united Sri Lanka, the Tamil Tigers drive for independence will fail. To succeed, therefore, a terrorist organization must rst convince moderate citizens that their government needs to be replaced or that independence from the central government is the only acceptable outcome. Provocation helps shift citizen support away from the incumbent regime. In a provocation strategy, terrorists seek to goad the target government into a military response that harms civilians within the terrorist organizations home territory.54 The aim is to convince them that the government is so evil that the

54. Fromkin, The Strategy of Terrorism.

Contending with Terrorism 114

radical goals of the terrorists are justied and support for their organization is warranted.55 This is what the Basque Fatherland and Liberty group (ETA) sought to do in Spain. For years, Madrid responded to ETA attacks with repressive measures against the Basque community, mobilizing many of its members against the government even if they did not condone those attacks. As one expert on this conict writes, Nothing radicalizes a people faster than the unleashing of undisciplined security forces on its towns and villages.56 David Lake argues that moderates are radicalized because government attacks provide important information about the type of leadership in power and its willingness to negotiate with more moderate elements.57 Ethan Bueno de Mesquita and Eric Dickson develop this idea and show that if the government has the ability to carry out a discriminating response to terrorism but chooses an undiscriminating one, it reveals itself to be unconcerned with the welfare of the countrys citizens. Provocation, therefore, is a way for terrorists to force an enemy government to reveal information about itself that then helps the organization recruit additional members.58 conditions favorable to provocation. Constraints on retaliation and regime type are again important in determining when provocation is successful. For provocation to work, the government must be capable of middling levels of brutality. A government willing and able to commit genocide makes a bad target for provocation, as the response will destroy the constituency the terrorists represent. At the opposite pole, a government so committed to human rights and the rule of law that it is incapable of inicting indiscriminate punishment also makes a bad target, because it cannot be provoked. Such a government might be an attractive target for an attrition strategy if it is not very good at stopping attacks, but provocation will be ineffective. What explains why a government would choose a less discriminating counterstrategy over a more precise one? In some instances, a large-scale military response will enhance the security of a country rather than detract from it. If the target government is able to eliminate the leadership of a terrorist orga-

55. Crenshaw, The Causes of Terrorism, p. 387; and Price, The Strategy and Tactics of Revolutionary Terrorism, p. 58. 56. Paddy Woodworth, Why Do They Kill? The Basque Conict in Spain, World Policy Journal, Vol. 18, No. 1 (Spring 2001), p. 7. 57. David A. Lake, Rational Extremism: Understanding Terrorism in the Twenty-rst Century, Dialog-IO, Vol. 56, No. 2 (Spring 2002), pp. 1529. 58. Ethan Bueno de Mesquita and Eric S. Dickson, The Propaganda of the Deed: Terrorism, Counterterrorism, and Mobilization, Washington University and New York University, 2005. Bueno de Mesquita and Dickson also argue that government violence lowers economic prosperity, which favors extremists in their competition with moderates.

The Strategies of Terrorism 115

nization and its operatives, terrorism is likely to cease or be greatly reduced even if collateral damage radicalizes moderates to some extent. A large-scale military response may also enhance the security of a country, despite radicalizing some moderates, if it deters additional attacks from other terrorist groups that may be considering a war of attrition. Target governments may calculate that the negative consequences of a provocation strategy are acceptable under these conditions. Domestic political considerations are also likely to inuence the type of response that the leadership of a target state chooses. Democracies may be more susceptible to provocation than nondemocracies. Populations that have suffered from terrorist violence will naturally want their government to take action to stop terrorism. Unfortunately, many of the more discriminating tools of counterterrorism, such as inltrating terrorist cells, sharing intelligence with other countries, and arresting individuals, are not visible to the publics these actions serve to protect. Bueno de Mesquita has argued that democratic leaders may have to employ the more public and less discriminating counterterror strategies to prove that their government is taking sufcient action against terrorists, even if these steps are provocative.59 Pressure for a provocative counterresponse may also be particularly acute for more hard-line administrations whose constituents may demand greater action.60 Counterstrategies, therefore, are inuenced in part by the political system from which they emerge. The United States in September 2001 was ripe for provocation, and al-Qaida appears to have understood this. The new administration of George W. Bush was known to be hawkish in its foreign policy and in its attitude toward the use of military power. In a November 2004 videotape, bin Laden bragged that al-Qaida found it easy for us to provoke this administration.61 The strategy appears to be working. A 2004 Pew survey found that international trust in the United States had declined signicantly in response to the invasion of Iraq.62 Similarly, a 2004 report by the International Institute for Strategic Studies found that al-Qaidas recruitment and fundraising efforts had been given a
59. Ethan Bueno de Mesquita, Politics and the Suboptimal Provision of Counterterror, International Organization (forthcoming). 60. On the other hand, more dovish regimes might feel political pressure to take strong visible actions, whereas a regime with hawkish credentials could credibly claim that it was pursuing effective but nonvisible tactics. For a similar logic, see Kenneth A. Schultz, The Politics of Risking Peace: Do Hawks or Doves Deliver the Olive Branch? International Organization, Vol. 59, No. 1 (Winter 2005), pp. 138. 61. Bin Laden, Messages to the World, pp. 241242. 62. Pew Research Center for the People and the Press, Pew Global Attitudes Project: Nine-Nation Survey, A Year after the Iraq War: Mistrust of America in Europe Ever Higher, Muslim Anger Persists, March 16, 2004.

Contending with Terrorism 116

major boost by the U.S. invasion of Iraq.63 In the words of Shibley Telhami, What were seeing now is a disturbing sympathy with al-Qaida coupled with resentment toward the United States.64 The Bush administrations eagerness to overthrow Saddam Hussein, a desire that predated the September 11 attacks, has, in the words of bin Laden, contributed to these remarkable results for al-Qaida.65 best responses to provocation. The best response to provocation is a discriminating strategy that inicts as little collateral damage as possible. Countries should seek out and destroy the terrorists and their immediate backers to reduce the likelihood of future terror attacks, but they must carefully isolate these targets from the general population, which may or may not be sympathetic to the terrorists.66 This type of discriminating response will require superior intelligence capabilities. In this regard, the United States efforts to invest in information-gathering abilities in response to September 11 have been underwhelming. Even the most basic steps, such as developing a deeper pool of expertise in the regional languages, have been slow in coming.67 This stands in contrast to U.S. behavior during the Cold War, when the government sponsored research centers at top universities to analyze every aspect of the Soviet economic, military, and political system. The weakness of the U.S. intelligence apparatus has been most clearly revealed in the inability of the United States to eliminate bin Laden and al-Zawahiri, and in the United States decision to invade Iraq.68 Faulty U.S. intelligence has simultaneously protected al-Qaida leaders from death and led to the destruction of thousands of Muslim civiliansexactly the response al-Qaida was likely seeking. spoiling: sabotaging the peace The goal of a spoiling strategy is to ensure that peace overtures between moderate leaders on the terrorists side and the target government do not suc-

63. See International Institute for Strategic Studies, Strategic Survey, 2003/4: An Evaluation and Forecast of World Affairs (London: Routledge, 2005). 64. Quoted in Dafna Linzer, Poll Shows Growing Arab Rancor at U.S., Washington Post, July 23, 2004. 65. Bob Woodward, Plan of Attack (New York: Simon and Schuster, 2004), pp. 2123; and bin Laden, Messages to the World, pp. 241242. 66. A program of economic and social assistance to these more moderate elements would provide counterevidence that the target is not malicious or evil as the terrorist organizations had claimed. 67. Farah Stockman, Tomorrows Homework: Reading, Writing, and Arabic, Boston Globe, January 6, 2006. 68. For an analysis of obstacles to innovation in U.S. intelligence agencies, see Amy B. Zegart, September 11 and the Adaptation Failure of U.S. Intelligence Agencies, International Security, Vol. 29, No. 4 (Spring 2005), pp. 78111.

The Strategies of Terrorism 117

ceed.69 It works by playing on the mistrust between these two groups and succeeds when one or both parties fail to sign or implement a settlement. It is often employed when the ultimate objective is territorial change. Terrorists resort to a spoiling strategy when relations between two enemies are improving and a peace agreement threatens the terrorists more far-reaching goals. Peace agreements alarm terrorists because they understand that moderate citizens are less likely to support ongoing violence once a compromise agreement between more moderate groups has been reached. Thus, Iranian radicals kidnapped fty-two Americans in Tehran in 1979 not because relations between the United States and Iran were becoming more belligerent, but because three days earlier Irans relatively moderate prime minister, Mehdi Bazargan, met with the U.S. national security adviser, Zbigniew Brzezinski, and the two were photographed shaking hands. From the perspective of the radicals, a real danger of reconciliation existed between the two countries, and violence was used to prevent this.70 A similar problem has hampered Arab-Israeli peace negotiations, as well as talks between Protestants and Catholics in Northern Ireland. A spoiling strategy works by persuading the enemy that moderates on the terrorists side cannot be trusted to abide by a peace deal. Whenever two sides negotiate a peace agreement, there is uncertainty about whether the deal is self-enforcing. Each side fears that even if it honors its commitments, the other side may not, catapulting it back to war on disadvantageous terms. Some Israelis, for example, feared that if Israel returned an additional 13 percent of the West Bank to the Palestinians, as mandated by the 1998 Wye accord, the Palestinian Authority would relaunch its struggle from an improved territorial base. Extremists understand that moderates look for signs that their former enemy will violate an agreement and that targeting these moderates with violence will heighten their fears that they will be exploited. Thus terrorist attacks are designed to persuade a targeted group that the seemingly moderate opposition with whom it negotiated an agreement will not or cannot stop terrorism, and hence cannot be trusted to honor an agreement. Terrorist acts are particularly effective during peace negotiations because opposing parties are naturally distrustful of each others motives and have limited sources of information about each others intentions. Thus, even if moderate leaders are willing to aggressively suppress extremists on their side,

69. Stephen John Stedman, Spoiler Problems in Peace Processes, International Security, Vol. 22, No. 2 (Fall 1997), pp. 553. 70. Lewis, The Revolt of Islam, p. 54.

Contending with Terrorism 118

terrorists know that isolated violence might still convince the target to reject the deal. A reason for this is that the targeted group may not be able to readily observe the extent of the crackdown and must base its judgments primarily on whether terrorism occurs or not. Even a sincere effort at self-policing, therefore, will not necessarily convince the targeted group to proceed with a settlement if a terrorist attack occurs. conditions favorable to spoiling. Terrorists pursuing a spoiling strategy are likely to be more successful when the enemy perceives moderates on their side to be strong and therefore more capable of halting terrorism.71 When an attack occurs, the target cannot be sure whether moderates on the other side can suppress their own extremists but choose not to, or are weak and lack the ability to stop them. Israelis, for example, frequently questioned whether Yasser Arafat was simply unable to stop terrorist attacks against Israel or was unwilling to do so. The weaker the moderates are perceived to be, the less impact a terrorist attack will have on the other sides trust, and the less likely such an attack is to convince them to abandon a peace agreement. The Israeli-Palestinian conict, and in particular the Oslo peace process, has been plagued by spoilers. On the Palestinian side, Hamass violent attacks coincided with the ratication and implementation of accordsoccasions when increased mistrust could thwart progress toward peace. Hamas also stepped up its attacks prior to Israeli elections in 1996 and 2001, in which Labor was the incumbent party, in an effort to persuade Israeli voters to cast their votes for the less cooperative and less trusting hard-line Likud Party.72 Terrorism was especially effective after Arafats 1996 electoral victory, when it became clear to the Israelis that Arafat was, at the time, a popular and powerful leader within the Palestinian community.73 This in turn suggested to the Israelis that Arafat was capable of cracking down aggressively on terrorist violence but was unwilling to do so, a sign that he could not be trusted to keep the peace. best responses to spoiling. When mutual trust is high, a peace settlement can be implemented despite ongoing terrorist acts and the potential vulnerabilities the agreement can create. Trust, however, is rarely high after long conicts, which is why spoilers can strike with a reasonable chance that their

71. Andrew Kydd and Barbara F. Walter, Sabotaging the Peace: The Politics of Extremist Violence, International Organization, Vol. 56, No. 2 (Spring 2002), pp. 263296. 72. Claude Berrebi and Esteban F. Klor, On Terrorism and Electoral Outcomes: Theory and Evidence from the Israeli-Palestinian Conict, Princeton University and Hebrew University of Jerusalem, 2004. 73. Kydd and Walter, Sabotaging the Peace, pp. 279289.

The Strategies of Terrorism 119

attack will be successful. Strategies that build trust and reduce vulnerability are, therefore, the best response to spoiling. Vulnerabilities emerge in peace processes in two ways. Symmetric vulnerabilities occur during the implementation of a deal because both sides must lower their guard. The Israelis, for example, have had to relax controls over the occupied territories, and the Palestinians were obligated to disarm militant groups. Such symmetric vulnerabilities can be eased by third-party monitoring and verication of the peace implementation process. Monitoring can help reduce uncertainty regarding the behavior of the parties. Even better, thirdparty enforcement of the deal can make reneging more costly, increasing condence in the deal and its ultimate success.74 Vulnerabilities can also be longer term and asymmetric. In any peace deal between Israel and the Palestinians, the ability of the Palestinians to harm Israel will inevitably grow as Palestinians build their own state and acquire greater military capabilities. This change in the balance of power can make it difcult for the side that will see an increase in its power to credibly commit not to take advantage of this increase later on. This commitment problem can cause conicts to be prolonged even though there are possible peace agreements that both sides would prefer to war.75 The problem of shifting power can be addressed in at least three ways. First, agreements themselves can be crafted in ways that limit the post-treaty shift in power. Power-sharing agreements such as that between the Liberals and Conservatives to create a single shared presidency in Colombia in 1957 are one example of this. Allowing the defeated side to retain some military capabilities, as Confederate ofcers were allowed to do after the surrender at Appomattox, is another example.76 Second, peace settlements can require the side about to be advantaged to send a costly signal of its honorable intentions, such as providing constitutional protections of minority rights. An example is the Constitutional Law on National Minorities passed in Croatia in 2002,
74. Barbara F. Walter, Committing to Peace: The Successful Settlement of Civil Wars (Princeton, N.J.: Princeton University Press, 2002); and Holger Schmidt, When (and Why) Do Brokers Have To Be Honest? Impartiality and Third-Party Support for Peace Implementation after Civil Wars, 1945 1999, Georgetown University, 2004. 75. Fearon, Rationalist Explanations for War; James D. Fearon, Commitment Problems and the Spread of Ethnic Conict, in David A. Lake and Donald Rothchild, eds., The International Spread of Ethnic Conict: Fear, Diffusion, and Escalation (Princeton, N.J.: Princeton University Press, 1998); and Robert Powell, The Inefcient Use of Power: Costly Conict with Complete Information, American Political Science Review, Vol. 98, No. 2 (May 2004), pp. 231241. 76. As part of the terms of surrender, Confederate ofcers were allowed to keep their sidearms and personal property (including their horses) and return home.

Contending with Terrorism 120

which protects the right of minorities to obtain an education in their own language. Finally, parties can credibly commit to an agreement by participating in international institutions that insist on the protection of minority rights. A government that is willing to join the European Union effectively constrains itself from exploiting a minority group because of the high costs to that government of being ejected from the group. outbidding: zealots versus sellouts Outbidding arises when two key conditions hold: two or more domestic parties are competing for leadership of their side, and the general population is uncertain about which of the groups best represents their interests.77 The competition between Hamas and Fatah is a classic case where two groups vie for the support of the Palestinian citizens and where the average Palestinian is uncertain about which side he or she ought to back. If citizens had full information about the preferences of the competing groups, an outbidding strategy would be unnecessary and ineffective; citizens would simply support the group that best aligned with their own interests. In reality, however, citizens cannot be sure if the group competing for power truly represents their preferences. The group could be a strong and resolute defender of the cause (zealots) or weak and ineffective stooges of the enemy (sellouts). If citizens support zealots, they get a strong champion but with some risk that they will be dragged into a confrontation with the enemy that they end up losing. If citizens support sellouts, they get peace but at the price of accepting a worse outcome than might have been achieved with additional armed struggle. Groups competing for power have an incentive to signal that they are zealots rather than sellouts. Terrorist attacks can serve this function by signaling that a group has the will to continue the armed struggle despite its costs. Three reasons help to explain why groups are likely to be rewarded for being more militant rather than less. First, in bargaining contexts, it is often useful to be represented by an agent who is more hard-line than oneself. Hard-line agents will reject deals that one would accept, which will force the adversary

77. For the most extensive treatment of terrorism and outbidding, see Mia Bloom, Dying to Kill: The Allure of Suicide Terrorism (New York: Columbia University Press, 2005). See also Stuart J. Kaufman, Spiraling to Ethnic War: Elites, Masses, and Moscow in Moldovas Civil War, International Security, Vol. 21, No. 2 (Fall 1996), pp. 108138.

The Strategies of Terrorism 121

to make a better offer than one would get by representing oneself in the negotiations.78 Palestinians might therefore prefer Hamas as a negotiating agent with Israel because it has a reputation for resolve and will reject inferior deals. Second, uncertainty may also exist about the type of adversary the population and its competing groups are facing. If the population believes there is some chance that their adversary is untrustworthy (unwilling to compromise under any condition), then they know that conict may be inevitable, in which case being represented by zealots may be advantageous.79 A third factor that may favor outbidding is that ofce-holding itself may produce incentives to sell out. Here, the problem lies with the benets groups receive once in ofce (i.e., income and power). Citizens fear that their leaders, once in ofce, may betray important principles and decide to settle with the enemy on unfavorable terms. They know that holding ofce skews ones preferences toward selling out, but they remain unsure about which of their leaders is most likely to give in. Terrorist organizations exploit this uncertainty by using violence to signal their commitment to a cause. Being perceived as more extreme than the median voter works to the terrorists benet because it balances out the tempering effect of being in ofce. An interesting aspect of the outbidding strategy is that the enemy is only tangentially related to the strategic interaction. In fact, an attack motivated by outbidding may not even be designed to achieve any goal related to the enemy, such as inducing a concession or scuttling a peace treaty. The process is almost entirely concerned with the signal it sends to domestic audiences uncertain about their own leadership and its commitment to a cause. As such, outbidding provides a potential explanation for terrorist attacks that continue even when they seem unable to produce any real results. conditions favorable to outbidding. Outbidding will be favored when multiple groups are competing for the allegiance of a similar demographic base of support. In Peru, the 1970s saw the development of a number of leftist groups seeking to represent the poor and indigenous population. When the military turned over power to an elected government in 1980, the Shining Path took up an armed struggle to distinguish itself from groups that chose to pur78. Abhinay Muthoo, Bargaining Theory with Applications (Cambridge: Cambridge University Press, 1999), p. 230. 79. Rui J.P. de Figueiredo Jr. and Barry R. Weingast, The Rationality of Fear: Political Opportunism and Ethnic Conict, in Barbara F. Walter and Jack Snyder, eds., Civil Wars, Insecurity, and Intervention (New York: Columbia University Press, 1999), pp. 261302.

Contending with Terrorism 122

sue electoral politics.80 It also embarked on an assassination campaign designed to weaken rival leftist groups and intimidate their followers. When organizations encounter less competition for the support of their main constituents, outbidding will be less appealing. best responses to outbidding. One solution to the problem of outbidding would be to eliminate the struggle for power by encouraging competing groups to consolidate into a unied opposition. If competition among resistance groups is eliminated, the incentive for outbidding also disappears. The downside of this counterstrategy is that a unied opposition may be stronger than a divided one. United oppositions, however, can make peace and deliver, whereas divided ones may face greater structural disincentives to do so. An alternative strategy for the government to pursue in the face of outbidding is to validate the strategy chosen by nonviolent groups by granting them concessions and attempting to satisfy the demands of their constituents. If outbidding can be shown to yield poor results in comparison to playing within the system, groups may be persuaded to abandon the strategy. As in the case of the Shining Path, this may require providing physical protection to competing groups in case the outbidder turns to intimidation in its competition with less violent rivals. In general, any steps that can be taken to make the non-outbidding groups seem successful (e.g., channeling resources and government services to their constituents) will also help undermine the outbidders. The high turnout in the December 2005 Iraqi election in Sunni-dominated regions may indicate that outbidding is beginning to fail in the communities most strongly opposed to the new political system.81

Conclusion
Terrorist violence is a form of costly signaling by which terrorists attempt to inuence the beliefs of their enemy and the population they represent or wish to control. They use violence to signal their strength and resolve in an effort to produce concessions from their enemy and obedience and support from their followers. They also attack both to sow mistrust between moderates who
80. James Ron, Ideology in Context: Explaining Sendero Luminosos Tactical Escalation, Journal of Peace Research, Vol. 38, No. 5 (September 2001), p. 582. 81. Dexter Filkins, Iraqis, Including Sunnis, Vote in Large Numbers on Calm Day, New York Times, December 16, 2005.

The Strategies of Terrorism 123

might want to make peace and to provoke a reaction that makes the enemy appear barbarous and untrustworthy. In this article, we have laid out the ve main goals terrorist organizations seek and the ve most important terrorist strategies, and we have outlined when they are likely to be tried and what the best counterstrategies might look like. What becomes clear in this brief analysis is that a deeper study of each of the ve strategies is needed to reveal the nuanced ways in which terrorism works, and to rene responses to it. We conclude by highlighting two variables that will be important in any such analysis, and by a nal reection on counterterror policies that are strategically independent or not predicated on the specic strategy being used. The rst variable is information. It has long been a truism that the central front in counterinsurgency warfare is the information front. The same is true in terrorism. Costly signaling is pointless in the absence of uncertainty on the part of the recipient of the signal. Attrition is designed to convince the target that the costs of maintaining a policy are not worth the gains; if the target already knew this, it would have ceded the issue without an attack being launched. Provocation is designed to goad the target into retaliating indiscriminately (because it lacks information to discriminate), which will persuade the population that the target is malevolent (because it is uncertain of the targets intentions). The other strategies are similarly predicated on uncertainty, intelligence, learning, and communication. Thus, it bears emphasizing that the problem of terrorism is not a problem of applying force per se, but one of acquiring intelligence and affecting beliefs. With the right information, the proper application of force is comparatively straightforward. The struggle against terrorism is, therefore, not usefully guided by the metaphor of a war on terrorism any more than policies designed to alleviate poverty are usefully guided by the metaphor of a war on poverty or narcotics policy by a war on drugs. The struggle against terrorism can more usefully be thought of as a struggle to collect and disseminate reliable information in environments fraught with uncertainty. The second important variable is regime type. Democracies have been the sole targets of attritional suicide bombing campaigns, whereas authoritarian regimes such as those in Algeria routinely face campaigns by rebel groups pursuing an intimidation strategy. Democracies also seem to be more susceptible to attrition and provocation strategies. This type of variation cries out for deeper analysis of the strengths and weakness of different regime types in the

Contending with Terrorism 124

face of different terrorist strategies. Our analysis suggests that democracies are more likely to be sensitive to the costs of terrorist attacks, to grant concessions to terrorists so as to limit future attacks, to be constrained in their ability to pursue a lengthy attritional campaign against an organization, but also to be under greater pressure to do something. This does not mean that all democracies will behave incorrectly in the face of terrorist attacks all the time. Democratic regimes may possess certain structural features, however, that make them attractive targets for terrorism. Finally, we realize that our discussion is only a beginning and that further elaboration of each of the strategies and their corresponding counterstrategies awaits future research. We also understand that not all counterterrorism policies are predicated on the specic strategy terrorists pursue. Our analysis is at the middling level of strategic interaction. At the tactical level are all the tools of intelligence gathering and target defense that make sense no matter what the terrorists strategy is. At the higher level are the primary sources of terrorism such as poverty, education, international conict, and chauvinistic indoctrination that enable terrorist organizations to operate and survive in the rst place. Our aim in this article has been to try to understand why these organizations choose certain forms of violence, and how this violence serves their larger purposes. The United States has the ability to reduce the likelihood of additional attacks on its territory and citizens. But it will be much more successful if it rst understands the goals terrorists are seeking and the underlying strategic logic by which a plane ying into a skyscraper might deliver the desired response.

Why Terrorism Does Not Work

Why Terrorism Does Not Work

Max Abrahms

errorist groups attack civilians to coerce their governments into making policy concessions, but does this strategy work?1 If target countries systematically resist rewarding terrorism, the international community is armed with a powerful message to deter groups from terrorizing civilians. The prevailing view within the eld of political science, however, is that terrorism is an effective coercive strategy. The implications of this perspective are grim; as target countries are routinely coerced into making important strategic and ideological concessions to terrorists, their victories will reinforce the strategic logic for groups to attack civilians, spawning even more terrorist attacks.2 This pessimistic outlook is unwarranted; there has been scant empirical research on whether terrorism is a winning coercive strategy, that is, whether groups tend to exact policy concessions from governments by attacking their
Max Abrahms is a doctoral candidate in political science at the University of California, Los Angeles. He wrote this article when he was a Research Associate in terrorism studies at the Belfer Center for Science and International Affairs in the John F. Kennedy School of Government at Harvard University. The author would like to thank the following individuals for offering valuable advice and comments: Peter Krause, Deborah Larson, Marcy McCullaugh, David Rapoport, Steven Spiegel, Marc Trachtenberg, Micah Zenko, and especially Robert Goldenberg, Evan Spring, and the anonymous reviewers for International Security. 1. For decades, terrorism specialists and political scientists have recognized that groups use terrorism to achieve policy objectives. See, for example, Martha Crenshaw, The Causes of Terrorism, Comparative Politics, Vol. 13, No. 4 (July 1981), p. 379. 2. The logic of this argument is explored in Ehud Sprinzak, Rational Fanatics, Foreign Policy, No. 120 (SeptemberOctober 2000), p. 68; Robert A. Pape, The Strategic Logic of Suicide Terrorism, American Political Science Review, Vol. 97, No. 3 (August 2003), pp. 1314; and Robert A. Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (New York: Random House, 2005), pp. 61, 7576. See also Alan Dershowitz, Why Terrorism Works: Understanding the Threat, Responding to the Challenge (New Haven, Conn.: Yale University Press, 2002). In recent years, numerous writers have argued that terrorism is both rational and effective. See, for example, David A. Lake, Rational Extremism: Understanding Terrorism in the Twenty-rst Century, Dialogue-IO, Vol. 1 (January 2002), p. 15; Andrew Kydd and Barbara Walter, Sabotaging the Peace: The Politics of Extremist Violence, International Organization, Vol. 56, No. 2 (April 2002), p. 263; Andrew Kydd and Barbara Walter, The Strategies of Terrorism, International Security, Vol. 31, No. 1 (Summer 2006), p. 49; Arie W. Kruglanski and Shira Fishman, The Psychology of Terrorism: Syndrome versus Tool Perspectives, Terrorism and Political Violence, Vol. 18, No. 2 (Summer 2006), p. 207; Noam Chomsky, The New War against Terror (transcribed from audio), recorded at the Technology and Culture Forum at the Massachusetts Institute of Technology, October 18, 2001, http:/ /www.chomsky.info/talks/ 20011018.htm; and Scott Atran, Trends in Suicide Terrorism: Sense and Nonsense, paper presented to the World Federation of Scientists Permanent Monitoring Panel on terrorism, Erice, Sicily, August 2004, http:/ /www.sitemaker.umich.edu/satran/les/atran-trends.pdf.
International Security, Vol. 31, No. 2 (Fall 2006), pp. 4278 2006 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

125

Contending with Terrorism 126

civilian populations. In the 1980s, Martha Crenshaw observed that the outcomes of campaigns of terrorism have been largely ignored, as most analyses have emphasized the causes and forms rather than the consequences of terrorism.3 Ted Robert Gurr added that terrorisms policy effectiveness is a subject on which little national-level research has been done, systematically or otherwise.4 This lacuna within terrorism studies is both a symptom and a cause of the lack of data sets with coded information on the outcomes of terrorist campaigns.5 Within the past several years, numerous scholars have purported to show that terrorism is an effective coercive strategy, but their research invariably rests on game-theoretic models, single case studies, or a handful of well-known terrorist victories.6 To date, political scientists have neither analyzed the outcomes of a large number of terrorist campaigns nor attempted to specify the antecedent conditions for terrorism to work. In light of its policy relevance, terrorisms record in coercing policy change requires further empirical analysis. This study analyzes the political plights of twenty-eight terrorist groups the complete list of foreign terrorist organizations (FTOs) as designated by the U.S. Department of State since 2001.7 The data yield two unexpected ndings. First, the groups accomplished their forty-two policy objectives only 7 percent of the time. Second, although the groups achieved certain types of policy objectives more than others, the key variable for terrorist success was a tactical one: target selection. Groups whose attacks on civilian targets outnumbered attacks on military targets systematically failed to achieve their policy objectives, regardless of their nature. These ndings suggest that (1) terrorist groups rarely achieve their policy objectives, and (2) the poor success rate is inherent

3. Martha Crenshaw, ed., Terrorism, Legitimacy, and Power: The Consequences of Political Violence (Middletown, Conn.: Wesleyan University Press, 1983), p. 5. 4. Ted Robert Gurr, Empirical Research on Political Terrorism, in Robert O. Slater and Michael Stohl, eds., Current Perspectives on International Terrorism (New York: St. Martins, 1988), p. 125. 5. See ibid., p. 120. 6. See, for example, Max Abrahms, Are Terrorists Really Rational? The Palestinian Example, Orbis, Vol. 48, No. 3 (Summer 2004), pp. 533549; Max Abrahms, Al-Qaidas Scorecard: A Progress Report on al-Qaidas Objectives, Studies in Conict and Terrorism; Sprinzak, Rational Fanatics, p. 68; Lake, Rational Extremism, p. 15; Kydd and Walter, The Strategies of Terrorism, p. 49; Pape, Dying to Win, p. 40; and Pape, The Strategic Logic of Suicide Terrorism, p. 13. The sample Pape used to determine the motivations of suicide terrorist groups is impressive, incorporating data from every suicide attack from 1980 to 2003. His secondary argument on terrorisms effectiveness relies on a less convincing sample of cases. For a critique of the methods Pape used to analyze the effectiveness of suicide terrorist groups, see Max Abrahms, Dying to Win, Middle East Policy, Vol. 12, No. 4 (Winter 2005), pp. 176178. 7. U.S. Department of State, Foreign Terrorist Organizations (FTOs), October 11, 2005, http:/ / www.state.gov/s/ct/rls/fs/37191.htm.

Why Terrorism Does Not Work 127

to the tactic of terrorism itself. Together, the data challenge the dominant scholarly opinion that terrorism is strategically rational behavior.8 The bulk of the article develops a theory to explain why terrorist groups are unable to achieve their policy objectives by targeting civilians. This article has ve main sections. The rst section summarizes the conventional wisdom that terrorism is an effective coercive strategy and highlights the decit of empirical research sustaining this position. The second section explicates the methods used to assess the outcomes of the forty-two terrorist objectives included in this study and nds that terrorist success rates are actually extremely low. The third section examines the antecedent conditions for terrorism to work. It demonstrates that although terrorist groups are more likely to succeed in coercing target countries into making territorial concessions than ideological concessions, groups that primarily attack civilian targets do not achieve their policy objectives, regardless of their nature. The fourth section develops a theory derived from the social psychology literature for why terrorist groups that target civilians are unable to compel policy change. Its external validity is then tested against three case studies: the September 1999 Russian apartment bombings, the September 11, 2001, attacks on the United States, and Palestinian terrorism in the rst intifada. The article concludes with four policy implications for the war on terrorism and suggestions for future research.

The Notion That Terrorism Works

Writers are increasingly contending that terrorism is an effective coercive strategy. In his 2002 best-seller, Why Terrorism Works, Alan Dershowitz argues that Palestinian gains since the early 1970s reveal that terrorism works and is thus an entirely rational choice to achieve a political objective.9 David Lake
8. Herbert Simon noted that there are two types of rationality. Substantive rationality is based on the actual achievement of given goals. By contrast, procedural rationality depends only on the [thought] process that generated it. In other words, substantive rationality is concerned with the consequences of the decision, whereas procedural rationality makes no claim that the actor correctly anticipates the consequences of his decision. Simon, From Substantive to Procedural Rationality, in Spiro Latsis, ed., Method and Appraisal in Economics (Cambridge: Cambridge University Press, 1976), pp. 130131; and Simon, Models of Bounded Rationality, Vol. 3: Empirically Grounded Economic Research (Cambridge, Mass.: MIT Press, 1997), pp. 89. My article addresses the question of whether terrorism is substantively rational behavior, not whether it is procedurally rational. 9. Dershowitz, Why Terrorism Works, p. 86. In the past several years, a urry of editorials has likewise warned that the tendency for states to reward terrorism is encouraging other groups to use it. See John Derbyshire, Terrorism Works, National Review Online, October 12, 2000; Evelyn Gordon, Terrorism Works, Jerusalem Post, July 14, 2005; and R.W. Johnson, Why Bush Might Yet Give In

Contending with Terrorism 128

recently adapted James Fearons rationalist bargaining model to argue that terrorism is a rational and strategic tactic because it enables terrorists to achieve a superior bargain by increasing their capabilities relative to those of target countries.10 Based on their game-theoretic model and case study on Hamas, Andrew Kydd and Barbara Walter likewise conclude that terrorist groups are surprisingly successful in their aims.11 According to Scott Atran, terrorist groups generally achieve their policy objectives. As evidence, he notes that the Lebanese-based Shiite terrorist group, Hezbollah, successfully compelled the United States and France to withdraw their remaining forces from Lebanon in 1984 and that in 1990 the Tamil Tigers of Sri Lanka wrested control of Tamil areas from the Sinhalese-dominated government.12 For Ehud Sprinzak, the plights of Hezbollah and the Tamil Tigers testify to terrorisms gruesome effectiveness, which explains its growing popularity since the mid-1980s.13 Robert Pape has developed this thesis in a prominent article that was recently expanded into a major book.14 Pape contends that over the past two decades, suicide terrorism has been rising largely because terrorists have learned that it pays.15 He reports that from 1980 to 2003, six of the eleven terrorist campaigns in his sample were associated with signicant policy changes by the target state and that a 50 percent success rate is remarkable.16 The perception that terrorism is an effective method of coercion, he afrms, is thus grounded in reasonable assessments of the relationship between terrorists coercive efforts and the political gains that the terrorists have achieved.17 Papes research, although conned to suicide terrorist groups, is frequently cited as evidence that terrorism in general is effective in achieving a terrorist groups political aims.18 This emerging consensus lacks a rm empirical basis. The notion that terrorism is an effective coercive instrument is sustained by either single case studies or a few well-known terrorist victories, namely, by Hezbollah, the Tamil Tigers, and Palestinian terrorist groups. Papes research appears to offer the
to the Terrorists, Daily Telegraph (London), September 29, 2001. See also Kruglanski and Fishman, The Psychology of Terrorism, p. 207; and Chomsky, The New War against Terror. 10. Lake, Rational Extremism, pp. 20, 15. 11. Kydd and Walter, Sabotaging the Peace, p. 264. 12. Atran, Trends in Suicide Terrorism. 13. Sprinzak, Rational Fanatics, p. 68. 14. Pape, The Strategic Logic of Suicide Terrorism; and Pape, Dying to Win. 15. Pape, The Strategic Logic of Suicide Terrorism, p. 343. 16. Ibid., p. 9; and Pape, Dying to Win, pp. 6465. 17. Pape, Dying to Win, pp. 61, 6465. 18. See, for example, Kydd and Walter, The Strategies of Terrorism, p. 49.

Why Terrorism Does Not Work 129

strongest evidence that terrorist groups regularly accomplish their policy objectives, but on closer analysis his thesis is also empirically weak. Not only is his sample of terrorist campaigns modest, but they targeted only a handful of countries: ten of the eleven campaigns analyzed were directed against the same three countries (Israel, Sri Lanka, and Turkey), with six of the campaigns directed against the same country (Israel).19 More important, Pape does not examine whether the terrorist campaigns achieved their core policy objectives. In his assessment of Palestinian terrorist campaigns, for example, he counts the limited withdrawals of the Israel Defense Forces from parts of the Gaza Strip and the West Bank in 1994 as two separate terrorist victories, ignoring the 167 percent increase in the number of Israeli settlers during this periodthe most visible sign of Israeli occupation.20 Similarly, he counts as a victory the Israeli decision to release Hamas leader Sheik Ahmed Yassin from prison in October 1997, ignoring the hundreds of imprisonments and targeted assassinations of Palestinian terrorists throughout the Oslo peace process.21 Papes data therefore reveal only that select terrorist campaigns have occasionally scored tactical victories, not that terrorism is an effective strategy for groups to achieve their policy objectives. The two sections that follow are intended to help bridge the gap between the growing interest in terrorisms efcacy and the current weakness of empirical research on this topic.

Measuring Terrorisms Effectiveness

Terrorist campaigns come in two varieties: strategic terrorism aims to coerce a government into changing its policies; redemptive terrorism is intended solely to attain specic human or material resources such as prisoners or money.22 Because my focus is on terrorisms ability to compel policy change, terrorism in this study refers only to strategic terrorism campaigns. Terrorisms effectiveness can be measured along two dimensions: combat effectiveness describes the level of damage inicted by the coercing power; strategic effectiveness refers to the extent to which the coercing power achieves its policy objectives.23 This study is conned to analyzing the notion that terrorism is strategically ef19. Pape, The Strategic Logic of Suicide Terrorism, p. 9. 20. Abrahms, Dying to Win, p. 177. 21. See Abrahms, Dying to Win. 22. Gary C. Gambill, The Balance of Terror: War by Other Means in the Contemporary Middle East, Journal of Palestine Studies, Vol. 28, No. 1 (Autumn 1998), p. 61. 23. Robert A. Pape, Bombing to Win: Air Power and Coercion in War (Ithaca, N.Y.: Cornell University Press, 1996), pp. 5657.

Contending with Terrorism 130

fective, not whether it succeeds on an operational or tactical level.24 Finally, because this study is concerned with terrorisms effect on the target country, intermediate objectivesnamely, the ability of terrorist groups to gain international attention and supportare outside the scope of analysis.25 This study analyzes the strategic effectiveness of the twenty-eight terrorist groups designated by the U.S. Department of State as foreign terrorist organizations since 2001. The only selection bias would come from the State Department. Using this list provides a check against selecting cases on the dependent variable, which would articially inate the success rate because the most well known policy outcomes involve terrorist victories (e.g., the U.S. withdrawal from southern Lebanon in 1984). Furthermore, because all of the terrorist groups have remained active since 2001, ample time has been allowed for each group to make progress on achieving its policy goals, thereby reducing the possibility of articially deating the success rate through too small a time frame. In fact, the terrorist groups have had signicantly more time than ve years to accomplish their policy objectives: the groups, on average, have been active since 1978; the majority has practiced terrorism since the 1960s and 1970s; and only four were established after 1990. For terrorist groups, policy outcomes are easier to assess than policy objectives. Instead of arbitrarily dening the objectives of the terrorist groups in this study, I dene them as the terrorists do. In general, the stated objectives of terrorist groups are a stable and reliable indicator of their actual intentions. This assumption undergirds the widely accepted view within terrorism studies that groups use terrorism as a communication strategy to convey to target countries the costs of noncompliance.26 Because these groups seek political change
24. See James T. Tedeschi, A Reinterpretation of Research on Aggression, Psychological Bulletin, Vol. 81, No. 9 (September 1974), p. 562. 25. There is little debate that terrorism often facilitates the achievement of intermediate objectives. This position gained acceptance after the 1967 Arab-Israeli war, when a spate of terrorist attacks galvanized Palestinian nationalist sentiment and propelled the Palestinian cause onto the international agenda. In September 1969, just three months after the Popular Front for the Liberation of Palestine hijacked a U.S. jet departing from Athens, the United Nations General Assembly recognized the inalienable rights of the Palestinian people. Eighteen months after Palestinian terrorists killed eleven Israeli Olympians at the 1972 Munich Games, Yassir Arafat was ofcially invited to speak before this body. Bruce Hoffman has observed, It is doubtful whether the terrorists could ever have received this success had they not resorted to international terrorism. Hoffman, Inside Terrorism (New York: Columbia University Press, 1998), p. 75. See also Abrahms, Are Terrorists Really Rational? p. 341; and Thomas C. Schelling, What Purposes Can International Terrorism Serve? in R.G. Frey and Christopher W. Morris, eds., Violence, Terrorism, and Justice (New York: Cambridge University Press, 1991), p. 20. 26. See Crenshaw, The Causes of Terrorism, p. 379; Ronald D. Crelinsten, Terrorism as Political Communication: The Relationship between the Controller and the Controlled, in Paul Wilkinson and Alasdair M. Stewart, eds., Contemporary Research on Terrorism (Aberdeen, Scotland: Aberdeen

Why Terrorism Does Not Work 131

and because their stated objectives represent their intentions, terrorisms effectiveness is measured by comparing their stated objectives to policy outcomes. A potential objection to this approach is that terrorists possess extreme policy goals relative to those of their supporters, and thus terrorist campaigns may be judged unsuccessful even when they compel policy changes of signicance to their broader community. What distinguish terrorists from moderates, however, are typically not their policy goals, but the belief that terrorism is the optimal means to achieve them.27 As Pape has observed, It is not that terrorists pursue radical goals relative to those of their supporters. Rather, it is that terrorists are simply the members of their societies who are the most optimistic about the usefulness of violence for achieving goals that many, and often most, support.28 There are no broadly based data sets with coded information on the objectives of terrorist campaigns, but those ascribed to the terrorist groups in this study are all found in standard descriptions of them, such as in RANDs MIPT Terrorism Knowledge Base and the Federation of American Scientists Directory of Terrorist Organizations (see Table 1).29 To capture the range of policy outcomes, this study employs a four-tiered rating scale. A total success denotes the full attainment of a terrorist groups policy objective. Conversely, no success describes a scenario in which a terrorist group does not make any perceptible progress on realizing its stated objective. Middling achievements are designated as either a partial success or a limited success in descending degrees of effectiveness. Several groups are counted more than once to reect their multiple policy objectives. Hezbollah, for example, is credited with two policy successes: repelling the multinational peacekeepers and Israelis from southern Lebanon in 1984 and again in 2000. By contrast, Revolutionary Nuclei is tagged with two policy failures: its inability either to spark a communist revolution in Greece or to sever U.S.-Greek relations.
University Press, 1987), pp. 331; Bruce Hoffman and Gordon H. McCormick, Terrorism, Signaling, and Suicide Attack, Studies in Conict and Terrorism, Vol. 27, No. 4 (July/August 2004), pp. 243281; Max Abrahms, Al-Qaedas Miscommunication War: The Terrorism Paradox, Terrorism and Political Violence, Vol. 17, No. 4 (Autumn 2005), pp. 529549; and Kydd and Walter, The Strategies of Terrorism, p. 59. 27. U.S. Department of Defense, The National Military Strategic Plan for the War on Terrorism, February 1, 2006, p. 3. 28. Pape, The Strategic Logic of Suicide Terrorism, p. 7. 29. Gurr, Empirical Research on Political Terrorism, p. 120; RAND, MIPT Terrorism Knowledge Base: A Comprehensive Databank of Global Terrorist Incidents and Organizations, Groups Subcategories, http:/ /www.tkb.org/Category.jsp?catID?1; and Federation of American Scientists, Liberation Movements, Terrorist Organizations, Substance Cartels, and Other Para-state Entities, http:/ /www.fas.org/irp/world/para/. See also Kydd and Walter, The Strategies of Terrorism, pp. 5455.

Table 1. Terrorist Groups: Objectives, Targets, and Outcomes Type Maximalist Maximalist Limited Idiosyncratic Idiosyncratic Idiosyncratic Maximalist Idiosyncratic Maximalist Maximalist Idiosyncratic Maximalist Maximalist Maximalist Limited Maximalist Maximalist Limited Idiosyncratic Limited Limited Maximalist Maximalist Maximalist Maximalist Maximalist Idiosyncratic Maximalist Maximalist Maximalist Maximalist Maximalist Civilian Civilian Civilian Military Military Military Military Military Military Civilian Civilian Civilian Civilian Military Civilian Civilian Military Military No success Limited success No success No success No success Total success Total success No success No success No success Limited success No success No success No success No success No success No success No success Civilian Civilian Civilian Civilian Civilian Civilian Civilian Civilian Civilian Civilian Civilian Civilian Civilian Military No success No success Limited success No success No success No success No success No success No success No success No success No success No success Limited success Main Target Outcome

Group

Objective

Destroy Israel Establish Islamic state in Philippines Expel the United States from Persian Gulf Sever U.S.-Israel relations Sever U.S.-apostate relations Spare Muslims from Crusader wars Establish Islamic state in Algeria Eliminate left-wing insurgents Establish utopian society in Japan Establish Marxism in Turkey Sever U.S.-Turkish relations Establish Islamic state in Egypt Establish Marxism in Colombia Establish peasant rule in Colombia

Contending with Terrorism 132

Abu Nidal Organization Abu Sayyaf Group Al-Qaida Al-Qaida Al-Qaida Al-Qaida Armed Islamic Group United Forces of Colombia Aum Shinrikyo Peoples Liberation Front Peoples Liberation Front Egyptian Islamic Jihad National Liberation Army Revolutionary Armed Forces of Colombia Fatherland and Liberty Hamas Hamas Harakat ul-Mujahidin Harakat ul-Mujahidin Hezbollah (Lebanese) Hezbollah (Lebanese) Hezbollah (Lebanese) Islamic Movement of Uzbekistan Islamic Group Islamic Jihad Islamic Jihad Kach Mujahideen-e-Khalq Popular Front for the Liberation of Palestine (PFLP) PFLP PFLPGeneral Command PFLPGeneral Command

Establish Basque state Establish state in historic Palestine Destroy Israel Rule Kashmir Eliminate Indian insurgents Expel peacekeepers Expel Israel Destroy Israel Establish Islamic state in Uzbekistan Establish Islamic state in Egypt Establish state in historic Palestine Destroy Israel Transfer Palestinians from Israel End clerical rule in Iran Destroy Israel

Establish Marxist Palestine Destroy Israel Establish Marxist Palestine

Table 1. Continued Type Limited Maximalist Maximalist Limited Maximalist Idiosyncratic Maximalist Idiosyncratic Maximalist Limited Civilian Civilian Civilian Military Military Military Civilian Civilian Civilian Military Main Target Outcome No success No success No success No success No success No success No success No success No success Partial success

Group

Objective

Kurdistan Workers Party Kurdistan Workers Party Palestine Liberation Front Real Irish Republican Army Revolutionary Nuclei Revolutionary Nuclei Seventeen November Seventeen November Shining Path Tamil Tigers

Establish Kurdish state in Middle East Establish communism in Turkey Destroy Israel Establish Irish unification Establish Marxism in Greece Sever U.S.-Greek relations Establish Marxism in Greece Sever U.S.-Greek relations Establish communism in Peru Establish Tamil state

SOURCES: RAND, MIPT Terrorism Knowledge Base, http://www.tkb.org/Home.jsp; and Federation of American Scientists, Liberation

Why Terrorism Does Not Work 133

Movements, Terrorist Organizations, Substance Cartels, and Other Para-state Entities, http://www.fas.org/irp/world/para.

Contending with Terrorism 134

To construct a hard test for the argument that terrorism is an ineffective means of coercion, I afforded generous conditions to limit the number of policy failures. First, for analytic purposes both a total success and a partial success are counted as policy successes, while only completely unsuccessful outcomes (no successes) are counted as failures. A limited success is counted as neither a success nor a failure, even though the terrorist group invariably faces criticism from its natural constituency that the means employed have been ineffective, or even counterproductive. Thus, a policy objective is deemed a success even if the terrorist group was only partially successful in accomplishing it, whereas an objective receives a failing grade only if the group has not made any noticeable progress toward achieving it. Second, an objective is judged successful even if the group accomplished it before 2001, the year the State Department assembled its ofcial list of foreign terrorist organizations. Third, all policy successes are attributed to terrorism as the causal factor, regardless of whether important intervening variables, such as a peace process, may have contributed to the outcome. Fourth, terrorist groups are not charged with additional penalties for provoking responses from the target country that could be considered counterproductive to their policy goals.30 Fifth, the objectives of al-Qaida afliates are limited to their nationalist struggles. Groups such as the Kashmiri Harakat ul-Mujahidin and the Egyptian Islamic Jihad are not evaluated on their ability to sever U.S.-Israeli relations, for example, even though many of their supporters claim to support this goal. Based on their policy platforms, the twenty-eight terrorist groups examined in this study have a combined forty-two policy objectives, a healthy sample of cases for analysis. Several well-known terrorist campaigns have accomplished their objectives. As frequently noted, Hezbollah successfully coerced the multinational peacekeepers and Israelis from southern Lebanon in 1984 and 2000, and the Tamil Tigers won control over the northern and eastern coastal areas of Sri Lanka from 1990 on. In the aggregate, however, the terrorist groups achieved their main policy objectives only three out of forty-two timesa 7 percent success rate.31 Within the coercion literature, this rate of success is considered extremely low. It is substantially lower, for example, than even the success rate of economic sanctions, which are widely regarded as only mini30. That the Israel Defense Forces reoccupied large sections of the West Bank in April 2002 in response to terrorist activity, for example, is relevant only insofar as it may have prevented Palestinian terrorist organizations from accomplishing their stated objectives. 31. Even when limited successes are counted as policy successes, the success rate is only 17 percent.

Why Terrorism Does Not Work 135

mally effective.32 The most authoritative study on economic sanctions has found a success rate of 34 percentnearly ve times greater than the success rate of the terrorist groups examined in my studywhile other studies have determined that economic sanctions accomplish their policy objectives at an even higher rate.33 Compared to even minimally effective methods of coercion, terrorism is thus a decidedly unprotable coercive instrument.34

When Terrorism Works: The Paramountcy of Target Selection

The terrorist groups in this study were far more likely to achieve certain types of policy objectives than others. Yet predicting the outcomes of terrorist campaigns based on their policy goals is problematic. The objectives of terrorist groups are sometimes difcult to code. More important, the terrorist groups did not tend to achieve their policy aims regardless of their nature. The key variable for terrorist success was a tactical one: target selection. Groups whose attacks on civilian targets outnumbered attacks on military targets systematically failed to achieve their policy objectives. Below I examine the effects of objective type and target selection on the outcomes of the forty-two terrorist campaigns included in this study. importance of objective type Since the mid-1960s, international mediation theorists have asserted that limited objectives are more conducive to locating a mutually acceptable resolution than disputes over maximalist objectives, which foreclose a bargaining range.35 In the international mediation literature, limited objectives typically refer to demands over territory (and other natural resources); maximalist ob32. Robert A. Pape, Why Economic Sanctions Do Not Work, International Security, Vol. 22, No. 2 (Fall 1997), p. 99. 33. Robert A. Hart, Democracy and the Successful Use of Economic Sanctions, Political Research Quarterly, Vol. 53, No. 2 (June 2000), p. 279. 34. To say that the coercive skills of terrorist groups are poor relative to those of states because these groups are considerably weaker is to concede the point that compelling policy change is a low-probability affair for terrorist groups. 35. Robert A. Dahl, Who Governs? Democracy and Power in an American City (New Haven, Conn.: Yale University Press, 1961); Theodore J. Lowi, American Business, Public Policy, Case Studies, and Political Theory, World Politics, Vol. 16, No. 3 (July 1964), pp. 677715; and Marvin Ott, Mediation as a Method of Conict Resolution, Two Cases, International Organization, Vol. 26, No. 4 (Autumn 1972), p. 613. The distinction between limited and maximalist issues is also expressed in terms of tangible versus intangible issues, respectively. See John A. Vasquez, The Tangibility of Issues and Global Conict: A Test of Rosenaus Issue Area Typology, Journal of Peace Research, Vol. 20, No. 2 (Summer 1983), p. 179; and I. William Zartman, Elusive Peace: Negotiating an End to Civil Conicts (Washington, D.C.: Brookings, 1995).

Contending with Terrorism 136

jectives, on the other hand, refer to demands over beliefs, values, and ideology, which are more difcult to divide and relinquish.36 Empirical research on interstate bargaining has demonstrated that limited issues are more likely to be resolved than demands over maximalist issues; in one study, Jacob Bercovitch, Theodore Anagnoson, and Donnette Willie showed that in the latter half of the twentieth century only one out of ten Cold War disputes resulted in political compromise, compared to thirteen of thirty-one nonideological disputes in which the coercing party succeeded in winning concessions.37 More recently, scholars have applied the distinction between limited and maximalist objectives to civil wars. Unlike traditional interstate conicts, which often end in territorial compromise, civil wars were found to defy political resolution because they are frequently fought over competing ideologies where the costs of retreating are comparatively high.38 Disaggregating the terrorist campaigns by objective type offers preliminary evidence that it inuences their success rate (see Figure 1). As in other political contexts, a terrorist group is said to have limited objectives when its demands are over territory. Specically, the group is ghting to either (1) evict a foreign military from occupying another country, or (2) win control over a piece of territory for the purpose of national self-determination. By contrast, a terrorist group has maximalist objectives when its demands are over ideology. In this scenario, the group is attacking a country to either (1) transform its political system (usually to either Marxist or Islamist), or (2) annihilate it because of its values.39 The data suggest that, for terrorist groups, limited objectives are far more likely to be conciliated than maximalist objectives. Coercion succeeded in three out of eight cases when territory was the goal, but it failed in all twentytwo cases when groups aimed to destroy a target states society or values. This

36. Kalevi J. Holsti, Resolving International Conicts: A Taxonomy of Behavior and Some Figures on Procedure, Journal of Conict Resolution, Vol. 10, No. 3 (September 1966), p. 272; Robert Jervis, Perception and Misperception in International Politics (Princeton, N.J.: Princeton University Press, 1976), p. 101; and Daniel Druckman and Benjamin J. Broome, Value Differences and Conict Resolution: Facilitation or Delinking? Journal of Conict Resolution, Vol. 32, No. 3 (September 1988), p. 491. 37. Jacob J. Bercovitch, Theodore J. Anagnoson, and Donnette L. Willie, Some Conceptual Issues and Empirical Trends in the Study of Successful Mediation in International Relations, Journal of Peace Research, Vol. 28, No. 1 (June 1991), pp. 717; Druckman and Broome, Value Differences and Conict Resolution, pp. 489510; and Jacob J. Bercovitch and Jeffrey Langley, The Nature of the Dispute and the Effectiveness of International Mediation, Journal of Conict Resolution, Vol. 37, No. 4 (Autumn 1993), pp. 670691. 38. Jacob J. Bercovitch and Karl DeRouen, Managing Ethnic Civil Wars: Assessing the Determinants of Successful Mediation, Civil Wars, Vol. 7, No. 1 (Spring 2005), p. 100. 39. There is no suggestion that groups with limited objectives lack ideological convictions, only that the object of change is territorial possession, not the target countrys ideology.

Why Terrorism Does Not Work 137

Figure 1. Terrorist Objectives Matter

result is not only consistent with previous studies on interstate and civil conict mediation; it is intuitively understandable that target countries would resist making concessions to groups believed to hold maximalist intentions. There are, however, major limitations to predicting the outcomes of terrorist campaigns based on the nature of their policy objectives. First, even when their objectives are territorial, terrorist groups do not usually achieve them. The Kurdistan Workers Party (PKK), Harakat ul-Mujahidin (HUM), Basque terrorists (ETA), and the Real Irish Republican Army (RIRA) have all failed to end what they regard as foreign occupations: the PKKs aspirations of an independent Kurdish state remain elusive; HUM has had little success establishing a Kashmiri state; the ETA has made some progress gaining civil and political rights, but not on its core demand of sovereignty; and Irish unication is not imminent. Second, in some cases terrorist objectives can be difcult to code. As an explanatory variable, objective type lacks robustness; terrorist objectives frequently do not conform to the territory-ideology organizing scheme. In this sample, 20 percent of the policy objectives are termed idiosyncratic; campaigns aiming to eliminate other militant groups (e.g., HUM) or sever relations between states (e.g., Revolutionary Nuclei) do not readily t into the territoryideology dichotomy (see Table 1). Furthermore, terrorist groups sometimes have ambiguous policy objectives. The al-Aqsa Martyrs Brigades, for example, routinely makes contradictory claims about whether its goal is to destroy Israel or merely establish a Palestinian state in the West Bank and Gaza Strip.40
40. The al-Aqsa Martyrs Brigades is not in the sample because the State Department began listing it as an FTO after 2001.

Contending with Terrorism 138

the key: target selection Target selection is a superior explanatory variable for predicting the outcomes of terrorist campaigns. The Department of State denes foreign terrorist organizations as groups that engage in premeditated, politically motivated violence perpetrated against noncombatant targets.41 Like other lists of terrorist groups, the Department of States does not distinguish between (1) groups that focus their attacks primarily on civilian targets and (2) those that mostly attack military targets, but occasionally attack civilians. By convention, any group whose strategy includes the intentional targeting of noncombatants is deemed a terrorist organization. This classication scheme may be defensible on normative grounds, but it obscures signicant differences in the coercion rates of guerrilla groups and what I call civilian-centric terrorist groups (CCTGs). Guerrilla groups, by denition, mostly attack military and diplomatic targets, such as military assets, diplomatic personnel, and police forces.42 CCTGs, on the other hand, primarily attack innocent bystanders and businesses. Conating the two types of groups contributes to the view that attacking civilians is an effective tactic for groups to attain their policy goals.43 In fact, for terrorist groups the targeting of civilians is strongly associated with policy failure. RANDs MIPT Terrorism Incident database provides statistics on the target selections of every terrorist group.44 When groups are classied by target selection, a trend emerges: guerrilla groupsthat is, groups whose attacks on military and diplomatic targets outnumber attacks on civilian targets accounted for all of the successful cases of political coercion. Conversely, CCTGs never accomplished their policy goals, even when they were limited, ambiguous, or idiosyncratic (see Figure 2). The remainder of the article develops a theory to explain why terrorist groups that target civilians systematically fail to achieve their policy objectives, even when they are not maximalist.

41. U.S. Department of State, Foreign Terrorist Organizations (FTOs). 42. The terms guerrilla warfare and insurgency are often used interchangeably to denote an asymmetric campaign by subnational actors against a conventional army. I have opted against using the term insurgency because it also denotes a separatist struggle. The term guerrilla group, by contrast, refers only to a subnational groups target selection. See Guerrilla, Encyclopedia of Terrorism (London: Sage, 2003), p. 54; and Cathal J. Nolan, The Greenwood Encyclopedia of International Relations (Westport, Conn.: Greenwood, 2002), p. 669. 43. In Papes research, for example, six of the thirteen terrorist campaigns are actually waged by guerrilla groups, which account for all of the terrorist victories in his sample. See Pape, Dying to Win, p. 40. 44. RAND, MIPT Terrorism Knowledge Base, http:/ /www.tkb.org/Home.jsp.

Why Terrorism Does Not Work 139

Figure 2. Paramountcy of Target Selection

Why Attacking Civilians Is Strategically Ineffective

Terrorism is a coercive instrument intended to communicate to target countries the costs of noncompliance with their policy demands. This notion has important implications for explaining the poor track record of terrorist groups. The following analysis develops a theory for why terrorist groupsespecially ones that primarily target civiliansdo not achieve their policy objectives. The basic contention is that civilian-centric terrorist groups fail to coerce because they miscommunicate their policy objectives. Even when a terrorist group has limited, ambiguous, or idiosyncratic objectives, target countries infer from attacks on their civilians that the group wants to destroy these countries values, society, or both. Because countries are reluctant to appease groups that are believed to harbor maximalist objectives, CCTGs are unable to win political concessions (see Figure 3).45 This model is grounded in two ways. First, it is consistent with attributional
45. Thomas C. Schelling makes a related point that coercion stands to work only when the coerced party understands the coercing partys demands. See Schelling, Arms and Inuence (New Haven, Conn.: Yale University Press, 1966), p. 3. Several studies analyze how groups use terrorism to signal their capabilities and resolve. These studies tend to ignore the question of whether terrorism effectively conveys to the target government the terrorist groups policy objectives. See Harvey E. Lapan and Todd Sandler, Terrorism and Signaling, European Journal of Political Economy, Vol. 9, No. 3 (August 1993), pp. 383397; Per Baltzer Overgaard, The Scale of Terrorist Attacks as a Signal of Resources, Journal of Conict Resolution, Vol. 38, No. 3 (September 1994), pp. 452478; and Hoffman and McCormick, Terrorism, Signaling, and Suicide Attack, pp. 243281.

Contending with Terrorism 140

Figure 3. Contingency Model of Civilian-centric Terrorist Groups

research in the social psychology literature; correspondent inference theory offers a framework to show that target countries infer that CCTGsregardless of their policy demandshave maximalist objectives. Second, correspondent inference theory is applied to three case studies: the responses of Russia to the September 1999 apartment bombings, the United States to the September 11 attacks, and Israel to Palestinian terrorism in the rst intifada. The three cases offer empirical evidence that (1) target countries infer that groups have maximalist objectives when they target civilians, and (2) the resultant belief that terrorist groups have maximalist objectives dissuades target countries from making political concessions. The two methodological approaches combine to offer an externally valid theory for why terrorist groups, when their attacks are directed against civilians, do not achieve their policy goals regardless of their nature. objectives encoded in outcomes Correspondent inference theory provides a framework for understanding why target countries infer that CCTGs have maximalist objectives, even when their policy demands suggest otherwise. Correspondent inference theory was developed in the 1960s and 1970s by the social psychologist Edward Jones to ex-

Why Terrorism Does Not Work 141

plain the cognitive process by which an observer infers the motives of an actor. The theory is derived from the foundational work of Fritz Heider, the father of attributional theory. Heider saw individuals as nave psychologists motivated by a practical concern: a need to simplify, comprehend, and predict the motives of others. Heider postulated that individuals process information by applying inferential rules that shape their response to behavior. In laboratory experiments, he found that people attribute the behavior of others to inherent characteristics of their personalityor dispositionsrather than to external or situational factors.46 Correspondent inference theory attempted to resolve a crucial question that Heider left unanswered: How does an observer infer the motives of an actor based on its behavior?47 Jones showed that observers tend to interpret an actors objective in terms of the consequence of the action.48 He offered the following simple example to illustrate the observers assumption of similarity between the effect and objective of an actor: a boy notices his mother close the door, and the room becomes less noisy; the correspondent inference is that she wanted quiet.49 The essential point is what Jones called the attribute-effect linkage, whereby the objectives of the actor are presumed to be encoded in the outcome of the behavior.50 Levels of correspondence describe the extent to which the objectives of the actor are believed to be reected in the effects of the action.51 When an action has high correspondence, the observer infers the objectives of the actor directly from the consequences of the action. With low correspondence, the observer either does not perceive the behavior as intentional or attributes it to external factors, rather than to the actors disposition.52
46. Fritz Heider, The Psychology of Interpersonal Relations (New York: Wiley, 1958), p. 79. 47. Kathleen S. Crittenden, Sociological Aspects of Attribution, Annual Review of Sociology, Vol. 9 (1983), p. 426. 48. Edward E. Jones and Daniel McGillis, Correspondence Inferences and the Attribution Cube: A Comparative Reappraisal, in John H. Harvey, William J. Ickes, and Robert F. Kidd, ed., New Directions in Attribution Research, Vol. 1 (Hillsdale, N.J.: Erlbaum, 1976), pp. 389420; and Edward E. Jones and Richard E. Nisbett, The Actor and the Observer, in Jones, David E. Kanouse, Harold H. Kelley, Richard E. Nisbett, Stuart Valins, and Bernard Weiner, eds., Attribution: Perceiving the Causes of Behavior (Morristown, N.J.: General Learning Press, 1972), p. 87. 49. See A Conversation with Edward E. Jones and Harold H. Kelley, in Harvey, Ickes, and Kidd, New Directions in Attribution Research, p. 378; and Edward E. Jones and Keith E. Davis, From Acts to Dispositions: The Attribution Process in Person Perception, in Leonard Berkowitz, ed., Advances in Experimental Social Psychology, Vol. 2 (New York: Academic Press, 1965), p. 225. 50. Jones and Davis, From Acts to Dispositions, p. 227. 51. Crittenden, Sociological Aspects of Attribution, p. 427; and Jones and Davis, From Acts to Dispositions, p. 263. 52. Jones and Davis, From Acts to Dispositions, p. 264. Social psychologists stress two important points: rst, that an observer believes an action has high correspondence does not mean the effect of the action actually reects the actors objectives (correspondent inferences can lead ob-

Contending with Terrorism 142

high correspondence of terrorism The theory posited here is that terrorist groups that target civilians are unable to coerce policy change because terrorism has extremely high correspondence.53 Countries believe that their civilian populations are attacked not because the terrorist group is protesting unfavorable external conditions such as territorial occupation or poverty. Rather, target countries infer from the shortterm consequences of terrorismthe deaths of innocent citizens, mass fear, loss of condence in the government to offer protection, economic contraction, and the inevitable erosion of civil libertiesthe objectives of the terrorist group.54 In short, target countries view the negative consequences of terrorist attacks on their societies and political systems as evidence that the terrorists want them destroyed. Target countries are understandably skeptical that making concessions will placate terrorist groups believed to be motivated by these maximalist objectives. As a consequence, CCTGs are unable to coerce target countries into entering a political compromise, even when their stated goals are not maximalist. The three case studies that follow provide preliminary evidence that terrorism is a awed method of coercion because (1) terrorism has high correspondence, and (2) inferences derived from its effects militate against political compromise. To highlight the effect of the independent variableterrorist attackson the proclivity of target states to bargain, a supporting case must conform to ve empirical criteria: (1) the coercing party is not motivated by a maximalist objective, that is, the desire to destroy the target states values or society; (2) the coercing party either uses terrorism or is suspected of doing so to further its policy objectives; (3) the target country xates on the short-term effects of the terrorist acts, rather than the coercing partys policy demands;
servers to assign mistaken objectives to the actor); and second, perceptions believed to be real have real consequences. Regardless of their accuracy, inferences of the actors objectives inuence the observers attitude toward the actor. See Jones and McGillis, Correspondence Inferences and the Attribution Cube, p. 417; and Harold H. Kelley and John L. Michela, Attribution Theory and Research, Annual Review of Psychology, Vol. 31 (New York: Academic Press, 1980), p. 494. 53. Attributional theories have been empirically tested and conrmed in dozens of experiments since Heider introduced the basic framework almost fty years ago. Relatively few applications have been made outside the laboratory, however, and fewer still have been applied to international relations. Those applied to international relations have tended to describe the cognitive forces impinging on the aggressor, specically, how situational factors can induce seemingly normal people to participate in depraved foreign policy behavior. See, for example, Christopher R. Browning, Ordinary Men: Reserve Policy Battalion 101 and the Final Solution in Poland (New York: Harper Perennial, 1992); Susan T. Fiske, Lasana T. Harris, Amy J.C. Cuddy, Why Ordinary People Torture Enemy Prisoners, Science, Vol. 306, No. 5701 (Summer 2004), pp. 14821483; and Sam Keen, Faces of the Enemy: Reections of the Hostile Imagination (New York: HarperCollins, 1991). 54. These common consequences of terrorist attacks are carefully explored in Ami Pedahzur, Suicide Terrorism (New York: Polity, 2005).

Why Terrorism Does Not Work 143

(4) the target country infers from the effects of the terrorist acts that the coercing party has maximalist objectives; and (5) the target countrys inference that the coercing party wants to destroy its society, values, or both impedes it from making political concessions.55

Russias Response to the 1999 Apartment Bombings

Russias response to the three apartment bombings that killed 229 Russian civilians in September 1999 helps illustrate why terrorist groups that target civilians are strategically ineffective. Before news of the bombings reverberated throughout Russia, there was widespread agreement among Russians that Chechen objectives were limited to establishing an independent Chechen state. During this period, most Russians favored territorial compromise. After the apartment bombings, however, large segments of Russian society xated on their short-term consequences and inferred from them that the presumed perpetrators (the Chechens) surreptitiously wanted to destroy Russia. This view that the Chechens are irredeemably committed to destroying Russia has eroded support for granting Chechen independence. The attitudinal shift after the bombings offers preliminary evidence that (1) terrorism has high correspondence, and (2) inferences of Chechen objectives resulting from the terrorist attacks militated against making concessions. Below I detail this attitudinal shift by tracing the evolution of Russian opinion on Chechnya between the two Chechen wars. the first chechen war, 199496 With the collapse of the Soviet Union in the late 1980s, several North Caucasian republics declared sovereignty. In 1991 Chechnyas rst president, Dzhokar Dudayev, took the additional step of declaring independence. Federal forces invaded Chechnya in December 1994 to reestablish control in the breakaway republic. For the next twenty months, Russian federal forces battled Chechen guerrillas in an asymmetric war based in Chechnya. During this period of guerrilla warfare, Russians recognized that Chechen objectives were
55. A potential objection to this framework is that it lacks a time dimension: terrorist groups, even if not initially motivated by maximalist intentions, may adopt them upon achieving more limited objectives. Violent means are often self-sustaining and can distort the ultimate goals of perpetrators, rendering ends and means conceptually indistinct. With this proviso in mind, the framework is not meant to imply that the publics of target countries are necessarily in a state of false consciousness. Rather, the point is simply that the short-term consequences of terrorist acts inform target countries understanding of the perpetrators objectives and that this inference undercuts the terrorists ability to win political concessions.

Contending with Terrorism 144

limited to self-determination. John Russell has noted that in the rst Chechen war, the Russian public [believed] that Chechens perceived their struggle as one of national liberation.56 Michael McFaul has similarly observed that the Russian military and the Russian people believed that the [Chechen] rationale for this war was self-defense.57 The Russian military shared the view that Chechen objectives were territorial, calling the rst Chechen war the War for the Restoration of Territorial Integrity.58 During this period, most Russians were prepared to make signicant concessions over the status of Chechnya. When the war broke out, the Russian public and even the secret police perceived it as precipitate, believing diplomatic solutions had not been exhausted.59 Boris Yeltsins position on the war did not gain popularity as it unfolded. Top military commanders openly resigned and condemned the president for not pursuing negotiations.60 From the onset of military operations until the cease-re in August 1996, some 70 percent of Russians opposed the war.61 Disdain for the war manifested itself most clearly in public attitudes toward Defense Minister Pavel Grachev. Opinion polls rated his approval at only 3 percent, just a few points lower than the Russian publics support for Yeltsins handling of the Chechen problem in general.62 By early 1996, domestic opposition to ghting the guerrillas imperiled Yeltsins electoral prospects. The Economist predicted in February that Mr. Yeltsin can scarcely afford to let the conict drift violently on if he hopes to win a second term of ofce in Junes presidential election.63 Yeltsin folded to domestic pressure, calling for an end to all troop operations in Chechnya and the immediate commencement of negotiations with Dudayev over its future status. Yeltsins approval rating climbed from 21 percent in February 1996 to 52 percent three months later.64 In May Yeltsin admitted in an interview that he

56. John Russell, Mujahedeen, Maa, Madmen: Russian Perceptions of Chechens during the Wars in Chechnya, 19941996 and 19992001, Journal of Post-Communist Studies and Transition Politics, Vol. 18, No. 1 (March 2002), p. 80. 57. Michael McFaul, Russia under Putin: One Step Forward, Two Steps Back, Journal of Democracy, Vol. 11, No. 3 (July 2000), p. 21. 58. Russell, Mujahedeen, Maa, Madmen, p. 77. 59. Mike Bowker, Russia and Chechnya: The Issue of Secession, Nations and Nationalism, Vol. 10, No. 4 (Winter 2004), pp. 469, 473. 60. Liliia Federovna Shevtsova, Yeltsins Russia: Myths and Reality (Washington, D.C.: Carnegie Endowment for International Peace, 1999), p. 117. 61. John Russell, Terrorists, Bandits, Spooks, and Thieves: Russian Demonization of the Chechens before and since 9/11, Third World Quarterly, Vol. 26, No. 1 (July 2005), p. 105. 62. Shevtsova, Yeltsins Russia, p. 117. 63. Russia and Chechnya: Yeltsins Vietnam, Economist, February 10, 1996, p. 51. 64. Mid-month Polls, VCIOM, posted by the Russian Independent Institute of Social and National Problems, http:/ /www.cs.indiana.edu/~dmiguse/Russian/polls.html.

Why Terrorism Does Not Work 145

would lose the upcoming election if he did not proceed with granting Chechnya de facto sovereignty.65 The Khasavyurt agreement of August 1996 committed the Russian Federation to relinquishing Chechnya by December 2001.66 In the interim period, Chechnya would have a de facto state with free elections, a parliament, a president, and autonomy over its nances and resources.67 Noting the generous terms of the accord, the BBC remarked that the status reserved for Chechnya essentially signied its independence.68 In Khasavyurt, Yeltsin formally acknowledged that Russians preferred territorial compromise to ghting the guerrillas. the second chechen war, september 1999present On September 8, 1999, Russia experienced its rst taste of modern-day international terrorism.69 A large bomb was detonated on the ground oor of an apartment building in southeast Moscow, killing 94 civilians. On September 13, another large bomb blew up an apartment building on the Kashmirskoye highway, killing 118 civilians. On September 16, a truck bomb exploded outside a nine-story apartment complex in the southern Russian city of Volgodonsk, killing 17 civilians. The Kremlin quickly ngered the Chechens as the perpetrators.70 Russians responded to the terrorist attacks by xating on their effects, while ignoring the Chechens persistent policy demands. Russia watchers noted during the bombing campaign that attention is being directed more at the actual perpetration of terrorist acts, their visible, external and horric effects with

65. Hope for Chechnya, Economist, June 1, 1996, p. 16. 66. Mark A. Smith, The Second Chechen War: The All-Russian Context, Occasional Paper, No. 40 (London: Strategic and Combat Studies Institute, September 2000), p. 5. 67. Chechnyas Truce, Maclean Hunter, June 10, 1996. 68. Kremlin Deputy Chief of Staff Elaborates on Putins Political Reform Proposals, BBC Monitor, October 4, 2004, http:/ /www.russiaprole.org/cdi/article.wbp?article-id 5FEB1101-585F4CDB-841B-7D467FF7F83B. 69. Russell, Mujahedeen, Maa, Madmen, p. 77. The September 1999 apartment bombings were not the rst terrorist attacks against Russian civilians, but there is general agreement that the bombings represented a watershed. Henceforth, Russian civilians would become a primary target. See Mia Bloom, Dying to Kill: The Allure of Suicide Terror (New York: Columbia University Press, 2005), p. 127. 70. Since the bombings, a conspiracy theory has continued to circulate that the Federal Security Services framed the Chechens to build support for a counteroffensive. Less than 10 percent of Russians accept this theory as credible. Even though the identity of the perpetrators has never been denitively established, the essential point is that the Russian public believed that the Chechens were responsible for the terrorist attacks. See A. Petrova, The Bombing of Houses in Russian Towns: One Year Later, Public Opinion Foundation, poll conducted September 14, 2000, http:/ / bd.english.form.ru/report/cat/az/A/explosion_house/eof003701.

Contending with Terrorism 146

perhaps not so much emphasis on the [stated] cause.71 Indeed, polls showed that following the terrorist acts only 15 percent of Russians believed the Chechens were ghting for independence.72 The Public Opinion Foundation reported that this motive is mentioned less frequently now. . . . [It] is gradually dying out.73 As Timothy Thomas observed after the bombings, The Chechens were no longer regarded as a small separatist people struggling to defend their territory.74 Yeltsins successor, Vladimir Putin, contributed to the view that the Chechens ceased to be motivated by the desire for national selfdetermination, declaring that the question of Chechnyas dependence on, or independence from, Russia is of absolutely no fundamental importance anymore. Following the terrorist attacks, he stopped referring to Chechens altogether, instead labeling them as terrorists. The campaign for the restoration of territorial integrity became the campaign against terrorism. Russian counterattacks became counter-terrorist operations.75 Putins focus on the effects of Chechen violence was most evident after September 11, 2001, when he told Western countries that we have a common foe, given that the World Trade Center attacks and the apartment bombings appear to bear the same signature.76 After the bombings, Russians concluded that Chechen objectives had suddenly become maximalist. Polls conducted after the terrorist attacks showed that Russians were almost twice as likely to believe that Chechen motives were now to kill Russians, bring Russia to its knees, destabilize the situation in

71. Quoted in Charles W. Blandy, Military Aspects of the Two Russo-Chechen Conicts in Recent Times, Central Asian Survey, Vol. 22, No. 4 (December 2003), p. 422. 72. Russell, Mujahedeen, Maa, Madmen, p. 88. 73. Public Opinion Foundation, Chechen Labyrinth, June 6, 2002, http:/ /bd.english.fom.ru/ report/cat/az/A/chechenian/ed022239. 74. Timothy Thomas, Manipulating the Mass Consciousness: Russian and Chechen Information War Tactics in the Second Chechen-Russian Conict, in A.C. Aldis, ed., The Second Chechen War (Washington, D.C.: Conict Studies Research Center, 2000), p. 116. 75. Quoted in Roman Khalilov, The Russian-Chechen Conict, Central Asia Survey, Vol. 21, No. 4 (January/February 2002), p. 411. 76. Quoted in John OLoughlin, Gearoid O. Tuathail, and Vladimir Kolossov, A Risky Western Turn? Putins 9/11 Script and Ordinary Russians, Europe-Asia Studies, Vol. 56, No. 1 (January 2004), p. 9. There is a tendency to dismiss Putins comparison of September 11 to the apartment bombings as instrumental rhetoric designed to articially shore up Russian relations with the West. This assertion, though true, misses an important point: even Russians who opposed stronger relations with the West after September 11, 2001, believed the attacks were analogous to the apartment bombings, despite their belief that the Chechens were the perpetrators. See John OLoughlin, Gearoid O. Tuathail, and Vladimir Kolossov, Russian Geopolitical Storylines and Public Opinion in the Wake of 9-11, n.d., p. 13, http:/ /www.colorado.edu/IBS/PEC/johno/pub/ Russianstorylines.pdf.

Why Terrorism Does Not Work 147

Russia, destroy and frighten Russian society, and bring chaos to Russian society than to achieve the independence of Chechnya.77 Putins public statements suggest that he too inferred Chechen objectives from the effects of the terrorism, asserting that the presumed perpetrators are attacking Russia so it goes up in ames.78 This post-bombing belief that Chechen objectives had become maximalist was accompanied by an abrupt loss of interest in making concessions and unprecedented support for waging war. The Public Opinion Foundation found that a strong majority of Russians (71 percent) supported the idea of trading land for peace but had come to believe that the Chechens are not trustworthy.79 When Russians were asked to explain why they no longer trusted the Chechens to abide by a land-for-peace deal, the most common explanation given was because of the terrorist acts.80 Whereas Russians had demanded Yeltsins impeachment over the rst Chechen conict, after the apartment bombings they were baying for blood.81 In the rst Chechen war, Russians favored, by a two to one margin, an independent Chechen state over battling the guerrillas in the breakaway republic; after the bombings, these numbers were reversed, even when respondents were told that federal forces would suffer heavy losses.82 Popular support for war remained remarkably stable after the bombings; six months after they occurred, 73 percent of Russians favored the advance of federal forces into Chechnya, compared with only 19 percent of Russians who wanted peaceful negotiations with the Chechen leadership.83 Since 2000, support for President Putins Chechnya policy has not dropped below 67 percent.84 the apartment bombings high correspondence In the mid-1990s, foreign jihadists began using Chechen territory as a safe haven, but links between the two groups have been exaggerated. Russia scholars
77. The Bombing of Houses in Russian Towns, Public Opinion Foundation, poll conducted September 14, 2000, http:/ /bd.english.fom.ru/report/cat/az/A/explosion_house/eof003701. 78. Quoted in Yana Dlugy, Putin Calls on Religious Leaders to Aid in Anti-terror Fight, Agence France-Presse, September 29, 2004. 79. S. Klimova, Negotiating with Maskhadov, Public Opinion Foundation, poll conducted November 21, 2002, http:/ /bd.english.fom.ru/report/cat/societas/chechnya/peace_conference/ ed024629. 80. S. Klimova, Attitude to Chechens: Pity and Fear, Public Opinion Foundation, January 30, 2003, http:/ /bd.english.form.ru/report/cat/az/A/Chechenian/ed030429. 81. Graeme P. Herd, Information Warfare and the Second Chechen Campaign, in Sally Cummings, ed., War and Peace in Post-Soviet Eastern Europe (Washington, D.C.: Conict Studies Research Center, 2000), p. 32. 82. ChechnyaTrends, 20002005, Levada Center (formerly VCIOM), poll conducted September 30, 2000, http:/ /www.russiavotes.org/Mood_int_tre.htm. 83. Smith, The Second Chechen War, p. 6. 84. OLoughlin, Tuathail, and Kolossov, Russian Geopolitical Storylines, p. 22.

Contending with Terrorism 148

widely agree that since the Soviet Union unraveled in the late 1980s, the Chechens objective has remained constantto establish an independent Chechen state.85 Russian perceptions of Chechen aims changed profoundly, however, as a result of the apartment bombings. There is supporting evidence that the barrage of attacks on Russian civilians in September 1999 had high correspondence. Russians xated on the short-term consequences of the bombings and suddenly concluded that the suspected attackers evidently want Russia destroyed. Once Russians believed that the Chechens goal was no longer conned to achieving national self-determination, enthusiasm for compromise abruptly declined while support for a military solution increased.

U.S. Response to the September 11 Terrorist Attacks

The response of the United States to the September 11 attacks further illustrates why terrorist groups that target civilians are unable to coerce policy change. The U.S. response provides supporting evidence that (1) terrorism has high correspondence, and (2) inferences derived from the effects of the attacks have not been conducive to offering concessions. The following case study shows that Americansespecially in the immediate aftermath of the terrorist attackshave tended to ignore al-Qaidas rationale for violence. Instead of focusing on al-Qaidas policy demands, they have xated on the effects of the terrorist attacks and inferred from them that the terrorists are targeting the United States to destroy its society and values. These inferences have hampered al-Qaida from translating its violence into policy successes in the Muslim world. al-qaidas stated objectives Al-Qaida describes its use of terrorism as a communication strategy to demonstrate to the United States the costs of maintaining its unpopular foreign policies in the Muslim world.86 Osama bin Laden has implored Americans to rid
85. See Gail W. Lapidus, Putins War on Terrorism: Lessons from Chechnya, Post-Soviet Affairs, Vol. 18, No. 3 (January/March 2002), p. 45; Roman Khalilov, The Russian-Chechen Conict, Central Asia Survey, Vol. 21, No. 4 (JanuaryFebruary 2002), p. 411; Richard Pipes, Give the Chechens a Homeland of Their Own, New York Times, September 9, 2004; and Pedahzur, Suicide Terrorism, p. 113. 86. In his May 1997 interview with CNN, bin Laden described terrorism as a message with no words. Four years later he characterized the coordinated strikes on Washington and New York as speeches that overshadowed all other speeches. In his treatise for waging jihad, bin Ladens ideological counterpart, Ayman al-Zawahiri, asserted that terrorism is the only language understood by the West. An al-Qaida spokesman referred to the 2002 attack on a French

Why Terrorism Does Not Work 149

themselves of their spiritless materialistic life, but a comprehensive perusal of al-Qaidas public statements reveals scant references to American popular culture. Bin Laden has threatened that freedom and human rights in America are doomed, but American political values are also not a recurrent theme in al-Qaida communiqus. The relative silence on these issues suggests that American values are not a principal grievance.87 In fact, bin Laden has explicitly rejected the claim that al-Qaidas goal is to change these values. On multiple occasions, he has warned American audiences that those who repeat this lie either suffer from confusion or are intentionally misleading you.88 Since bin Laden declared war on the United States in February 1998, his policy demands have remained notably consistent.89 First, his most well

tanker off the coast of Yemen as a political message to Washington. See CNN, Interview with Osama bin Laden, May 10, 1997, http:/ /fl1.findlaw.com/news.findlaw.com/cnn/docs/ binladen/binladenintvw-cnn.pdf; videotape from Afghanistan made in November 2001, quoted in Brigitte L. Nacos, The Terrorist Calculus behind 9/11: A Model for Future Terrorism? Studies in Conict and Terrorism, Vol. 26, No. 8 (January/February 2003), p. 1; and Ayman al-Zawahiri, Knights under the Prophets BannerMeditations on the Jihadist Movement (London: Al-Sharq alAwsat, 2001), quoted in Middle East/Afghanistan: Al Qaida Threat, Oxford Analytica, May 14, 2002, http:/ /www.ciaonet.org/pbei/oxan/oxa05142002.html. For an almost identical statement, see Al-Qaida Urges More Attacks, al-Zawahiri Tape, Aljazeerah.com, May 21, 2003. 87. Osama bin Ladens two-minute audiotape broadcast of October 6, 2002, on Aljazeera, quoted in Rohan Gunaratna, Defeating Al QaedaThe Pioneering Vanguard of the Islamic Movements, in Russell D. Howard and Reid L. Sawyer, eds., Defeating Terrorism: Shaping the New Security Environment (New York: McGraw-Hill, 2003), p. 26. See also Clark McCauley, Psychological Issues in Understanding Terrorism and the Response to Terrorism, in Chris E. Stout, ed., Psychology of Terrorism: Coping with the Continuing Threat (Westport, Conn.: Praeger, 2004), p. 51. Several al-Qaida experts have remarked that al-Qaida leaders almost always emphasize their hatred of U.S. foreign policies, not American values. See, for example, Peter L. Bergen, Holy War, Inc.: Inside the Secret World of Osama Bin Laden (New York: Free Press, 2001), p. 223; and Rohan Gunaratna, Inside Al Qaeda: Global Network of Terror (New York: Columbia University Press, 2002), p. 45. For a detailed analysis of al-Qaidas objectives, see Max Abrahms, Al-Qaedas Scorecard: A Progress Report on al-Qaedas Objectives, Studies in Conict and Terrorism, Vol. 29, No. 4 (JulyAugust, 2006), pp. 509 529. 88. Bin Laden: Your Security Is in Your Own Hands, October 30, 2004, http:/ /edition.cnn.com/ 2004/WORLD/meast/10/29/bin.laden.transcript. 89. Most terrorism experts believe that bin Ladens policy demands have stayed consistent since the late 1990s. Even those who place less emphasis on his policy demands do not dispute that the jihadists are committed to realizing them. The debate is over whether granting bin Ladens demands would alleviate the terrorism threat or merely transfer the violence to a new theater or cause. This article does not take a position on this important question. There is no suggestion here that the global jihadists would lay down their arms upon accomplishing the objectives laid out in bin Ladens 1998 fatwa. For a sample of works on this debate, see Anonymous (Michael Scheuer), Imperial Hubris: Why the West Is Losing the War on Terror (Washington, D.C.: Brasseys, 2004); Anonymous (Michael Scheuer), Through Our Enemies Eyes: Osama bin Laden, Radical Islam, and the Future of America (Washington, D.C.: Brasseys, 2003); Daniel L. Byman, Al-Qaeda as an Adversary: Do We Understand Our Enemy? World Politics, Vol. 56, No. 1 (October 2003), pp. 139163; Bergen, Holy War, Inc.; Jessica Stern, Errors in Fighting Al Qaeda Have Worsened the Danger, Boston

Contending with Terrorism 150

known ultimatum is for the United States to withdraw its troops from Saudi Arabia, Land of the Two Holy Places. His statements indicate that he objects not only to the U.S. stationing of troops in the holiest of places, but also to U.S. bases serving as a spearhead through which to ght the neighboring Muslim peoples. In al-Qaida communiqus, criticisms of U.S. military interference in Saudi Arabia have invariably been coupled with complaints about the treatment of its neighbors, especially Iraq. For the al-Qaida leadership, deploying U.S. troops to Saudi Arabia during the lead-up to the 1991 Persian Gulf War was not only an egregious provocation in itself; the bases represented and facilitated the occupation of its most powerful neighboring Arab state. Bin Laden and his lieutenants have thus threatened that the United States will remain a target until its military forces withdraw from the entire Persian Gulf.90 Second, al-Qaida spokesmen say that its terrorist acts are intended to dissuade the United States from supporting military interventions that kill Muslims around the world. In the 1990s these interventions included Crusader wars in Chechnya, Bosnia, and East Timor. Bloodshed in Israel and Iraq during this period generated the most intense opposition. Since the September 11 attacks, al-Qaidas condemnation of the United States has focused on events in these two countries.91 Third, al-Qaida communiqus emphasize the goal of ending U.S. support for pro-Western Muslim rulers who suppress the will of their people. Al-Qaida leaders routinely denounce the House of Saud and President Pervez Musharrafs Pakistan in particular as the most oppressive, corrupt, and

Globe, March 25, 2004; Assaf Moghadam, The New Martyrs Go Global, Boston Globe, November 18, 2005; and Steve Simon and Daniel Benjamin, The Terror, Survival, Vol. 43, No. 4 (Winter 2001), pp. 517. 90. World Islamic Front, Jihad against Jews and Crusaders, February 22, 1998, http:/ /www .fas.org/irp/world/para/docs/980223-fatwa.htm; Osama bin Laden, Declaration of War against the Americans Occupying the Land of the Two Holy Places, August 1996, http:/ /www.pbs.org/ newshour/terrorism/international/fatwa_1996.html; Middle East Media Research Institute, Osama bin Laden Speech Offers Peace Treaty with Europe, Says Al-Qaida Will Persist in Fighting the U.S., Special Dispatch, No. 695, posted April 15, 2004, http:/ /memri.org/bin/ articles.cgi?Page archives&Area sd&ID \SP69504. Interview rst broadcast on Aljazeera, April 15, 2004; and Al-Qaeda Urges More Attacks, al-Zawahiri Tape, Agence France-Presse, May 21, 2003. 91. See, for example, Bin Laden Rails against Crusaders and UN, November 3, 2002, BBC.com, http:/ /news.bbc.co.uk/1/hi/world/monitoring/media_reports/1636782.stm. Statement rst broadcast on Aljazeera.com, November 3, 2002. See also Bin Ladens Warning: Full Text, October 7, 2001, BBC.com, http:/ /news.bbc.co.uk/1/hi/world/south_asia/1585636.stm. Statement rst broadcast on Aljazeera.com, July 7, 2001.

Why Terrorism Does Not Work 151

tyrannical regimes whose very existence depends on the supervision of America.92 A prominent al-Qaida website has equated U.S. nancial and political support of Saudi Arabia and Pakistan to colonization.93 Fourth, al-Qaida leaders describe Israel in similar terms, as a colonial outpost. Based on the organizations communiqus, al-Qaidas nal objective is thus to destroy the Zionist-Crusader alliance, which enables Israel to maintain its occupation of Jerusalem and murder Muslims there.94 effects trump rationale Americans have focused on the effects of al-Qaida violence, not on al-Qaidas stated purpose. Ronald Steel noted in the New Republic after the June 1996 attack on the Khobar Towers in Saudi Arabia, which killed 19 Americans, that American journalists xated on who or what bin Laden attacked and the method of attack. By contrast, what bin Laden had been saying about why he and his al-Qaida forces were attacking was given short shrift.95 The British journalist Robert Fisk similarly observed that after the August 1998 attacks on the U.S. embassies in Kenya and Tanzania, U.S. leaders emphasized the carnage and devastation, but not in a single press statement, press conference, or interview did a U.S. leader or diplomat explain why the enemies of America hate America.96 Since September 11, 2001, major Western journalists have devoted generous coverage to the fallout of terrorist attacks, but only since 2004, with the publication of Michael Schueurs Imperial Hubris, have they consistently published excerpts of al-Qaidas communiqus.97

92. ABC News, Interview: Osama Bin Laden, May 1998, http:/ /www.pbs.org/wgbh/pages/ frontline/shows/binladen/who/interview.html; and Ayman Zawahiri, quoted in Bergen, Holy War, Inc., p. 208. See also Observer, Full Text: Bin Ladens Letter to America, November 24, 2002, http:/ /observer.guardian.co.uk/worldview/story/0,11581,845725,00.html. The letter rst appeared on the internet and was then circulated by supporters in Britain. 93. This statement is based on conclusions reported by the Center for Islamic Studies and Research in The Operation of 11 Rabi al-Awwal: The East Riyadh Operation and Our War with the United States and Its Agents, published in 2003. The book was translated by the Foreign Broadcast Information Service. For excerpts, see http:/ /www.why-war.com/les/2004/01/qaeda_east_ riyadh.html. 94. Middle East Media Research Institute, Osama bin Laden Speech Offers Peace Treaty with Europe; and World Islamic Front, Jihad against the Jews and the Crusaders, February 23, 1998, http:/ /www.fas.org/irp/world/para/ladin.htm. 95. Ronald Steel, Blowback: Terrorism and the U.S. Role in the Middle East, New Republic, July 28, 1996, pp. 711. 96. Robert Fisk, As My Grocer Said: Thank You Mr. Clinton for the Fine Words . . . , Independent (London), August 22, 1998, p. 3. 97. See Anonymous, Imperial Hubris, p. 128.

Contending with Terrorism 152

high correspondence of september 11 President George W. Bushs public pronouncements indicate that he deduces al-Qaidas motives directly from the short-term consequences of the terrorist attacks of September 11. According to Bush, We have seen the true nature of these terrorists in the nature of their attacks, rather than in their professed political agenda.98 For Bush, September 11 demonstrated that the enemy hates not our policies, but our existence.99 In the resulting panic weeks after the attacks, he concluded, These acts of mass murder were intended to frighten our nation into chaos. With Americans hesitant to y after the four planes were hijacked, he asserted, They [the terrorists] want us to stop ying.100 The toppling of the World Trade Center and the economic contraction that followed revealed that the terrorists wanted our economy to stop. With American civil liberties inevitably restricted in the wake of the attacks, he proclaimed that al-Qaidas goals, inter alia, were to curtail our freedom of religion, our freedom of speech, our freedom to vote and assemble and disagree with each other.101 Given that al-Qaida and its afliates are mute on these topics, it is difcult to imagine Bush ascribing them to the terrorists had Americans not been greatly frightened for their safety, hesitant to y, and worried about their political and economic future in the wake of the terrorist attacks. For President Bush, any group that deliberately attacks American civilians is evidently motivated by the desire to destroy American society and its democratic values. When asked by a reporter in October 2001 if there was any direct connection between the September 11 attacks and the spate of anthrax attacks that followed, he replied: I have no direct evidence but there are some links . . . both series of actions are motivated to disrupt Americans way of life.102 This interpretation of the motives of the unknown terrorist perpetrator(s) is revealing: the identity of the person(s) who sent the anthrax is irrelevant because

98. George W. Bush, speech by the president to the Warsaw Conference, Warsaw, Poland, November 6, 2001. 99. George W. Bush, remarks by the president to the United Nations General Assembly, New York, November 10, 2001, http:/ /www.usunnewyork.usmission.gov/01_162.htm. 100. Richard A. Clarke, Against All Enemies: Inside Americas War on Terror (New York: Free Press, 2004), p. 17; and George W. Bush, presidential speech to the California Business Association, Sacramento, November 17, 2001. 101. Quoted in Clarke, Against All Enemies, p. 17; and George W. Bush, address by the president to a joint session of Congress, Washington, D.C., September 23, 2001. 102. George W. Bush, remarks by the president to the Dixie Printing Company, Glen Burnie, Maryland, October 24, 2001, http:/ /www.globalsecurity.org/military/library/news/2001/10/ mil-011024-usa03c.htm.

Why Terrorism Does Not Work 153

all terrorists who disrupt the American way of life must be motivated by this maximalist objective.103 The American public has tended to share President Bushs interpretation of the terrorists motives. Polls conducted after September 11 show that most Americans believed that al-Qaida was not responding to unpopular U.S. foreign policies. After the attacks, only one in ve respondents agreed with the statement that there is any way that the United States has been unfair in its dealings with other countries that might have motivated the terrorist attacks.104 In a separate poll, only 15 percent of Americans agreed that American foreign policies are partly the reason for al-Qaida terrorism.105 Instead of attributing al-Qaida terrorism to U.S. foreign policies, large segments of American society shared Bushs belief that the goal of the terrorists was to destroy American society and values. Since September 11, more Americans have polled that the terrorists are targeting the United States because of its democracy, freedom, values, and way of life than because of its interference in the Muslim world.106 al-qaidas miscommunication strategy Bin Laden and his lieutenants frequently complain that the United States has failed to understand the true reason for the September 11 attacks. Instead of attacking because we hate freedom, the attacks are a response to the fact that you spoil our security and attack us.107 Attributional research provides a framework to explain why al-Qaidas communication strategy has failed. As correspondent inference theory predicts, supporting evidence suggests that President Bush and large segments of American society focused on the disastrous effects of al-Qaidas behavior and inferred from them that the terrorists must want to destroy American society and its valuesdespite al-Qaidas relative silence on these issues.108
103. President Bush is not the only U.S. president to infer these objectives from terrorist attacks. The claim that al-Qaida struck the United States to destroy its society and values was also espoused by President Bill Clinton after the terrorist bombing of the World Trade Center in 1993. See Clarke, Against All Enemies, pp. 129130. 104. Pew Research Center, Roper Center, September 21, 2001. Seventy percent of Americans rejected the idea that unfair U.S. foreign policies contributed to the terrorist attacks. 105. IPSOS-REID, Roper Center, September 21, 2001. 106. Roper Center, Harris poll, September 1924, 2001, http:/ /www.pollingreport.com/ terror9.htm, October 20, 2005. 107. Quoted in Anonymous, Imperial Hubris, p. 153; and Aljazeera.com, Bin Laden: Your Security Is in Your Own Hands. 108. Ronald Spiers, Try Clearer Thinking About Terrorists, International Herald Tribune, January

Contending with Terrorism 154

To be sure, even if terrorism had not delegitimized al-Qaidas policy demands, it is inconceivable the United States would have ever fully complied with them. Paul Wilkinson has observed that in deciding whether to negotiate with terrorists, the target government must rst decide whether their demands are corrigible or incorrigible. When demands are perceived as corrigible, the target government engages in a roots debatean assessment of the pros and cons of appeasing the terrorists. When terrorists are perceived as incorrigible, concessions are rejected outright because the demands are deemed so extreme that they fall outside of the realm of consideration. In Wilkinsons model, incorrigible terrorists are not categorically implacable, but placating them would exact a prohibitive cost.109 In the discourse of international relations theory, realists would support the view that the United States has not entered a postSeptember 11 roots debate because it is strategically wedded to the Middle East. Realists are on strong ground in their prediction that the worlds most powerful country would not willingly concede a geographically vital region of the world to terrorists. But it is doubtful that had Americans viewed al-Qaidas stated grievances as credible, they would have embraced a counterterrorism strategy after September 11 that systematically aggravated them. In response to the September 11 attacks, the United States took four steps: (1) increased troop levels in the Persian Gulf fteenfold; (2) strengthened military relations with pro-U.S. Muslim rulers, especially in Pakistan and Saudi Arabia; (3) supported counterterrorism operationseither directly or indirectlythat have killed tens of thousands of Muslims around the world; and (4) became an even less partial mediator in the Israeli-Palestinian conict.110 Although the
14, 2003; Bergen, Holy War, Inc., p. 223; and Gunaratna, Inside Al Qaeda, p. 45. See also Anonymous, Imperial Hubris, p. x. 109. Paul Wilkinson, Security Challenges in the New Reality, lecture, Tufts University, Medford, Massachusetts, October 16, 2002. 110. For a detailed analysis of al-Qaidas effect on U.S. policies in the Muslim world, see Abrahms, Al-Qaedas Scorecard. In this study, al-Qaida is tagged with failures in three of the core policy objectives outlined in its 1998 declaration of war: ending U.S. support for Muslim apostate regimes, Israel, and what it derides as Crusader wars, such as Operation Iraqi Freedom. AlQaidas policy effectiveness in the Persian Gulf is designated as a limited success. Overall, the September 11 attacks did not reduce U.S. involvement in the Gulf. On the contrary, the attacks served as the critical impetus for the American publics decision to support the operation, which has led to the long-term occupation of Iraq and unprecedented U.S. military cooperation with the Gulf monarchy countries. The one modest success was the U.S. decision to draw down its troop presence in the Saudi Arabian Peninsula after September 11, 2001. Al-Qaida does not regard this policy outcome as noteworthy, for two reasons. First, the decision to withdraw hundreds of American troops from the Saudi desert after September 11 palls in comparison to the roughly 150,000 additional U.S. troops that were deployed to the same theater during this period. Second, U.S. in-

Why Terrorism Does Not Work 155

September 11 attacks achieved al-Qaidas intermediate objectives of gaining supporters and attention, its postSeptember 11 policy failures are a testament, at least in part, to its awed communication strategy.

Israels Response to the First Intifada

The rst intifada may seem like an unlikely case study to illustrate the limitations of terrorism as a coercive strategy. The mass uprising in the Gaza Strip and the West Bank was an exceptionally moderate period in the history of Palestinian terrorism. The revolt from December 1987 to January 1991 killed only twenty Israeli civilians. Compared with the Revolutionary Violence campaign of the 1970s and the outbreak of the second intifada in September 2000, the rst intifada was a peaceful interlude.111 Furthermore, the spontaneous insurrection was a bottom-up initiative. It circumvented Palestinian terrorist groups, which were ideologically opposed to a two-state solution. These groups were momentarily sidelined for three reasons. First, the Marxist groups (e.g., the Popular Front for the Liberation of Palestine and the PFLPGeneral Command) were reeling from the recent loss of their Soviet patron with the end of the Cold War. Second, the Islamist groups (e.g., Hamas and Islamic Jihad) did not yet pose a signicant challenge to the Palestine Liberation Organization (PLO). Third, the PLO was based in Tunis during this period, largely detached from Palestinian life in the territories.112 Facing relatively little competition from other Palestinian groups, the PLO co-opted the mass uprising in the late 1980s by recognizing the Israeli state within its pre-1967 borders and formally renouncing terrorism. Despite the unusually moderate tactics and objectives of the intifada, the Israeli response to it underscores that (1) the limited use of Palestinian terrorism had high correspondence, and (2) Israeli inferences of Palestinian objectives undermined support for making concessions. Edy Kaufman has noted that the primary purpose of the rst intifada was
terference in the political affairs of the Saudi kingdom increased markedly after the September 11 attacks, owing to the joint U.S.-Saudi interest in ghting the jihadists. I thank Michael Knights of Janes Intelligence Review for offering his expertise in this area. For supporting analysis, see Chaim Kaufmann, Threat Ination and the Failure of the Marketplace of Ideas: The Selling of the Iraq War, International Security, Vol. 29, No. 1 (Summer 2004), p. 31; and Pape, Dying to Win, pp. 46, 84. 111. BTselem, Fatalities in the First Intifada, http:/ /www.btselem.org/english/statistics/rst_ Intifada_Tables.asp. 112. See Bloom, Dying to Kill, p. 24; and Council on Foreign Relations, Terrorism: Questions and Answers, October 31, 2005, http:/ /cfrterrorism.org/groups/pp.html.

Contending with Terrorism 156

to communicate to Israelis the need to end the occupation of the territories.113 Terrorist acts, even in small numbers, interfered with the message. Throughout the intifada, only 15 percent of Palestinian demonstrations were violent.114 Yet an absolute majority of Israelis (80 percent) believed that the means employed by the Palestinians to protest Israeli rule were mainly violent. Of the violent Palestinian acts, the vast majority consisted of rock throwing against the Israel Defense Forces in the territories, with few incidences of terrorism inside the Green Line. An even broader consensus of Israelis (93 percent) felt that the intifada was directed both towards civilians and towards the army.115 Notwithstanding the intifadas restrained use of violence, Israelis appear to have xated on the intermittent attacks against Israeli civilians. The Louis Guttman Israel Institute of Applied Social Research conducted a series of polls in December 1990 to assess the Israeli publics views of Palestinian objectives in the rst intifada. As correspondent inference theory predicts, a strong majority of the respondents surveyed (85 percent) believed its purpose was to cause damage and injuryas it surely didwhile only a fraction (15 percent) believed the goal was to express protest. Similarly, the majority (66 percent) believed that the intifada was directed against the existence of the state of Israel, while a minority (34 percent) believed the purpose was to liberate the West Bank and Gaza Strip.116 The disconnect between the PLOs policy demands and Israeli perceptions of Palestinian objectives has been explained by (1) inconsistent rhetoric on the part of Palestinian leaders about the aims of the intifada, and (2) Jewish apprehension that contemporary violence against Israel is akin to previous traumatic experiences in which Jewish survival in the Diaspora was threatened. Compelling evidence suggests, however, that terrorism informed the Israeli view of Palestinian objectives. In a fascinating study based on the polling data contained in the Guttman report, Kaufman observed that the respondents who perceived Palestinian tactics as mainly violent were more likely to believe that the Palestinian goal was to destroy Israel. Conversely, the more Israelis perceived Palestinian tactics as nonviolent, the more they believed the goal

113. Edy Kaufman, Israeli Perceptions of the Palestinians Limited Violence in the Intifada, Terrorism and Political Violence, Vol. 3, No. 4 (Winter 1991), p. 4. 114. Gene Sharp, The Intifada and Nonviolent Struggle, Journal of Palestine Studies, Vol. 19, No. 1 (Autumn 1989), p. 7. 115. Kaufman, Israeli Perceptions of the Palestinians Limited Violence in the Intifada, p. 4. 116. Louis Guttman Israel Institute of Applied Social Research, Jerusalem, Public Assessment of the Activities and Violence of the Intifada, No. (s)IK1124/E&H (December 1990).

Why Terrorism Does Not Work 157

was to liberate the territories. The positive correlation between perceived Palestinian terrorism and maximalist objectives existed independent of the respondents political afliation, suggesting that the association was not a function of their preexisting political attitudes.117 Not surprisingly, Israelis were twice as likely to believe less in the idea of peace than before the intifada.118 Because the majority of Israelis regarded the intifada as a protracted terrorist campaign, and Israelis inferred from Palestinian terrorism their intentions of wanting to destroy Israel, the intifada undermined Israeli condence in the Palestinians as a credible partner for peace. In the early 1990s, Israeli Prime Ministers Yitzhak Shamir and Yitzhak Rabin came under increased pressure to trade land for peace with the Palestinians. The sources of pressure were twofold. First, President George H.W. Bush, determined to improve U.S.-Arab relations after Israel had lost its strategic utility as a Cold War satellite, forced the Israelis to the negotiating table by linking U.S. nancial assistance to Shamirs participation in the Madrid peace conference in October 1991. Second, Israeli military strategists recognized that the Jewish state faced a long-term demographic problem in occupying a growing and restive Palestinian population.119 In September 1993 Israel consented to the land-for-peace formula outlined in the Declaration of Principles known as the Oslo accords, but the pattern persisted: although Palestinian terrorism demonstrated to Israel the costs of the occupation, it undercut Israeli condence in the Palestinians as a credible partner for peace, reducing support for making territorial concessions.120 Throughout the 1990s, the Jaffee Center
117. Kaufman, Israeli Perceptions of the Palestinians Limited Violence in the Intifada, p. 13. 118. Jerusalem Post International Edition, No. 1 (weekend ed., August 27, 1988), p. 451. 119. Avi Shlaim, When Bush Comes to Shove: America and the Arab-Israeli Peace Process, Oxford International Review, Vol. 3, No. 2 (Spring 1992), p. 4. 120. The limited objectives of the rst intifada should not be confused with the maximalist objectives of the Palestinian terrorist organizations. The six Palestinian terrorist organizations in this study have been largely ineffective in accomplishing their policy objectives. Hamas is typically regarded as the greatest beneciary of Palestinian terrorism. Even for Hamas, however, the gulf between its policy demands and their outcome is vast. Its Covenant of the Islamic Resistance Movement stresses two goals. First, Israel or any part of it should not be squandered; it or any part of it should not be given up. According to Hamas, all of historic Palestine is an Islamic waqf, which translates as a prohibition from surrendering or sharing. Hamass only territorial acquisition is in the Gaza Strip, which represents less than 2 percent of the total land area Hamas claims as its own. Even if one assumes that Hamas may ultimately accept a Palestinian state in the West Bank and Gaza Strip, its territorial achievements to date have been minor. The Gaza Striponenineteenth the size of the West Bankrepresents less than 6 percent of the disputed land outside the Green Line. Even the most moderate Palestinian nationalist would be categorically opposed to establishing a Palestinian state in such a small territory. Furthermore, according to international law Gaza remains under occupation because Israel still controls the airspace and Palestinian movement on land and by sea. Palestinians share the view that Gaza remains under occupation, commonly referring to it as the prison. The second major policy objective is to destroy Israeli so-

Contending with Terrorism 158

for Strategic Studies periodically polled Israeli respondents on their perceptions of Palestinian aspirations. The dominant response was that the Palestinians wanted to conquer Israel and destroy a large portion of the Jewish population, a position that peaked during heightened levels of terrorist activity.121 The perception that the Palestinians hold maximalist aspirations has been the principal impediment to Israels willingness to make signicant territorial concessions. Since 1994 the Tami Steinmetz Center for Peace Research has polled a representative sample of Israelis on two questions: Do you believe the Palestinians are viable partners for peace? And do you support the peace process? Instances of Palestinian terrorism systematically incline Israelis to answer no to both questions.122 In sum, since the rst intifada, Palestinian violence has created pressure on Israel to change the status quo. Paradoxically, terrorism has simultaneously convinced Israelis that the Palestinians are not committed to a two-state solution, which has eroded support for making the territorial concessions necessary to achieving it.

Conclusion
Thomas Schelling asserted more than a decade ago that terrorists frequently accomplish intermediate means toward political objectives . . . but with a few exceptions it is hard to see that the attention and publicity have been of much value except as ends in themselves.123 This study corroborates that view; the twenty-eight groups of greatest signicance to U.S. counterterrorism policy have achieved their forty-two policy objectives less than 10 percent of the time. As the political mediation literature would predict, target countries did not make concessions when terrorist groups had maximalist objectives. Yet even
ciety. Despite Palestinian violence, Israels gross domestic product per capita ($18,000) is higher than those of New Zealand, Spain, Portugal, and Greece. Far from destroying Israel, Palestinians consistently rate Israel as the country to which the goal of Palestinian statehood should most aspire. See Central Intelligence Agency, Gaza Strip, The World Fact Book, http:/ /www.cia.gov/cia/ publications/factbook/index.html; Covenant of the Islamic Resistance Movement, August 18, 1988, http:/ /www.yale.edu/lawweb/avalon/mideast/hamas.htm; and Palestinian Center for Policy and Survey Research, Palestinians Expect It to Grow, June 6, 2005, http:/ /www.pcpsr .org/ survey/polls/2005/p16ejoint.html. 121. Asher Arian, Israeli Public Opinion on National Security, 2000 (Tel Aviv: Jaffee Center for Strategic Studies, 2000), p. 14. 122. Tami Steinmetz Center for Peace Research, Tel Aviv University, Peace Index, http:/ / www.tau.ac.il/peace. See also David Fielding and Madeline Penny, What Causes Changes in Opinion about the Israeli-Palestinian Peace Process? Economics Discussion Paper, No. 0601 (Dunedin, New Zealand: School of Business, University of Otago, March 2006), p. 8. 123. Schelling, What Purposes Can International Terrorism Serve? p. 20.

Why Terrorism Does Not Work 159

when groups expressed limited, ambiguous, or idiosyncratic policy objectives, they failed to win concessions by primarily attacking civilian targets. This suggests not only that terrorism is an ineffective instrument of coercion, but that its poor success rate is inherent to the tactic of terrorism itself. Why are terrorist groups unable to coerce governments when they primarily attack civilian targets? Terrorism miscommunicates groups objectives because of its extremely high correspondence. The responses of Russia to the September 1999 apartment bombings, the United States to the attacks of September 11, and Israel to Palestinian terrorism in the rst intifada provide evidence that target countries infer the objectives of terrorist groups not from their stated goals, but from the short-term consequences of terrorist acts. Target countries view the deaths of their citizens and the resulting turmoil as proof that the perpetrators want to destroy their societies, their publics, or both. Countries are therefore reluctant to make concessions when their civilians are targeted irrespective of the perpetrators policy demands. Four policy implications follow for the war on terrorism. First, terrorists will nd it extremely difcult to transform or annihilate a countrys political system. Second, the jihadists stand to gain from restricting their violence to military targets. Already, mounting U.S. casualties in Iraq and the absence of a postSeptember 11 attack on the homeland have eroded U.S. support for maintaining a military presence in Iraq.124 Terrorist strikes on the U.S. homeland will only undermine the terrorists message that their purpose is to alter unpopular U.S. policies in the Muslim world. Even sporadic attacks on American civiliansif seen as the dominant component of al-Qaidas overall strategy will undermine support for an exit strategy. Third, the self-defeating policy consequences of terrorism will ultimately dissuade potential jihadists from supporting it. Although guerrilla attacks against U.S. forces in Iraq show no signs of abating, polling data from Muslim countries suggest that the terrorism backlash is already under way. The Pew Research Center reported in its July 2005 Global Attitudes Project that compared with its polls conducted in 2002, In most majority-Muslim countries surveyed support for suicide bombings and other acts of violence in defense of Islam has declined signicantly, as has condence in Osama bin Laden to do the right thing in world affairs.125 Similarly, major Islamist groups and leaders are increasingly denouncing terrorist
124. See Bush Urges Patience as Support for War Shrinks, CNN.com, October 30, 2005. 125. Since 2002, public support for terrorism has dropped by 64 percent in Pakistan, 80 percent in Indonesia, 87 percent in Lebanon, and 200 percent in Morocco. Pew Global Attitudes Project, Support for Terror Wanes among Muslim Publics, July 14, 2005, pp. 2, 6, http:/ /www.pewglobal .org.reports/pdf/248.pdf.

Contending with Terrorism 160

attacks as counterproductive, even as they encourage guerrilla warfare against the Iraqi occupation.126 Fourth, it is commonly said that terrorists cannot be deterred because they are willing to die for their cause and that they lack a return address to threaten with a retaliatory strike.127 But perhaps the greatest reason deterrence breaks down is because of the widespread, albeit erroneous, belief that attacking civilians is an effective strategy for terrorist groups to advance their policy goals. Disabusing terrorists of this notion would go a long way toward defusing the cycles of violent reprisal. Further research is needed in three areas. First, why do terrorist groups target civilians if doing so is strategically ineffective? Testing of the following four hypotheses could yield useful results: (1) groups have an exaggerated sense of terrorisms ability to coerce policy change;128 (2) terrorist groups attach equal importance to achieving their intermediate objectives; (3) even though terrorism almost never pays, it is a superior strategy to the alternatives, such as conducting a peaceful protest; and (4) only comparatively weak groups target civilians, because attacking military targets requires a higher level of combat sophistication. Of these hypotheses, only the fourth one appears empirically dubious. Nascent terrorist groups generally focus their attacks on military targets and then graduate to attacking civilian targets. This progression from military to civilian targets was evident between the two Chechen wars, between al-Qaidas declaration of war on the United States in 1998 and the September 11 attacks, and from the beginning of the rst intifada to its more violent conclusion. In each campaign, the terrorists initially conned their attacks to military targets and then, upon becoming stronger organizationally and technologically, took aim at civilians. Second, future research may demonstrate that in international relations the attribute-effect linkage diminishes over time. In this study, the target countries inferred from attacks on their civilians that the perpetrators held maximalist objectives that could not be satised. As time elapsed from the terrorist attacks, however, the publics of Russia and the United States began expressing

126. Fareed Zakaria, How We Can Prevail, Newsweek, July 18, 2005, p. 38. 127. See, for example, George W. Bush, National Security Strategy of the United States of America (Washington, D.C.: White House, September 17, 2002), http:/ /www.whitehouse.gov/nsc/ nss.html. 128. Bin Laden, for example, has frequently said that terrorism works, especially against the United States, such as when it withdrew from Lebanon following the 1983 U.S. Marine barracks bombing and from Somalia in 1993 after the deaths of the eighteen U.S. Army Rangers. See Bruce Hoffman, Rethinking Terrorism and Counterterrorism since 9/11, Studies in Conict and Terrorism, Vol. 25, No. 5 (September 2002), p. 310.

Why Terrorism Does Not Work 161

greater receptivity to curtailing their countrys inuence in Chechnya and the Muslim world, respectively.129 Third, correspondent inference theory may have prescriptive utility for conducting a more strategic and humane war on terrorism. If countries impute terrorists motives from the consequences of their actions, then the communities in which terrorists thrive may impute states motives from the consequences of their counterterrorism policies, reinforcing the strategic logic of minimizing collateral damage. Correspondent inference theory can explain not only why terrorist campaigns rarely work, but also perhaps why counterterrorism campaigns tend to breed even more terrorism.

129. See A. Petrova, Approval for Russian Military Actions in Chechnya Is Steadily Declining, Public Opinion Foundation Database, September 5, 2002, http:/ /bd.english.fom.ru/report/cat/ societas/Chechnya/truck_war/eof023305; and Susan Page, Poll: American Attitudes on Iraq Similar to Vietnam Era, USA Today, December 15, 2005.

Correspondence: Does Terrorism Ever Work?

William Rose and Rysia Murphy Does Terrorism Ever Work? The 2004 Max Abrahms Madrid Train Bombings

Correspondence

To the Editors:
Max Abrahmss argument that terrorism rarely works is compelling.1 He is not correct, however, that terrorist groups that primarily attack civilians never achieve their political objectives. The March 2004 Madrid train bombings offer an exception to Abrahmss thesis. The terrorist group that carried out the attack sought to compel Spain to withdraw its troops from Afghanistan and especially Iraq. The result was a partial success, because Spain did withdraw its forces from Iraq. This case study, developed below, helps to identify the uncommon conditions under which at least partial terrorist success is possible, and the ndings have implications for counterterrorism policy. Two additional arguments follow from this case. First, Abrahmss concentration on ofcial foreign terrorist organizations (FTOs) is too narrow to capture the emerging phenomenon of ad hoc terrorist networks that do not have formal afliation as a cell of a recognized FTO. Second, his focus on compelling governments to make policy concessions misses an important distinction between the impact of a terrorist attack on a government and on a countrys citizens. The Madrid attack never compelled the government led by the Popular Party to change policy on Spanish troops in Iraq. Instead it mobilized voters to elect a new government led by the Socialist Party because, in large part, this party campaigned on the promise to pull Spanish troops from Iraq.2 To be fair, the strengths of Abrahmss analysis outweigh its weaknesses. Contrary to some recent scholarship on terrorism, he convincingly shows that terrorist organizations rarely achieve their political objectives. Although he is not the only scholar to argue that terrorism usually fails, he is the rst to analyze systematically a large number of terrorist organizations and camWilliam Rose is Professor and Chair of the Government Department at Connecticut College. Rysia Murphy is a recent graduate of the college. The authors thank Julie Berson, Raj Chari, Nicholas Culver, Bruce Hoffman, Javier Jordn, Chaim Kaufmann, Arang Keshavarzian, Petter Nesser, and David Patton for their comments on an earlier draft. Max Abrahms is a doctoral candidate in political science at the University of California, Los Angeles. 1. Max Abrahms, Why Terrorism Does Not Work, International Security, Vol. 31, No. 2 (Fall 2006), pp. 4278. 2. For a more extensive treatment of this case study, see http:/ /www.conncoll.edu/academics/ web_proles/rose-murphy_long_IS_letter.pdf.
International Security, Vol. 32, No. 1 (Summer 2007), pp. 185192 2007 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

162

Correspondence: Does Terrorism Ever Work? 163

paigns. Further, he claries the conditions under which success can occur: when terrorist groups have limited objectives and, more important, when their main targets are military and not civilian. His analysis is theory informed, as he adapts correspondent inference theory to explain his observations. His article also stimulates contemplation, discussion, and new research projects including our Madrid case study. Abrahms asserts that terrorism rarely succeeds in achieving its political objectives, especially when a terrorist organization primarily targets civilians. To support his thesis, he sensibly forwards historical examples that demonstrate the arguments plausibility. All twenty-eight FTOs that he examines meet his expectations. He then turns to three case studies that match his predictions perfectly, and he explores them in depth to show how causation unfolded. His research is a shining example of a plausibility probe. We looked in vain for any acknowledgment that his thesis might be wrong at least sometimes, however, or that the degree of certainty of his conclusions is not high. Likewise, we hoped to nd a section in which he would encourage scholars to nd cases that challenge his thesis. Finding none, in a brainstorming session we thought of a case that does not t: the Madrid train bombings. On March 11, 2004, Spain was the site of the most devastating terrorist attack in Europe since World War II. Ten bombs exploded on three commuter trains full of passengers making the morning trip into Madrid. The attack resulted in 191 deaths and 1,500 wounded.3 Suspicion initially fell on the Basque separatist movement ETA, which had been the only active terrorist group in Spain. Evidence soon indicated, however, that an Islamic terrorist organization was responsible for the attack. The reality was that the individuals who helped plan, fund, and carry out the attack constituted an ad hoc jihadist network and not a particular terrorist organization.4 A number of these individuals had ties with high-prole members of the Moroccan Islamic Combatant Group, but some were linked to other groups or were unafliated. Many were homegrown radicals acting on their own rather than being directed by al-Qaida, the Moroccan Group, or any other organization. Spanish journalists and scholars refer to the ad hoc group that carried out the March 11, 2004, attacks as the 11-M network.
3. Raj S.Chari, The 2004 Spanish Election: Terrorism as a Catalyst for Change? West European Politics, Vol. 27, No. 5 (November 2004), p. 954. 4. Javier Jordn and Nicola Horsburgh, Spain and Islamism: Analysis of the Threat and Response 19952005, Mediterranean Politics, Vol. 11, No. 2 (July 2006), pp. 209229; and Gordon Corera, The Legacy of the Madrid Bombings, BBC News, February 14, 2007.

Contending with Terrorism 164

The primary political objective of the perpetrators was to compel Spain to end its military support for the U.S.-led occupations in Afghanistan and especially Iraq.5 The terrorists may also have hoped that Spain would be the weakest link in the U.S. coalition in Iraq, whereby its withdrawal would cause other coalition partners to follow.6 The document supporting this latter assertion was posted on an Islamist message board four months before the 11-M attack, and it contained a rational analysis of politics in Britain, Poland, and Spain and implications for the jihadist policy agenda. For two reasons, the documents author sensibly concluded that attacks on the Spanish would be most effective: public opposition to the war was greatest in Spain, and Spain was thought to have lower tolerance for casualties than Britain or Poland. Although the analysis focused on harming Spaniards in Iraq rather than in Europe, it set the stage for the 11-M network to predict at least some success. Translating the terrorists desire to compel Spanish troop withdrawal into terms consistent with Abrahmss analysis, it was a limited objective associated with demands over territory.7 Spain complied partially when it pulled its troops out of Iraq. This response does not follow two key predictions of Abrahmss theory. First, with civilians as the primary target, the public is expected to interpret the attack in maximalist terms. This predicted consequence did not occur. A day after the attack, newspapers throughout Spain conveyed that many people viewed the attack as a result of Spains involvement in Iraq war.8 One Elcano poll found that 49 percent of Spaniards believed that troop withdrawal would make Islamist attacks less likely. This nding supports the assertion that a signicant segment of the Spanish public correctly saw the limited objectives of the terrorist group.9 Second, Abrahmss theory predicts that countries whose citizens are targeted will not make policy concessions. Again

5. Javier Jordn and Robert Wesley, The Madrid Attacks: Results of Investigations Two Years Later, Terrorism Monitor, Vol. 4, No. 5 (March 9, 2006), pp. 14. 6. See Brynjar Lia and Thomas Hegghammer, Jihad Strategic Studies: The Alleged Al Qaida Policy Study Preceding the Madrid Bombings, Studies in Conict & Terrorism, Vol. 27, No. 5 (SeptemberOctober 2004), pp. 368371. 7. Abrahms writes, A terrorist group is said to have limited objectives when its demands are over territory. Specically, the group is ghting to either (1) evict a foreign military from occupying another country, or (2) win control over a piece of territory for the purpose of national self-determination. Abrahms, Why Terrorism Does Not Work, p. 53. 8. Quotations from two newspapers can be found in Jose A. Olmeda, Fear or Falsehood? Framing the 3/11 Terrorist Attacks in Madrid and Electoral Accountability, Defense and Security Working Paper, No. 24 (Madrid: Real Instituto Elcano, May 5, 2005), p. 25. 9. Charles Powell, Did Terrorism Sway Spains Election? Current History, November 2004, p. 380.

Correspondence: Does Terrorism Ever Work? 165

it is wrong. In national elections three days after the attack, voters deed earlier polls and voted out the government (led by the Peoples Party) that supported Spains intervention in Iraq. The surprise winner was the Socialist Party, which during the campaign had called for removing Spanish troops from Iraq. The troops were withdrawn several months later. Earlier in March the incumbent Popular Party led the polls by 5 percent, and commentators agree that it would have won the election had it not been for the terrorist attack. The opposition Socialist Party, headed by Jos Luis Rodrguez Zapatero, won 42.64 percent of the votes compared with 37.64 percent for the Popular Party. With 164 out of 350 seats in the Congress of Deputies, the Socialists were in a position to form a minority government.10 During the campaign Zapatero had promised to remove Spanish troops from Iraq by June 30, 2004. Twenty-eight percent of voters said that the bombings had inuenced their vote; and when Spanish troops were withdrawn a month earlier than expected, a strong majority supported the move.11 A methodical analysis concludes that the Madrid attack and the political atmosphere surrounding it affected election results by mobilizing about 1,700,000 voters who had not planned to vote and by discouraging approximately 300,000 voters from votingleading to a net 4 percent increase in voter turnout. In addition, more than 1 million voters switched their vote to the Socialist Party.12 Scholars highlight three factors to explain the surprise election outcome, although they may disagree on their relative importance. First, the Popular Party continued to blame ETA even after the leadership learned that an Islamic group was responsible for the attack. Admitting that Islamists were to blame would have exposed Prime Minister Jos Mara Aznar to the charge that his foreign policy contributed to the terrorist attack. Segments of the Spanish population sought to punish his party for the apparent deception. This factor is linked to a highly partisan framing contest between the government and the opposition, whereby the opposition presented the situation to the public more effectively than did the government.13 Second, a signicant part of the public
10. Chari, The 2004 Spanish Election, pp. 957958. 11. Powell, Did Terrorism Sway Spains Election? pp. 379, 381. 12. Narciso Michavila, War, Terror and Elections: Electoral Impact of the Islamist Terror Attacks on Madrid, Public Opinion Working Paper, No. 13 (Madrid: Real Instituto Elcano April 6, 2005), p. 31. 13. Olmeda, Fear or Falsehood? pp. 34. Narciso Michavila calls this circumstance the dual news manipulation hypothesis. Michavila, War, Terrorism, and Elections, pp. 2022. Javier Jordn and Nicola Horsburgh are concerned about the partisan politics that this factor revealed, whereby the two parties were playing politics with terrorism. Jordn and Horsburgh, Politics

Contending with Terrorism 166

mistrusted the Popular Party even before the Madrid attack. In reaction to several negative events since the party assumed an absolute majority in 2000, many Spaniards had begun to sense that the party was complacent and arrogant and that its decisionmaking process lacked transparency.14 Third, when the terrorists effectively signaled devastating punishment for Spains involvement in Iraq, the publics antiwar views became more pressing. Previously, many had opposed the war but supported Aznar. Following the 11-M attack, however, a signicant segment was no longer willing to support the Popular Party and its war policy. The conuence of conditions present in the Madrid case is probably quite rare and possibly unique. There may be additional, less obvious factors, and perhaps sometimes they carry more causal weight than these three. Although the generalizability of our ndings is low, the case does show that Abrahmss theory is not always right. Further research is needed to learn more about the reliability of his ndings. We encourage scholars to conduct studies with a larger collection of cases as well as to seek out cases that may be exceptions to the rule. If more exceptions can be found, scholarly understanding of variations in outcomes can widen and deepen. If successful terrorist attacks against civilians are indeed rare like we expect, then the impact of our ndings on Abrahmss four policy implications would not be extensive. Our ndings, however, suggest additional policy implications. They derive from the above-mentioned observation that the 11-M terrorist network probably targeted Spain because it was the weakest link among European members of the U.S.-led coalition in Iraq. To the extent that such an analysis inuenced the terrorists, therefore, several implications follow. First, more such attacks are likely in the future, because even partial success breeds support for them. Thus at least some terrorist groups will concentrate on targets where success is believed possible. Second, several relevant elements for countering terrorism are prescribed. One is that scholars and analysts worried about terrorism should conduct parallel analyses that could provide early warning signals for countries at higher risk for terrorist attacks. Because the terrorists are somewhat more likely to succeed in such countries, another recommendation is to give higher priority to preventing or foiling attacks. If

and Terrorism: The Madrid Attacks Case, in George Kassimeris, ed., Playing Politics with Terrorism: A Users Guide (London: Hurst, 2007), p. 2. 14. For an elaboration of these negative events, see Chari, The Spanish 2004 Election, pp. 954 955.

Correspondence: Does Terrorism Ever Work? 167

an attack occurs, it should be handled with less partisanship over framing the issue than occurred after the 11-M attack. Finally, because distrust of the government led by the Popular Party contributed to the terrorists success, governments have an interest in building and maintaining the trust of their publics. William Rose New London, Connecticut Rysia Murphy New London, Connecticut

The Author Replies:

I appreciate William Rose and Rysia Murphys thoughtful comments on my recent article in International Security.1 We agree on two main points: (1) terrorist groups that primarily target civilians fail to coerce their governments into making policy concessions; and (2) future research is needed to determine if there are any exceptions to the rule. Rose and Murphy focus on the second point and purport to identify an important outlier that does not t the rule: the March 2004 Madrid train bombings. The authors claim that the Madrid case undermines my article in two ways. First, they believe the attack shows that democracies are uniquely vulnerable to coercion because terrorists can sometimes inuence policies by scaring the electorate into ousting the incumbent leader. Specically, they argue that the Madrid attack represents a successful case of coercion because it bombed to power the antiwar candidate for prime minister, Jos Luis Rodrguez Zapatero, who kept his campaign pledge to withdraw Spanish troops from Iraq. Second, the authors assert that the Madrid case does not conform to correspondent inference theory because the Spanish public interpreted the bombings as evidence of 11-Ms intent to end the occupation, rather than to destroy the Spanish way of life, making coercion possible. The rst claim is weaker than the second: the Madrid case is an empirically problematic example of terrorist coercion, but it helps to delimit the antecedent conditions in which terrorist attacks on civilians might theoretically be effective. The Madrid case is an empirically problematic example of terrorist coercion
1. Max Abrahms, Why Terrorism Does Not Work, International Security, Vol. 31, No. 2 (Fall 2006), pp. 4278.

Contending with Terrorism 168

for three reasons. First, the argument that the 11-M attack coerced Spain into withdrawing from Iraq is questionable, because Zapatero might have won the election and then altered Spanish policy even in the absence of the attack. The surprise defeat of Prime Minister Jos Mara Aznar was actually not that surprising. In the days preceding the attack, Aznar held a narrow lead in most surveys, but the differences between the candidates voter estimates usually fell within the margin of error.2 Indeed, by early March the gap between the two candidates had closed: some surveys put Aznar ahead by a single point, while others had Zapatero winning by a razor-thin margin. In their study on the 2004 Spanish election, Ignacio Lago and Jos Ramn Montero state, It must be remembered that if the attacks had not taken place, either the PP [Aznars Popular Party] or the PSOE [Zapateros Socialist Party] could have won the election: only days before 11-M, the polls pointed to a technical tie.3 Furthermore, in the lead-up to the attack, the majority of Spaniards believed the country needed a change of government; a large percentage of the electorate was undecided; and in Spain undecided voters tend to vote for left-wing candidates such as Zapatero.4 Postelection returns conrm that Aznar did not lose any electoral support after the attack; as expected, the undecided voters gravitated toward the left-leaning candidates.5 The extent of electoral change or interelectoral volatility was not atypical for Spanish national elections.6 The claim that the 11-M attack successfully coerced Spain into withdrawing from Iraq is based on the counterfactual argument that without the attack, Zapatero would have lost the election, which is uncertain from the polling data. Second, Rose and Murphy imply that undecided voters gravitated toward Zapatero after the attack because it revealed the escalating costs of maintaining troops in Iraq, but Aznar compromised his electoral viability primarily by blaming the bombings on ETA. In testimony before the U.S. Senate Foreign Relations Committee, Philip Gordon stated, The [Aznar] government appears to have paid more of a price for misleading the public than for its policy on Iraq.7 In her study on the 2004 election, Georgina Blakeley found that the
2. Ignacio Lago and Jos Ramn Montero, The 2004 Election in Spain: Terrorism, Accountability, and Voting, Taiwan Journal of Democracy, Vol. 2, No. 1 (July 2006), p. 17. 3. Ibid., p. 34. 4. Georgina Blakeley, Its Politics, Stupid! The Spanish General Election of 2004, Parliamentary Affairs, Vol. 59, No. 2 (March 2006), p. 339. 5. See Enric Ordeix i Rigo, Aznars Political Failure or Punishment for Supporting the Iraq War? Hypotheses About the Causes of the 2004 Spanish Election Results, American Behavioral Scientist, Vol. 49, No. 4 (December 2005), p. 613. 6. Ingrid van Biezen, Terrorism and Democratic Legitimacy: Conicting Interpretations of the Spanish Elections, Mediterranean Politics, Vol. 10, No. 1 (March 2005), p. 107. 7. Philip H. Gordon, Madrid Bombings and U.S. Policy, testimony before the Senate Foreign Re-

Correspondence: Does Terrorism Ever Work? 169

point, therefore, is not that the bombings affected the general election, but rather, that the governments handling of the bombings had such profound consequences.8 The BBC likewise reported, It is sometimes wrongly claimed that the bombings themselves led directly to the defeat of the Conservative government and its replacement just days later by the Socialists. In fact, it was the perception that the government was misleading the public about who was responsible that did [the] most damage.9 Other foreign outlets, including Spanish television networks and the French newspaper Le Monde, reached the same conclusion.10 In sum, the dominant interpretation is that Zapateros postattack election gains were due mostly to Aznars mismanagement of the attacknot the Iraqi occupation that elicited itundermining the claim that the attack itself bombed Zapatero into power and effectively coerced the Spanish withdrawal. Third, it is doubtful that enlarging my sample of terrorist organizations or including ad hoc groups afliated with al-Qaida would lend support to the claim that democracies are uniquely vulnerable to terrorist coercion. I agree with Rose and Murphy that the Madrid case is probably quite rare and possibly unique because terrorism historically shifts the electorate to the right not the leftthereby empowering hard-liners who oppose accommodating the perpetrators.11 The most obvious example is in Israel, but the trend is also evident in the United States, where the mere release of Osama bin Ladens videotape the weekend before the 2004 presidential election boosted George W. Bushs electoral lead by two percentage points over his comparatively dovish opponent, John Kerry.12 Rose and Murphys stronger claim is that the Madrid case does not conform to correspondent inference theory. They point out that the train bombings targeted Spanish civilians, and yet the public did not revise its perception that allations Committee, 108th Cong., 2d sess., March 31, 2004, p. 2, http:/ /www.senate.gov/?foreign/ testimony/2004/GordonTestimony040331.pdf. 8. The Legacy of the Madrid Bombings, BBC News, February 15, 2007, http:/ /news.bbc.co.uk/ 2/hi/europe/6357599.stm. 9. Blakeley, Its Politics, Stupid! p. 342. 10. See Ray Suarez, Jos Gijon, and Salvador Sala, Terrorism and Politics in Spain, NewsHour with Jim Lehrer, March 16, 2004; and LEspagne Sanctionne le Mensonge dEtat [Spain punishes the governments lie], Le Monde, March 16, 2004, http:/ /www.pbs.org/newshour/bb/ international/jan-june04/spain_03-16.html. 11. See Christopher Hewitt, Consequences of Political Violence (Sudbury, Mass.: Dartmouth, 1993), pp. 80, 9798. 12. Claude Berrebi and Esteban F. Klor, On Terrorism and Electoral Outcomes: Theory and Evidence from the Israeli-Palestinian Conict, Journal of Conict Resolution, Vol. 50, No. 6 (January 2006), pp. 899925. See also David Paul Kuhn, Who Will Osama Surprise Help? CBSNews.com, October 30, 2004, http:/ /www.cbsnews.com/stories/2004/10/30/politics/main652438.shtml.

Contending with Terrorism 170

Qaida and its afliates aimed to achieve the limited policy goal of ending the occupation of Iraq. The Madrid case suggests that when a target country has strong preexisting beliefs that the terrorists are motivated by limited policy objectives, it will not always infer from attacks on its civilians that the terrorists are driven by ideological or maximalist objectives. Before the September 11 attacks, most Americans had little knowledge of alQaida. They therefore inferred from the consequences of the terrorist acts that the perpetrators aimed to harm American society and its values. Similarly, until the September 1999 apartment bombings, the Russian public knew little about the Chechnya campaign and therefore inferred from them that the Chechens had maximalist objectives. By contrast, Spanish opinion of alQaidas limited policy objectives was broadly and intensely established prior to the train bombings. Before the attack, 90 percent of the public disagreed with Aznars position that participating in the Iraq war made Spain safer from terrorism, an entrenched disconnect highlighted by two of the largest antiwar protests in history.13 Whereas news of the Chechnya occupation was withheld from the Russian public until it was targeted in September 1999, Spanish combat deaths in Iraq in August, October, and November 2003 were front-page news, reinforcing the perception that the terrorists aimed to end the occupation rather than Spains way of life.14 The Madrid example suggests that, in theory, terrorist attacks on civilians may potentially lead to policy concessions if the target country has extremely rm preexisting beliefs that the enemy is motivated by limited policy objectives. When this is the case, attacksregardless of target selectionwill communicate the escalating costs of defying the terrorists limited policy goals, making coercion possible. Future research is still needed, however, to identify a case of coercion where these antecedent conditions are present. Such a case would demonstrate not only that the attack(s) on civilians stoked the publics preexisting fears of defying the terrorists limited policy objectives, but that these fears actually changed the countrys policy. Rose and Murphys case study on the 11-M attack provides convincing evidence of the former, but not the latter. It is a basic truism that insurgency works, but terrorism does not. Max Abrahms Los Angeles, California

13. Mar Roman, Spain Wont Bow to Calls to Pull Troops, Associated Press, December 1, 2003. 14. Ibid.

What Terrorists Really Want

What Terrorists Really Want

Terrorist Motives and Counterterrorism Strategy

Max Abrahms

hat do terrorists want? No question is more fundamental for devising an effective counterterrorism strategy. The international community cannot expect to make terrorism unprotable and thus scarce without knowing the incentive structure of its practitioners.1 The strategic modelthe dominant paradigm in terrorism studiesposits that terrorists are rational actors who attack civilians for political ends. According to this view, terrorists are political utility maximizers; people use terrorism when the expected political gains minus the expected costs outweigh the net expected benets of alternative forms of protest.2 The strategic model has widespread currency in the policy community; extant counterterrorism strategies are designed to defeat terrorism by reducing its political utility. The most common strategies are to mitigate terrorism by decreasing its political benets via a strict no concessions policy; decreasing its prospective political benets via appeasement; or decreasing its political benets relative to nonviolence via democracy promotion. Are any of these counterterrorism strategies likely to work? Can terrorism be neutralized by withholding political concessions, granting political concesMax Abrahms is a doctoral candidate in political science at the University of California, Los Angeles. He conducted research for this article when he was a Research Associate at the Belfer Center for Science and International Affairs in the John F. Kennedy School of Government at Harvard University. The author would like to thank the following individuals for their time and comments: Robert Goldberg, Matthew Gottfried, Rex Hudson, Peter Krause, Deborah Larson, Karen Levi, Charles Mahoney, David Rapoport, Steven Spiegel, Arthur Stein, Marc Trachtenberg, Robert Trager, Jeff Victoroff, and the anonymous reviewers. 1. See Louise Richardson, What Terrorists Want: Understanding the Enemy, Containing the Threat (New York: Random House, 2006), p. 44. 2. Martha Crenshaw refers to what I call the strategic model as the instrumental model. For summaries of this model, see Crenshaw, Theories of Terrorism: Instrumental and Organizational Approaches, in David C. Rapoport, ed., Inside Terrorist Organizations (New York: Columbia University Press, 1988), pp. 1331; Crenshaw, The Logic of Terrorism: Terrorist Behavior as a Product of Strategic Choice, in Walter Reich, ed., Origins of Terrorism: Psychologies, Ideologies, Theologies, States of Mind (New York: Cambridge University Press, 1990), pp. 724; Gordon H. McCormick, Terrorist Decision Making, Annual Review of Political Science, Vol. 6 (June 2003), p. 482; and Gary C. Gambill, The Balance of Terror: War by Other Means in the Contemporary Middle East, Journal of Palestine Studies, Vol. 28, No. 1 (Autumn 1998), pp. 5166. For applications of the strategic model, see Robert A. Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (New York: Random House, 2005); Max Abrahms, Why Terrorism Does Not Work, International Security, Vol. 31, No. 2 (Fall 2006), pp. 4278; Andrew H. Kydd and Barbara F. Walter, The Strategies of Terrorism, International Security, Vol. 31, No. 1 (Summer 2006), pp. 4980; and James DeNardo, Power in Numbers: The Political Strategy of Protest and Rebellion (Princeton, N.J.: Princeton University Press, 1985), p. 3.
International Security, Vol. 32, No. 4 (Spring 2008), pp. 78105 2008 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

171

Contending with Terrorism 172

sions, or providing peaceful outlets for political change? In other words, does the solution to terrorism reside in diminishing its political utility? The answer depends on whether the strategic model is externally valid, that is, on whether terrorists are in fact rational people who attack civilians for political gain. If the model is empirically grounded, then the international community can presumably combat terrorism by rendering it an ineffective or unnecessary instrument of coercion. If the model is unfounded, however, then current strategies to reduce terrorisms political utility will not defuse the terrorism threat. Despite its policy relevance, the strategic model has not been tested. This is the rst study to comprehensively examine its empirical validity.3 The strategic model rests on three core assumptions: (1) terrorists are motivated by relatively stable and consistent political preferences; (2) terrorists evaluate the expected political payoffs of their available options, or at least the most obvious ones; and (3) terrorism is adopted when the expected political return is superior to those of alternative options. Does the terrorists decisionmaking process conform to the strategic model? The answer appears to be no. The record of terrorist behavior does not adhere to the models three core assumptions. Seven common tendencies of terrorist organizations atly contradict them. Together, these seven terrorist tendencies represent important empirical puzzles for the strategic model, posing a formidable challenge to the conventional wisdom that terrorists are rational actors motivated foremost by political ends. Major revisions in the dominant paradigm in terrorism studies and the policy communitys basic approach to ghting terrorism are consequently in order. This article has four main sections. The rst section summarizes the strategic models core assumptions and the empirical evidence that would disconrm them.4 The second section demonstrates the empirical weakness of the strategic model. In this section, I present the seven puzzlesbased on the records of dozens of terrorist organizations from the late 1960s to the present,

3. Martha Crenshaw has raised important questions about the strategic models empirical validity. See, for example, Crenshaws Theories of Terrorism and The Logic of Terrorism. 4. There is a debate within the social sciences about whether a hypothesiss assumptions need to be empirically valid. Milton Friedman famously argued that the merit of a hypothesis depends strictly on its predictive power, whereas many other theorists believe that the core assumptions of a hypothesis must also be grounded in reality. For a summary of this theoretical debate, see Jack Melitz, Friedman and Machlup on the Signicance of Testing Economic Assumptions, Journal of Political Economy, Vol. 73, No. 1 (February 1965), pp. 3760. In the eld of international relations, most theory testing takes the assumptions as exogenous, but this is not always the case. For two important exceptions that criticize realism because of its assumption of anarchy, see David A. Baldwin, ed., Neorealism and Neoliberalism: The Contemporary Debate (New York: Columbia University Press, 1993); and Alexander Wendt, Anarchy Is What States Make of It: The Social Construction of Power Politics, International Organization, Vol. 46, No. 2 (Spring 1992), pp. 391425.

What Terrorists Really Want 173

supplemented with theoretical arguments from the bargaining and coercion literaturesthat cannot be reconciled with the models underlying assumptions. The third section develops an alternative explanation for terrorism. The argument is not that terrorists are crazy or irrational; as Louise Richardson notes, psychiatric proles of terrorists are virtually unanimous that their primary shared characteristic is their normalcy.5 Rather, I contend that the strategic model misspecies terrorists incentive structure; the preponderance of empirical and theoretical evidence reveals that terrorists are rational people who use terrorism primarily to develop strong affective ties with fellow terrorists.6 If terrorists generally attach utmost importance to the social benets of using terrorism, then extant strategies to reduce its political benets will fail to counter the terrorism threat. In the nal section, I suggest a reorientation of counterterrorism strategy in light of what terrorists really seem to want.

The Strategic Model

In classical economic theory, rational agents (1) possess stable and consistent preferences; (2) compare the costs and benets of all available options; and (3) select the optimal option, that is, the one that maximizes output.7 Modern decision theory recognizes that decisionmakers face cognitive and informational constraints. Rational actor models therefore typically relax each assumption such that the rational agent must only (1) possess relatively stable and consistent goals; (2) weigh the expected costs and benets of the most obvious options; and (3) select the option with the optimal expected utility.8 The strategic model is explicitly predicated on this trio of assumptions. First, the strategic model assumes that terrorists are motivated by relatively stable and consistent political goals, which are encoded in the political plat-

5. Richardson, What Terrorists Want, p. 14. 6. Sociologists routinely treat social objectives as rational. See, for example, Jeffrey Pfeffer, Organizations and Organization Theory (Boston: Pitman, 1982), pp. 9, 4243, 62, 72, 256. Rational choice theorists in economics and political science also frequently treat social objectives as rational. See, for example, Jon Elster, Introduction, in Elster, ed., Rational Choice (Oxford: Basil Blackwell, 1986), p. 1; Gary Becker, The Economic Approach to Human Behavior, in Elster, Rational Choice, pp. 115, 119; and John C. Harsanyi, Rational Choice Models of Political Behavior vs. Functionalist and Conformist Theories, World Politics, Vol. 21, No. 4 (July 1969), pp. 513538. 7. See David M. Kreps, A Course in Microeconomic Theory (Princeton, N.J.: Princeton University Press, 1990), p. 480; Elster, Introduction, pp. 4, 16; Sidney Verba, Assumptions of Rationality and Non-Rationality in Models of the International System, World Politics, Vol. 14, No. 1 (October 1961), pp. 93117; and Graham Allison and Philip Zelikow, Essence of Decision: Explaining the Cuban Missile Crisis, 2d ed. (New York: Longman, 1999), pp. 1718. 8. Elster, Introduction, p. 5; and Allison and Zelikow, Essence of Decision, p. 18.

Contending with Terrorism 174

form of the terrorist organization. That West Germanys Red Army Faction (RAF) identied itself as Marxist, for example, implies that RAF members participated in the organization to achieve its stated revolutionary agenda.9 Disconrming evidence would therefore reveal that the RAF expressed a protean set of political objectives, fought mainly against other groups with its identical political platform, or continued using terrorism after its stated political grievances had been resolved. Second, the strategic model assumes that terrorism is a calculated course of action and that efcacy is the primary standard by which terrorism is compared with other methods of achieving political goals.10 Specically, the model assumes that terrorist groups weigh their political options and resort to terrorism only after determining that alternative political avenues are blocked.11 Disconrming evidence would therefore demonstrate that terrorism is not a strategy of last resort and that terrorist groups reexively eschew potentially promising nonviolent political alternatives. Third, the strategic model assumes that the decision to use terrorism is based on the logic of consequence, that is, its political effectiveness relative to alternative options.12 Specically, it is assumed that terrorist organizations achieve their political platforms at least some of the time by attacking civilians; that they possess reasonable expectations of the political consequences of using terrorism based on its prior record of coercive effectiveness; and that they abandon the armed struggle when it consistently fails to coerce policy concessions or when manifestly superior political options arise.13 Disconrming evidence would therefore reveal that terrorist organizations do not achieve their political platforms by attacking civilians; that they do not renounce terrorism in spite of consistent political failure or manifestly superior political options; or that they do not even use terrorism in a manner that could potentially coerce policy concessions from the target country. Below I identify and then describe seven tendencies of terrorist organizations that challenge the strategic model with disconrming evidence of its core assumptions.
9. See McCormick, Terrorist Decision Making, p. 482; and Crenshaw, Theories of Terrorism, pp. 15, 27. 10. McCormick, Terrorist Decision Making, p. 481. 11. Crenshaw, Theories of Terrorism, p. 16. See also Alex P. Schmid and Albert J. Jongman, Political Terrorism (Amsterdam: North-Holland, 1988), pp. 122123. 12. See James G. March, A Primer on Decision Making: How Decisions Happen (New York: Free Press, 1994), pp. 23. See also Crenshaw, The Logic of Terrorism, p. 20. 13. See Pape, Dying to Win, p. 62. See also Crenshaw, Theories of Terrorism, p. 16; and Schmid and Jongman, Political Terrorism, pp. 122123.

What Terrorists Really Want 175

The Seven Puzzling Tendencies of Terrorist Organizations

Seven empirical puzzles vitiate the strategic models premise that terrorists are rational people who are motivated mainly to achieve their organizations stated political goals. The seven puzzles contradicting the strategic model are (1) terrorist organizations do not achieve their stated political goals by attacking civilians; (2) terrorist organizations never use terrorism as a last resort and seldom seize opportunities to become productive nonviolent political parties; (3) terrorist organizations reexively reject compromise proposals offering signicant policy concessions by the target government; (4) terrorist organizations have protean political platforms; (5) terrorist organizations generally carry out anonymous attacks, precluding target countries from making policy concessions; (6) terrorist organizations with identical political platforms routinely attack each other more than their mutually professed enemy; and (7) terrorist organizations resist disbanding when they consistently fail to achieve their political platforms or when their stated political grievances have been resolved and hence are moot. puzzle #1: coercive ineffectiveness In the strategic model, people participate in a terrorist organization because they are deeply committed to achieving its political platform. The strategic model is explicit that success for a terrorist organization requires the attainment of its stated political goals.14 Even if all other strategies are blocked, terrorism is not based on the logic of consequence and is thus irrational according to the model unless organizations achieve their political platforms at least some of the time by attacking civilians.15 A major puzzle for the model then is that although terrorism is by denition destructive and scary, organizations rarely if ever attain their policy demands by targeting civilians.16 The Rand Corporation reported in the 1980s that terrorists have been unable to translate the consequences of terrorism into concrete political gains. . . . In that sense terrorism has failed. It is a fundamental failure.17 Martha
14. Crenshaw, Theories of Terrorism, p. 15. 15. Sun-Ki Chai, An Organizational Economics Theory of Antigovernment Violence, Comparative Politics, Vol. 26, No. 1 (October 1993), p. 100. 16. The strategic model focuses on strategic terrorism, not redemptive terrorism. The former aims to coerce a government into changing its policies, whereas the latter is intended solely to obtain specic human or material resources such as prisoners or money. On this distinction, see Abrahms, Why Terrorism Does Not Work, p. 46. 17. Bonnie Cordes, Bruce Hoffman, Brian M. Jenkins, Konrad Kellen, Sue Moran, and William Sater, Trends in International Terrorism, 1982 and 1983 (Santa Monica, Calif.: RAND, 1984), p. 49.

Contending with Terrorism 176

Crenshaw remarked at the time that terrorist organizations do not obtain the long-term ideological objectives they claim to seek, and therefore one must conclude that terrorism is objectively a failure.18 Thomas Schelling reached the same conclusion in the 1990s, noting that terrorist attacks never appear to accomplish anything politically signicant.19 In a study assessing terrorisms coercive effectiveness, I found that in a sample of twenty-eight well-known terrorist campaigns, the terrorist organizations accomplished their stated policy goals zero percent of the time by attacking civilians.20 Although several political scientists have developed theoretical models predicated on the notion that terrorism is an effective coercive instrument, their research fails to identify a single terrorist organization that has achieved its political platform by attacking civilians.21 Terrorist organizations may not realize their policy demands by targeting civilians, but do these attacks generally advance their political cause? Walter Laqueur notes that for terrorist organizations, the political consequences of their violence is nearly always negative.22 Polls show, for example, that after the Irish Republican Army (IRA) attacked the British public, the British people became signicantly less likely to favor withdrawing from Northern Ireland.23 Similar trends in public opinion have been registered after groups attacked civilians in Egypt, Indonesia, Israel, Jordan, the Philippines, and Russia.24 Although the international community frequently appeals for target countries to appease terrorists, terrorist attacks on civilians have historically empowered hard-liners who oppose, as a matter of principle, accommodating the perpetrators. For this reason, numerous studies have shown that terrorist attacks tend to closenot openthe bargaining space between what terrorist groups
18. Crenshaw, Theories of Terrorism, p. 15. 19. Thomas C. Schelling, What Purposes Can International Terrorism Serve? in R.G. Frey and Christopher W. Morris, eds., Violence, Terrorism, and Justice (New York: Cambridge University Press, 1991), p. 20. 20. Abrahms, Why Terrorism Does Not Work, pp. 4278. 21. Proponents of the strategic model claim that terrorism is an effective coercive instrument. Yet their conrming examples are limited to successful guerrilla campaigns, which are directed against military and diplomaticnot civiliantargets. See, for example, Pape, Dying to Win, p. 39; and Kydd and Walter, The Strategies of Terrorism, p. 49. On the distinction between terrorist and guerrilla campaigns, see Abrahms, Why Terrorism Does Not Work, pp. 4446. 22. Walter Laqueur, Terrorism (Boston: Little, Brown, 1977), p. 117. 23. Peter R. Neumann and Mike Smith, Strategic Terrorism: The Framework and Its Fallacies, Journal of Strategic Studies, Vol. 28, No. 4 (August 2005), p. 587. 24. See, for example, John Mueller, Overblown: How Politicians and the Terrorism Industry Inate National Security Threats, and Why We Believe Them (New York: Free Press, 2006), p. 184; and Claude Berrebi and Esteban F. Klor, On Terrorism and Electoral Outcomes: Theory and Evidence from the Israeli-Palestinian Conict, Journal of Conict Resolution, Vol. 50, No. 6 (Spring 2006), pp. 899 925.

What Terrorists Really Want 177

demand and what target governments are willing to offer.25 In sum, the strategic model posits that rational people participate in terrorist organizations to achieve their stated political goals. In practice, however, terrorism does not accomplish them. Predictably, terrorisms political ineffectiveness has led scholars to question its rationality and motives.26 puzzle #2: terrorism as the rst resort The strategic model assumes that groups turn to terrorism only after weighing their political options and determining they are blocked. In the parlance of the model, the decision to use terrorism is a last resort, a constrained choice imposed by the absence of political alternatives.27 In reality, terrorist groups do not embrace terrorism as a last resort and seldom elect to abandon the armed struggle to become nonviolent political parties. Terrorist groups never lack political alternatives.28 Large-n studies show, rst, that only the most oppressive totalitarian states have been immune from terrorism, and second, that the number of terrorist organizations operating in a country is positively associated with its freedom of expression, assembly, and associationconditions conducive to effecting peaceful political change.29 The paradox of terrorism is that terrorist groups tend to target societies with the greatest number of political alternatives, not the fewest.30 Case studies on terrorist organizations conrm that the decision to use terrorism is not a last resort.31 In their study of Italian terrorist organizations in the mid-1960s and early 1970s, for example, Donatella Della Porta and Sidney Tarrow found that terrorism was part of the protest repertoire from the very beginning, even

25. See, for example, Alan B. Krueger, What Makes a Terrorist: Economics and the Roots of Terrorism (Princeton, N.J.: Princeton University Press, 2007), pp. 130131; and Christopher Hewitt, Consequences of Political Violence (Sudbury, Mass.: Dartmouth, 1993), pp. 80, 9798. 26. See Ariel Merari, Terrorism as a Strategy of Insurgency, Terrorism and Political Violence, Vol. 5, No. 4 (Winter 1993), p. 229; Richardson, What Terrorists Want, p. 75; and Martha Crenshaw, How Terrorists Think: What Psychology Can Contribute to Understanding Terrorism, in Lawrence Howard, ed., Terrorism: Roots, Impact, Responses (New York: Praeger, 1992), p. 75. 27. See McCormick, Terrorist Decision Making, p. 483; Crenshaw, How Terrorists Think, p. 72; and DeNardo, Power in Numbers, p. 242. 28. Crenshaw, How Terrorists Think, p. 71. 29. See, for example, William L. Eubank and Leonard B. Weinberg, Does Democracy Encourage Terrorism? Terrorism and Political Violence, Vol. 6, No. 4 (Winter 1994), pp. 417443; and Leonard B. Weinberg and William L. Eubank, Terrorism and Democracy: What Recent Events Disclose, Terrorism and Political Violence, Vol. 10, No. 1 (Spring 1998), pp. 108118. See also Laqueur, Terrorism, p. 220. 30. Bonnie Cordes, When Terrorists Do the Talking: Reections on Terrorist Literature, in Rapoport, Inside Terrorist Organizations, p. 150. See also Walter Laqueur, Interpretations of Terrorism: Fact, Fiction, and Political Science, Journal of Contemporary History, Vol. 12, No. 1 (January 1977), p. 1. 31. Laqueur, Interpretations of Terrorism, p. 1; and Laqueur, Terrorism, p. 80.

Contending with Terrorism 178

though opportunity abounded for nonviolent, constitutionally protected political protest.32 More generally, the authors concluded that terrorism tended to appear from the very beginning of the protest cycle for the dozens of terrorist organizations operating in Western Europe during this period.33 Relatively few terrorist organizations have elected to abandon the armed struggle to become normal political parties.34 More commonly, terrorist organizations toil alongside peaceful parties, refuse to lay down their arms after participating in national elections, or sabotage open elections that would have yielded major political gains for the group, such as todays militant Sunni groups in Iraq.35 In many instances, nonviolent strategies are believed to be more policy effective, but terrorist organizations tend to retain, in one form or another, the path of armed resistance.36 For these reasons, Crenshaw has sensibly asked, Why use terrorism when it cannot be justied . . . as a last resort?37 The answer of most terrorism experts is that terrorist groups seem to possess an innate compulsion to engage in terrorism and an unswerving belief in its desirability over nonviolence, contradicting the strategic models assumption that groups employ terrorism only as a last resort upon evaluating their political options.38 puzzle #3: reexively uncompromising terrorists As a rule, terrorist organizations do not compromise with the target country. Bruce Hoffman has observed that terrorist organizations are notorious for their
32. Donatella Della Porta and Sidney Tarrow, Unwanted Children: Political Violence and the Cycle of Protest in Italy, 19661973, European Journal of Political Research, Vol. 14, Nos. 56 (November 1986), p. 616. See also Peter H. Merkl, ed., Political Violence and Terror: Motifs and Motivations (Los Angeles: University of California Press, 1986), p. 146. 33. Della Porta and Tarrow, Unwanted Children, pp. 14, 53. 34. Paul Wilkinson, Terrorism versus Democracy: The Liberal State Response (London: Frank Cass, 2000), p. 59. 35. Examples of the rst point include the dozens of United States and European-based Marxist terrorist organizations from the late 1960s to the late 1980s, such as Action Directe, the Communist Combatant Cells, the RAF, the Red Brigades, and the Weather Underground. Examples of the second point, including terrorist organizations overtly aligned with a parent political wing, are Aum Shinrikyo, the Communist Party of Nepal, the Communist Party of the Philippines, Dev Sol, ETA, Fatah, Hamas, Harakat ul-Mujahidin, Hezbollah, the IRA, the Japanese Red Army, Kach, the PKK, the Revolutionary Armed Forces of Colombia, and the Revolutionary United Front. On the relationship between terrorist organizations and political parties, see Leonard Weinberg and Ami Pedahzur, Political Parties and Terrorist Groups (London: Routledge, 2003). 36. See Maria Stephan and Erica Chenoweth, Does Terrorism Work? Comparing Strategies of Asymmetric Warfare, presentation to the Centre for Defence Studies, Kings College, London, March 2007. See also Crenshaw, How Terrorists Think, p. 71; and Laqueur, Interpretations of Terrorism, p. 1. 37. Crenshaw, How Terrorists Think, p. 72. 38. Bruce Hoffman, Inside Terrorism (New York: Columbia University Press, 1998), p. 174; and Audrey Kurth Cronin, How al-Qaida Ends: The Decline and Demise of Terrorist Groups, International Security, Vol. 31, No. 1 (Summer 2006), p. 11. See also Laqueur, Terrorism, p. 119.

What Terrorists Really Want 179

resolutely uncompromising demands.39 Crenshaw has likewise noted that terrorist organizations are characterized by an intransigent refusal to compromise.40 It is far more common for them to derail negotiations by ramping up their attacks.41 In fact, no peace process has transformed a major terrorist organization into a completely nonviolent political party.42 Proponents of the strategic model claim that terrorists are acting rationally in opposing compromise because their policy preferences are inherently extreme, precluding a mutually acceptable bargain solution with the target country.43 This argument is empirically and theoretically awed. First, terrorism is an extremism of means, not ends.44 Many terrorist organizations profess surprisingly moderate political positions. Russian terrorist groups of the mid-nineteenth century were known as liberals with a bomb because they sought a constitution with elementary civil freedoms.45 The expressed goal of the al-Aqsa Martyrs Brigades is to achieve a Palestinian state in the West Bank and Gaza Stripa policy preference held by most of the international community. Robert Pape points out that even in his sample of contemporary suicide terrorist organizations, the terrorists political aims, if not their methods, are often more mainstream than observers realize; they generally reect quite common, straightforward nationalist self-determination claims of their community . . . goals that are typically much like those of other nationalists within their community.46 Yet terrorist organizations rarely commit to negotiations, even when these would satisfy a signicant portion of their stated political grievances. The al-Aqsa Martyrs Brigades, for example, responded with an unprecedented wave of terror to Israeli Prime Minister Ehud Baraks January 2001 offer of the Gaza Strip and most of the West Bank.47 Second, even when terrorist groups are motivated by extreme policy preferences, a negotiated settlement is always preferable to political deadlock, ac-

39. Hoffman, Inside Terrorism, p. 128. 40. Martha Crenshaw, An Organizational Approach to the Analysis of Political Terrorism, Orbis, Vol. 29, No. 3 (Fall 1985), p. 481. 41. See Andrew Kydd and Barbara F. Walter, Sabotaging the Peace: The Politics of Extremist Violence, International Organization, Vol. 56, No. 2 (Spring 2002), pp. 263296. See also Stephen John Stedman, Spoiler Problems in Peace Processes, International Security, Vol. 22, No. 2 (Fall 1997), pp. 553. 42. Wilkinson, Terrorism versus Democracy, p. 59. 43. See, for example, David A. Lake, Rational Extremism: Understanding Terrorism in the Twenty-rst Century, Dialog-IO, Vol. 1, No. 1 (Spring 2002), pp. 1529. 44. Anthony Oberschall, Explaining Terrorism: The Contribution of Collective Action Theory, Sociological Theory, Vol. 22, No. 1 (March 2004), p. 26. On the types of political demands that terrorist organizations make, see Abrahms, Why Terrorism Does Not Work, pp. 5354. 45. Laqueur, Terrorism, p. 37. 46. Pape, Dying to Win, p. 43. 47. See Dennis Ross, The Missing Peace: The Inside Story of the Fight for Middle East Peace (New York:

Contending with Terrorism 180

cording to the logic of the strategic model.48 Most bargaining theorists do not accept issue indivisibility between rational adversaries as a viable explanation for conict because contested issues are typically complex and multidimensional, enabling the warring parties to nd linkages and side payments that create a mutually benecial bargain solution.49 Hamas, for example, has opposed surrendering claims to all of historic Palestine, but the Islamist group professes to value the West Bank and Gaza Strip. If acting solely to optimize its political platform, Hamas would therefore be expected to accept the Palestinian territories in exchange for peace. Hamas, however, acts as a spoiler, depriving its members of policy goals that the organization purports to support. In sum, bargaining theory dictates that the rational course of action is for terrorist organizations to compromiseeven if that means securing only partial concessions over continued deadlockbut they rarely do. The tendency for terrorist organizations to reexively oppose compromise undercuts the strategic models assumptions that terrorists weigh the most obvious political options and select terrorism because of its relative political effectiveness. puzzle #4: protean political platforms The strategic model assumes that terrorists are motivated by relatively stable and consistent goals reected in their organizations political platform. But terrorist organizations often have protean political platforms.50 The Rand Corporation described Frances Action Directe in the 1980s as a chameleon organization that rapidly refocused on a host of faddish policy issues, from opposing Israel to nuclear energy to the Catholic Church.51 For Ely Karmon, Action Directes hodgepodge of stated goals reected the organizations inability to agree on basic ideological principles.52 Action Directe was an unusually

Farrar, Straus and Giroux, 2004). See also Robert Malley, Israel and the Arafat Question, New York Review of Books, Vol. 51, No. 15 (October 7, 2004), pp. 1923. 48. See DeNardo, Power in Numbers, p. 90; and Navin A. Bapat, State Bargaining with Transnational Terrorist Groups, International Studies Quarterly, Vol. 50, No. 1 (March 2006), p. 214. For a seminal work on compromise from a rationalist bargaining perspective, see Robert Powell, Bargaining Theory and International Conict, Annual Review of Political Science, Vol. 5 (June 2002), pp. 130. 49. See James D. Fearon, Rationalist Explanations for War, International Organization, Vol. 49, No. 3 (Summer 1995), pp. 382, 390; and Robert Powell, War as a Commitment Problem, International Organization, Vol. 60, No. 1 (Winter 2006), pp. 176178, 180. For a contrarian perspective on issue indivisibility, see Monica Duffy Toft, Issue Indivisibility and Time Horizons as Rationalist Explanations for War, Security Studies, Vol. 15, No. 1 (JanuaryMarch 2006), pp. 3469. 50. See Crenshaw, Theories of Terrorism, p. 20. See also Cordes et al., Trends in International Terrorism, p. 50. 51. Quoted in Crenshaw, Theories of Terrorism, p. 20. 52. Ely Karmon, Coalitions between Terrorist Organizations: Revolutionaries, Nationalists, and Islamists (Leiden, The Netherlands: Koninkliijke Brill, 2005), p. 141.

What Terrorists Really Want 181

capricious terrorist organization, but even the crucial case of al-Qaida has purported to support a highly unstable set of political goals.53 In The Protean Enemy, Jessica Stern charts al-Qaidas transitory political agenda, as the movement morphed rapidly and unpredictably from waging defensive jihad against the Soviets in Afghanistan to ghting local struggles in Bosnia, the Philippines, Russia, Spain, and in Muslim countries to its eventual targeting of the far enemy in the late 1990s. The marked uidity of al-Qaidas political rationale is reected in the fatwas Osama bin Laden issued throughout the 1990s, which contain a litany of disparate grievances against Muslims.54 Only in his fourth call to arms on October 7, 2001, did he emphasize the Israeli occupation, which is known in policy circles as his belated concern.55 Al-Qaida members have frequently criticized the inconsistency of their organizations jihadi message. The al-Qaida military strategist, Abul-Walid, complained that with its hasty changing of strategic targets, al-Qaida was engaged in nothing more than random chaos.56 Other disgruntled al-Qaida members have reproached the organization for espousing political objectives that shift with the wind.57 Not surprisingly, the opportunistic nature of al-Qaidas political platform has led scholars to question the movements dedication to achieving it.58 Some of the most important terrorist organizations in modern history have pursued policy goals that are not only unstable but also contradictory. The Basque separatist group ETA, for example, is criticized for failing to produce a consistent ideology, as its political goals have wavered from ghting to overturn the Franco dictatorship in Spain to targeting the emergent democratic governmenta progression similar to that of the Shining Path, Perus most notorious terrorist organization.59 The Kurdistan Workers Party Turkeys most dangerous contemporary terrorist group (known by the Kurdish acronym PKK)has likewise vacillated between advocating jihad, a Marxist revolution, and a Kurdish homeland governed without Islamist or
53. For an excellent recent study on al-Qaidas protean nature, see Vahid Brown, Cracks in the Foundation: Leadership Schisms in al-Qaida from 19892006, CTC Report (West Point, N.Y.: Combating Terrorism Center, September 2007), p. 2. 54. Jessica Stern, The Protean Enemy, Foreign Affairs, Vol. 82, No. 4 (July/August 2003), p. 1. 55. Samuel R. Berger and Mona Sutphen, Commandeering the Palestinian Cause: Bin Ladens Belated Concern, in James F. Hoge Jr. and Gideon Rose, eds., How Did This Happen? Terrorism and the New War (New York: PublicAffairs, 2001), p. 123. 56. Quoted in Brown, Cracks in the Foundation, p. 10. 57. Omar Nasiri, Inside the Jihad: My Life with Al Qaeda, A Spys Story (New York: Basic Books, 2006), p. 295. 58. Cronin, How al-Qaida Ends, pp. 4142. 59. Crenshaw, An Organizational Approach to the Analysis of Political Terrorism, p. 71.

Contending with Terrorism 182

Marxist principles.60 The Abu Nidal Organization staged countless attacks against Syria in the 1980s and then almost overnight switched allegiance by becoming a Syrian proxy.61 According to Leonard Weinberg, the most feared international terrorist group of the 1980s was willing to carry out a terrorist attack on behalf of any cause, even conicting ones.62 Similarly, Laqueur points out that many well-known groups that began on the extreme right such as the Argentine Montoneros, Colombian M-19, and the Popular Front for the Liberation of Palestineended up on the left as far as their phraseology was concerned.63 Hoffman has likewise noted that in the 1980s, right-wing terrorist groups in West Germany temporarily adopted left-wing rhetoric and began attacking targets that are the traditional choice of left-wing groups. Predictably, the police initially suspected that dozens of their attacks were the work of communist groups.64 That terrorist organizations often pursue unstable, even inconsistent, political goals undermines the assumption that terrorist members are motivated by a stable and consistent utility function encoded in their organizations political platform. puzzle #5: anonymous attacks The strategic model assumes that terrorism is based on the logic of consequence, specically, its ability to coerce policy concessions from the target country by conveying the costs of noncompliance. For this reason, proponents of the model describe terrorism as a form of credible signaling or costly signaling.65 A basic principle of coercion, however, is that the coercer must convey its policy demands to the coerced party.66 A puzzle for the strategic model is that most of the time terrorist organizations neither issue policy demands nor even take credit for their attacks. Since the emergence of modern terrorism in 1968, 64 percent of worldwide terrorist attacks have been carried out by unknown perpetrators. Anonymous terrorism has been rising, with three out of four attacks going unclaimed since September 11, 2001.67 Anonymous terrorism is particularly prevalent in Iraq,
60. See Ami Pedahzur, Suicide Terrorism (Cambridge: Polity, 2005), pp. 87, 89. See also Mia Bloom, Dying to Kill: The Allure of Suicide Terror (New York: Columbia University Press, 2005), p. 112. 61. Walter Laqueur, The Age of Terrorism (Boston: Little, Brown, 1987), pp. 287288. 62. Leonard Weinberg, Global Terrorism: A Beginners Guide (Oxford: Oneworld, 2005), p. 83. 63. Laqueur, The Age of Terrorism, p. 205. 64. Bruce Hoffman, Right-Wing Terrorism in West Germany, No. P-7270 (Santa Monica, Calif.: RAND, 1986), pp. 815. 65. Pape, Dying to Win, p. 29; and Kydd and Walter, Strategies of Terrorism, p. 50. 66. Robert J. Art and Patrick M. Cronin, eds., The United States and Coercive Diplomacy (Washington, D.C.: United States Institute of Peace Press, 2003), p. 371. 67. Authors calculations from RANDs MIPT data set, http:/ /www.tkb.org.

What Terrorists Really Want 183

where the U.S. military has struggled to determine whether the violence was perpetrated by Shiite or Sunni groups with vastly different political platforms.68 Policy demands are rarely forthcoming, even when the terrorist organization divulges its identity to the target country.69 In the early 1990s, Schelling captured this point: Usually there is nothing to negotiate. A soldier is killed in a disco in Germany. A bomb explodes in front of an Israeli consulate. Japanese Black Septembrists unpack automatic weapons in the Lod airport and start shooting. The perpetrators dont ask anything, demand anything.70 The tendency for terrorist organizations to refrain from issuing policy demands increased in the late 1990s, leading Hoffman to conclude that the coercive logic of terrorism is seriously awed.71 After the attacks of September 11, David Lake also observed that the terrorists did not issue prior demands, and therefore a theory premised on coercion would seem ill-suited to explaining such violence.72 In sum, the strategic model assumes that terrorism is an effective coercive instrument. Yet terrorist groups rarely convey through violence their policy preferences to the target country, precluding even the possibility of successful coercion. puzzle #6: terrorist fratricide The strategic model assumes that terrorists are motivated by a consistent utility function reected in their organizations political platform, but terrorist organizations with the same political platform routinely undercut it in wars of annihilation against each other. Particularly in the early stages of their existence, terrorist organizations purporting to ght for a common cause frequently attack each other more than their mutually declared enemy. The Tamil Tigers, for example, did not target the Sinhalese government in the mid-1980s. Instead, it engaged in a systematic annihilation of other Tamil organizations espousing the same cause of national liberation.73 Pape

68. See Pedahzur, Suicide Terrorism, pp. 114115. 69. See Eqbal Ahmad, Comprehending Terror, MERIP, No. 140 (MayJune 1986), p. 3; and Bonnie Cordes, Euroterrorists Talk about Themselves: A Look at the Literature, in Paul Wilkinson and Alasdair M. Stewart, eds., Contemporary Research on Terrorism (Aberdeen, Scotland: Aberdeen University Press, 1987), p. 331. 70. Schelling, What Purposes Can International Terrorism Serve? p. 24. 71. Bruce Hoffman, Why Terrorists Dont Claim Credit, Terrorism and Political Violence, Vol. 9, No. 1 (Spring 1997), p. 1. See also Mark Juergensmeyer, The Logic of Religious Violence, in Rapoport, Inside Terrorist Organizations, p. 172. 72. Lake, Rational Extremism, p. 15. 73. Shri D.R. Kaarthikeyan, Root Causes of Terrorism? A Case Study of the Tamil Insurgency and

Contending with Terrorism 184

observes that the apparent implication of the Tigers target selection is that the violence had little to do with the political grievances of Tamil society or the relationship between the Tamils and their Sinhalese opponents.74 Ami Pedahzur alludes to the fact that the Tigers target selection is difcult to reconcile with the strategic model: In contrast to what might be expected from a guerrilla or a terrorist organization whose [expressed] goals were national liberation, the rst violent actions initiated by the Tigers were not aimed at any army forces or Sinhalese politicians. . . . The Tigers systematically liquidated leaders and sometimes activists of other [Tamil] organizations.75 Similarly, in the early years of the Algerian War, the National Algerian Movement (known by the French acronym MNA) and the National Liberation Front (FLN) mainly attacked each other, not their French occupiers.76 Proponents of the strategic model might reason that the MNA and the FLN were battling to determine the political future of Algeria. Benjamin Stora points out, however, that for both organizations the nature of the future independent Algerian society was not at issue.77 Predictably, the interorganizational violence had a devastating effect on the mutually expressed goal of the MNA and the FLN to end the French occupation.78 Terrorist organizations also undermined their political platforms by targeting each other more than their mutually declared enemy in the violent clashes in Aden between the Liberation of Occupied South Yemen and the National Liberation Front in 1967; in Argentina between Marxist terrorist organizations in the late 1970s; and in the Gaza Strip between Palestinian groups ghting for a common cause during the rst intifada.79 In recent years, the same phenomenon has been endemic in terrorist hot spots. In Chechnya, local terrorist organizations have been terrorizing each other despite their joint political platform to establish Chechen independence. And in southern Iraq, Shiite militias with a shared ideological stance have been mainly blowing each other up, to the obvious benet of the Sunnis.80 That terthe LTTE, in Tore Bjorgo, ed., Root Causes of Terrorism: Myths, Reality, and Ways Forward (New York: Routledge, 2006), p. 134. 74. Pape, Dying to Win, pp. 139140. 75. Pedahzur, Suicide Terrorism, pp. 8182. 76. Martha Crenshaw, ed., Terrorism in Context (University Park: Penn State University Press, 1995), p. 483. 77. Benjamin Stora, Algeria, 18302000: A Short History, trans. Jane Marie Todd (Ithaca, N.Y.: Cornell University Press, 2001), p. 59. 78. Crenshaw, Terrorism in Context, p. 484. 79. Pedahzur, Suicide Terrorism, p. 44. See also Jonathan Schanzer, Palestinian Uprisings Compared, Middle East Quarterly, Vol. 9, No. 3 (Summer 2002), pp. 2738; and Peter H. Merkl, Approaches to the Study of Political Violence, in Merkl, Political Violence and Terror, p. 45. 80. Ann Scott Tyson, Attacks in Iraq Continue to Decline, Washington Post, October 31, 2007; and Anthony H. Cordesman, Still Losing? The June 2007 Edition of Measuring Stability in Iraq,

What Terrorists Really Want 185

rorist organizations frequently undercut their stated political agenda is puzzling for the strategic model because terrorists are presumed to be primarily motivated to achieving it. puzzle #7: never-ending terrorism The strategic model assumes that terrorist organizations disband or renounce terrorism when it continuously fails to advance their political platforms.81 To act otherwise, Pape says, is deeply irrational because that would not constitute learning.82 Yet terrorist organizations survive for decades, notwithstanding their political futility.83 The primary explanation for war in the bargaining literature is that rational actors miscalculate the capability and resolve of their opponents.84 Proponents of the strategic model might speculate that terrorist organizations are acting rationally; they simply overestimate the likelihood that attacking civilians will coerce their governments into making policy concessions. The problem with this argument is that informational explanations provide a poor account of protracted conict. James Fearon has shown that after a few years of war, ghters on both sides are expected to develop accurate understandings of their relative capabilities and resolve.85 The idea that terrorists misjudge the coercive effectiveness of their violence therefore does not obtain because terrorist organizations exist for decades despite their political hopelessness. As Loren Lomasky observes, the strategic model impute[s] to terrorists no lesser rationality than that which social analysts routinely ascribe to other actors. . . . Rational agents are not systematically unable to distinguish efcacious from inefcacious activity.86 The longevity of terrorist organizations relative to

Working Paper (Washington, D.C.: Center for Strategic and International Studies, June 20, 2007), http:/ /www.thewashingtonnote.com/archives/IraqStab&Security06-20%5B1%5D.htm. 81. See Bruce Hoffman and Gordon H. McCormick, Terrorism, Signaling, and Suicide Attack, Studies in Conict and Terrorism, Vol. 27, No. 4 (July 2004), p. 252. See also Crenshaw, Theories of Terrorism, p. 16. 82. Pape, Dying to Win, pp. 6364. 83. Abrahms, Why Terrorism Does Not Work, p. 47. See also Martha Crenshaw, How Terrorism Declines, in Clark McCauley, ed., Terrorism Research and Public Policy (London: Frank Cass, 1991), p. 79. 84. Erik Gartzke, War Is in the Error Term, International Organization, Vol. 53, No. 3 (Summer 1999), p. 573. 85. James D. Fearon, Why Do Some Civil Wars Last So Much Longer than Others? Journal of Peace Research, Vol. 41, No. 3 (May 2004), p. 290. See also Branislav L. Slantchev, The Power to Hurt: Costly Conict with Completely Informed States, American Political Science Review, Vol. 97, No. 1 (February 2003), p. 123. 86. Loren E. Lomasky, The Political Signicance of Terrorism, in Frey and Morris, Violence, Terrorism, and Justice, p. 90.

Contending with Terrorism 186

their political accomplishments therefore conicts with the strategic models assumption that terrorism is based on the logic of consequence. Conversely, the strategic model assumes that because terrorists are motivated by relatively stable policy aims, the violence will cease when the organizations stated grievances have been lifted.87 A puzzle for the model then is that terrorist organizations resist disbanding when their political rationales have become moot.88 Papes research demonstrates that contemporary guerrilla campaigns have coerced major policy concessions from target countries; yet none of the organizations that also use terrorism have disbanded.89 Hezbollah, for example, remains an operational terrorist group, despite the fact that its guerrilla attacks on the Israel Defense Forces achieved the stated goal of liberating southern Lebanon in May 2000. When their political rationale is losing relevance, terrorist organizations commonly invent one. Klaus Wasmunds case study of the RAF shows, for example, that the German terrorists were aggravated when the Vietnam War ended because they suddenly faced a dilemma of nding a suitable revolutionary subject. Instead of abandoning the armed struggle, the RAF turned overnight into a militant advocate of the Palestinian cause.90 Similarly, the 9/11 commission explains that upon discovering in April 1988 that the Soviets were planning to withdraw from Afghanistan, the mujahideen made the collective decision to remain intact while they hunted for a new political cause.91 In this way, terrorist organizations contrive a new political raison dtre, belying the assumption that terrorists are motivated by relatively stable policy preferences reected in their organizations political platforms.

What Terrorists Really Want

These seven puzzles challenge the strategic model with disconrming evidence of its core assumptions that terrorists (1) are motivated by relatively consistent and stable political goals issued by the terrorist organization; (2) weigh the expected political costs and benets of the most obvious options; and (3) opt for a strategy of terrorism because of its expected political effectiveness
87. See Pape, Dying to Win, p. 94. 88. See Crenshaw, How Terrorism Ends, p. 80. See also Martha Crenshaw, The Causes of Terrorism, Comparative Politics, Vol. 13, No. 4 (July 1981), p. 397. 89. Pape, Dying to Win, p. 109. 90. Klaus Wasmund, The Political Socialization of West German Terrorists, in Merkl, Political Violence and Terror, p. 221. See also Hoffman, Inside Terrorism, p. 179. 91. The 9/11 Commission Report: Final Report of the National Commission on Terrorist Attacks upon the United States (New York: W.W. Norton, 2004), p. 56.

What Terrorists Really Want 187

Figure 1. The Empirical Weakness of the Strategic Model

NOTE: The strategic models assumptions are obviously interrelated; there is no implication

that each puzzle violates only one of them.

(see Figure 1). The puzzles suggest that the strategic model is awed in one of two ways: either terrorists are irrational people who minimize their utility or the model misspecies their incentive structure. Psychiatric studies reveal that terrorists are not irrational.92 This implies that the foremost objective of terrorists may not be to achieve their organizations political platform. The tremendous number and variation of terrorist organizations in the world preclude a single causal explanation for terrorism that obtains in every situation. The equinality of terrorism ensures that any causal explanation is necessarily probabilistic, not deterministic.93 This section demonstrates, however, that an alternative incentive structure has superior explanatory power. There is comparatively strong theoretical and empirical evidence that people become terrorists not to achieve their organizations declared political agenda, but to develop strong affective ties with other terrorist members. In other words, the preponderance of evidence is that people participate in terrorist organizations for the social solidarity, not for their political return. Organization theories are potentially useful for explaining terrorist motives because nearly all terrorist attacks are perpetrated by members of terrorist organizations.94 The natural systems model, a leading approach in organization theory, posits that people participate in organizations not to achieve their

92. Richardson, What Terrorists Want, p. 14. See also Marc Sageman, Understanding Terror Networks (Philadelphia: University of Pennsylvania Press, 2004), p. 81. 93. Karen Rasler, Review Symposium: Understanding Suicide Terror, Perspectives on Politics, Vol. 5, No. 1 (February 2007), p. 118. See also Krueger, What Makes a Terrorist, p. x. 94. See Christopher Hewitt, Understanding Terrorism in America: From the Klan to al-Qaida (London: Routledge, 2003), p. 57. See also Krueger, What Makes a Terrorist, p. 71.

Contending with Terrorism 188

ofcial goals, but to experience social solidarity with other members. After briey describing the natural systems model, I demonstrate its applicability to understanding terrorists motives.95 the natural systems model Organization theory has been dominated by two dueling models since the 1930s: the classical model and the natural systems model, which counts many more adherents.96 Classical organization theorists such as Max Weber and Frederick Taylor conceived of the organization as a set of arrangements oriented toward maximizing output. In the classical model, members participate in an organization solely to achieve its stated goals. According to this view, the effectiveness and rationality of an organization therefore depend entirely on the degree to which its actions advance its ofcial aims.97 In assuming that terrorists are motivated to achieving their organizations stated political goals, the strategic model is predicated on the antiquated views of the classical model, which faced almost immediate opposition. Chester Barnard, the father of the natural systems model, exposed the classical fallacy of equating the ofcial goals of an organization with the goals of its members. Barnard demonstrated that most individuals engage in a costbenet analysis of whether to participate in an organization based on its personal inducements, which have little if any connection to the organizations stated goals. For Barnard, the most important incentive is what he called the condition of communion, the sense of solidarity from participating in a social collectivity.98 The natural systems model stresses that there is often a disconnect between the ofcial goals of an organization and the latent social goals governing its behavior. The loose coupling of organizational practices with ofcial goals implies that the failure to achieve them may be entirely satisfactory from the perspective of its members.99 In fact, the model emphasizes that organizations
95. In the select cases where terrorism scholars have explicitly employed a variant of organization theory, they invariably present it as a secondary lens to complementnot contestthe strategic model. See, for example, Bloom, Dying to Kill, p. 3; Richardson, What Terrorists Want, p. 79; and Pedahzur, Suicide Terrorism, pp. 11, 25. 96. See Charles Perrow, Complex Organizations: A Critical Essay (Glenview, Ill.: Scott, Foresman, 1972), p. 75; and Pfeffer, Organizations and Organization Theory, p. 72. 97. Gibson Burrell and Gareth Morgan, Sociological Paradigms and Organisational Analysis (London: Heinemann, 1988), p. 149; Paul S. Goodman and Johannes M. Pennings, eds., New Perspectives on Organizational Effectiveness (London: Jossey-Bass, 1981), p. 3; and W. Richard Scott, Effectiveness of Organizational Effectiveness Studies, in Goodman and Pennings, New Perspectives on Organizational Effectiveness, p. 75. 98. Chester I. Barnard, The Functions of the Executive (Cambridge, Mass.: Harvard University Press, 1938), pp. 17, 85, 145146, 148. 99. Ibid., pp. 145146, 148. See also W. Richard Scott, Organizations: Rational, Natural, and Open Sys-

What Terrorists Really Want 189

will act to perpetuate their existenceeven when doing so undermines their ofcial goalswhenever members attach utmost importance to the social benets of the organization.100 If people participate in terrorist organizations primarily to achieve social solidarity, one would therefore expect to nd (1) evidence at the individual level that people are mainly attracted to terrorist organizations not to achieve their ofcial political platforms, but to develop strong affective ties with other terrorist members; and (2) evidence at the organizational level that terrorist groups consistently engage in actions to preserve the social unit, even when these impede their ofcial political agendas. There is compelling evidence at both levels of analysis. terrorists as social solidarity seekers Empirical evidence is accumulating in terrorism studies and political psychology that individuals participate in terrorist organizations not to achieve their political platforms, but to develop strong affective ties with fellow terrorists. First, psychologist Jeff Victoroff has concluded in a prcis of the terrorism literature that the claim that no individual factors identify those at risk for becoming terrorists is based on completely inadequate research.101 Terrorist organizations appeal disproportionately to certain psychological types of people, namely, the socially alienated. Melvin Seeman denes alienation broadly as the feeling of loneliness, rejection, or exclusion from valued relationships, groups, or societies.102 Demographic data show that the vast majority of terrorist organizations are composed of unmarried young men or widowed women who were not gainfully employed prior to joining them.103 Other demographic

tems, 3d ed. (Englewood Cliffs, N.J.: Prentice Hall, 1992), pp. 5, 51; and Walter W. Powell, Expanding the Scope of Institutional Analysis, in Walter W. Powell and Paul J. DiMaggio, eds., The New Institutionalism in Organizational Analysis (Chicago: University of Chicago Press, 1991), p. 183. 100. See David M. Austin, The Political Economy of Social Benet Organizations: Redistributive Services and Merit Goods, in Herman D. Stein, ed., Organization and the Human Services: CrossDisciplinary Reections (Philadelphia: Temple University Press, 1981), p. 170. See also David L. Clark, Emerging Paradigms: In Organizational Theory and Research, in Yvonna S. Lincoln, ed., Organizational Theory and Inquiry: The Paradigm Revolution (Beverly Hills, Calif.: Sage, 1985), p. 59. 101. See Jeff Victoroff, The Mind of the Terrorist: A Review and Critique of Psychological Approaches, Journal of Conict Resolution, Vol. 49, No. 1 (February 2005), p. 34. 102. Melvin Seeman, Alienation and Engagement, in Angus Campbell and Philip E. Converse, eds., The Human Meaning of Social Change (New York: Russell Sage, 1972), pp. 472473. 103. For research on the prevalence of these demographic characteristics in a wide variety of terrorist organizations, see Pedahzur, Suicide Terrorism, pp. 151152; Alex P. Schmid, Why Terrorism? Root Causes, Some Empirical Findings, and the Case of 9/11, presentation to the Council of Europe, Strasbourg, France, April 2627, 2007, p. 12; Ariel Merari, Social, Organizational, and Psychological Factors in Suicide Terrorism, in Bjorgo, Root Causes of Terrorism, p. 75; Sageman, Understanding Terror Networks, p. 95; Rex A. Hudson, The Sociology and Psychology of Terrorism (Washington, D.C.: Federal Research Division, Library of Congress, September 1999); Charles

Contending with Terrorism 190

studies show that terrorist organizations are frequent repositories for people undergoing dislocation from their native homeland who are therefore detached from family, friends, and the host society they are attempting to join. Marc Sagemans study of 172 global Sala jihadists demonstrates that these risk factors are particularly prevalent among the crucial case of al-Qaida members, 80 percent of whom are cultural outcasts living at the margins of society as unassimilated rst- or second-generation immigrants in nonMuslim countries.104 Analysts who study al-Qaida are increasingly nding that European Muslims are unassimilated in their host countries and represent a core constituency of al-Qaida, whereas Muslims in the United States are comparatively assimilated and detached from the al-Qaida network.105 Variation on the independent variable of alienation or social isolation can therefore explain variation on the dependent variable for joining al-Qaida. The high correlation of what Albert Bandura calls conducive social conditions among the hundreds of terrorist members for whom data exist is consistent with my argument that most individuals participate in terrorist organizations to achieve social solidarity.106 Second, members from a wide variety of terrorist groupsincluding ETA, the IRA, the Italian Communist Party, the RAF, the Red Brigades, Turkish terrorist organizations, and the Weather Undergroundsay that they joined these armed struggles not because of their personal attachment to their political or ideological agendas, but to maintain or develop social relations with other terrorist members.107 These are not the statements of a small number of terrorists; in the Turkish sample, for instance, the 1,100 terrorists interviewed
A. Russell and Bowman A. Miller, Prole of a Terrorist, in John D. Elliott and Leslie K. Gibson, eds., Contemporary Terrorism: Selected Readings (Gaithersburg, Md.: International Association of Chiefs of Police, 1978), pp. 8195; and Stern, The Protean Enemy, p. 6. In Sagemans sample, many of the jihadists are married, but most researchers believe that the jihadist population is overwhelmingly single. 104. Sageman, Understanding Terror Networks, p. 92. See also Olivier Roy, Terrorism and Deculturation, in Louise Richardson, ed., The Roots of Terrorism (New York: Routledge, 2006), pp. 159160; Stern, The Protean Enemy, p. 7; and The 9/11 Commission Report, p. 231. 105. See Roy, Terrorism and Deculturation, p. 166. 106. Albert Bandura, Psychological Mechanisms of Aggression, in Mario von Cranach, ed., Human Ethology: Claims and Limits of a New Discipline (Cambridge: Cambridge University Press, 1979). Proponents of the strategic model reject the idea that individuals turn to terrorism because they are socially alienated; their evidence, ironically, is that people who join a terrorist organization are sometimes embraced, even celebrated, by their surrounding communities. See, for example, Pape, Dying to Win, chap. 10. 107. See, for example, Schmid, Why Terrorism? p. 11; Robert W. White, Political Violence by the Nonaggrieved, in Donatella Della Porta, ed., International Social Movement Research, Vol. 4 (Greenwich, Conn.: Jai Press, 1992), p. 92; Wasmund, The Political Socialization of West German Terrorists, pp. 209212; Hudson, The Sociology and Psychology of Terrorism, p. 37; and Richard G. Braungart and Margaret M. Braungart, From Protest to Terrorism: The Case of the SDS and the Weathermen, in Della Porta, International Social Movement Research, p. 73.

What Terrorists Really Want 191

were ten times more likely to say that they joined the terrorist organization because their friends were members than because of the ideology of the group.108 Third, recent studies on al-Qaida, Fatah, Hamas, Hezbollah, Palestinian Islamic Jihad, and Turkish terrorists have found that the key scope condition for their joining the terrorist organization was having a friend or relative in ita conclusion consistent with prior research on ETA, the IRA, and both Italian and German right-wing and Marxist terrorist groups.109 These ndings are also consistent with a fascinating July 2007 study of Guantanamo Bay detainees. Researchers from West Points Combating Terrorism Center found in their sample of 516 detainees that knowing an al-Qaida member was a signicantly better predictor than believing in the jihad for turning to terrorismeven when a militant denition of jihad was used and other variables were held constant.110 The strategic model cannot explain why the vast majority of politically discontented people do not use terrorism. Yet the requirement of social linkages to the terrorist organization can explain the difference between the large pool of socially isolated people and the relatively small number who become terrorists.111 Fourth, case studies of al-Qaida, Aum Shinrikyo, Hezbollah, the IRA, the RAF, the Weather Underground, and Chechen and Palestinian terrorist groups have concluded that most of the terrorists in these groups participated in the armed struggle to improve their relationships with other terrorists or to reduce their sense of alienation from society, usually both.112 These studies emphasize that social bonds preceded ideological commitment, which was an effect, not a cause, of becoming a terrorist member.113

108. See Schmid, Why Terrorism? p. 11. 109. See, for example, White, Political Violence by the Nonaggrieved, p. 93; Jerrold M. Post, Ehud Sprinzak, and Laurita M. Denny, The Terrorists in Their Own Words: Interviews with 35 Incarcerated Middle Eastern Terrorists, Terrorism and Political Violence, Vol. 15, No. 1 (March 2003), pp. 171184; Sageman, Understanding Terror Networks, p. 92; and Schmid, Why Terrorism? p. 11. 110. Joseph Felter and Jarret Brachman, An Assessment of 516 Combatant Status Review Tribunal Unclassied Summaries, CTC Report (West Point, N.Y.: Combating Terrorism Center, July 15, 2007), pp. 2425, 34. 111. For discussion of the fundamental problem of specicity in terrorism studies, see Sageman, Understanding Terror Networks, chap. 4. See also Weinberg, Global Terrorism, p. 82. 112. See, for example, Hudson, The Sociology and Psychology of Terrorism, p. 148; Sageman, Understanding Terror Networks, p. 95; Merkl, Approaches to the Study of Political Violence, p. 42; Jerrold M. Post, The Socio-cultural Underpinnings of Terrorist Psychology: When Hatred Is Bred in the Bone, in Bjorgo, Root Causes of Terrorism, p. 55; The 9/11 Commission Report, p. 231; and Braungart and Braungart, From Protest to Terrorism, p. 68. 113. The studies on suicide terrorists devote extra attention to this point. One explanation for why suicide terrorists appear relatively apolitical is that organization leaders prefer expending members with no prior connection to the organization or its political cause. See Pedahzur, Suicide Terrorism, pp. 126, 131133, 152154. See also Sageman, Understanding Terror Networks, pp. 93, 135.

Contending with Terrorism 192

Fifth, many terrorist foot soldiers and even their leaders never develop a basic understanding of their organizations political purpose. This nding strengthens the argument that ideological commitment enters through the back door, if at all, of terrorist organizations. In his study of the IRA, for example, Robert White found that nearly half of the terrorists he interviewed were unaware of the discrimination in Northern Ireland against Catholics, despite the salience of this issue in IRA communiqus.114 According to Olivier Roy, Mia Bloom, and a former mujahideen, al-Qaida foot soldiers and their leaders are often ignorant about the basic tenets of Islam, if not bin Ladens political vision.115 Al-Qaida is unexceptional in this regard; Richardsons research shows that a striking and quite surprising aspect of terrorism is that the leaders of very different terrorist movements are unable to explain their basic political purpose.116 When asked to describe the society that their organizations hoped to achieve, the leader of the Shining Path conceded, We have not studied the question sufciently; the founder of the RAF responded, That is not our concern; the leader of the Japanese Red Army replied, We really do not know what it will be like; and the spokesman for the Revolutionary Armed Forces of Colombia acknowledged, I must admit that we have yet to dene this aspect.117 Audrey Cronin has found that leaders of both left-wing and anarchist terrorist groups are also notorious for their inability to articulate a clear vision of their [political] goals.118 That even terrorist leaders frequently cannot explain their organizations political purpose suggests that members have a different motive for participating in them.119 Sixth, terrorist organizations focus their recruitment on the socially isolated, not on people with a demonstrable commitment to their given political cause.

114. White, Political Violence by the Nonaggrieved, p. 83. See also Christopher Dobson and Ronald Payne, The Terrorists: Their Weapons, Leaders, and Tactics (New York: Facts on File, 1981), p. 32. 115. Roy, Terrorism and Deculturation, pp. 159160; Mia Bloom, The Transformation of Suicide Bombing Campaigns: Sectarian Violence and Recruitment in Pakistan, Afghanistan, and Iraq, paper presented at the Terrorist Organizations: Social Science Research on Terrorism conference, University of California, San Diego, May 4, 2007; and Nasiri, Inside the Jihad, p. 279. 116. Richardson, What Terrorists Want, pp. 8586. See also Laqueur, Terrorism, p. 81. 117. Quoted in Richardson, What Terrorists Want, pp. 8687. 118. Cronin, How al-Qaida Ends, p. 23. See also Hoffman, Inside Terrorism, p. 172. 119. That terrorist members often appear uninterested and uninformed regarding their organizations ofcial political agenda is actually not surprising. Terroristsbe it al-Qaida operatives, Red Brigadists, RAF members, the Weathermen, or the Tupamaros of Uruguayhave rarely hailed from the constituencies they claim to represent; many terrorist organizations do not train or indoctrinate their members in any ideology; and terrorists are often walk-ins who have no prior association with the terrorist organization or its political cause before volunteering for an operation. See The 9/11 Commission Report, pp. 228, 232; Dipak K. Gupta, Exploring Roots of Terrorism, in Bjorgo, Root Causes of Terrorism, p. 19; Pedahzur, Suicide Terrorism, pp. 132133; Crenshaw, Terrorism in Context, p. 15; and Crenshaw, How Terrorists Think, p. 73.

What Terrorists Really Want 193

Pedahzurs research, for example, shows that Hezbollah, the PKK, and Chechen and Palestinian groups recruit young, unemployed men who have never found their place in the community, not fervent nationalists committed to political change.120 Similarly, Peter Merkl shows that Marxist terrorist groups have historically recruited unemployed youth with failed personal lives who lacked political direction.121 Gregory Johnsen likewise suggests that al-Qaida, at least in Yemen, focuses its recruitment not on committed jihadists, but on young and largely directionless socially marginalized Muslim men.122 Seventh, terrorist organizations are particularly attractive outlets for those seeking solidarity. According to political psychologists, terrorist groups are far more tight-knit than other voluntary associations because of the extreme dangers and costs of participation, as well as their tendency to violate societal expectations.123 This observation may account for the fact that even when terrorist organizations fail to achieve their political platforms, committing acts of terrorism tends to generate new recruits, boost membership morale, and otherwise strengthen the social unit.124 Eighth, terrorists seem to prefer participating in terrorist groups and activities most conducive to developing strong affective ties with fellow terrorists. Jacob Shapiro has found that within the al-Qaida network, terrorists prefer operating in more centralized, cohesive clusters of cliques.125 Indeed, since the emergence of modern international terrorism, terrorists have ocked to where other terroristsregardless of their political orientationwere gathered. In the 1970s, thousands of terrorists from dozens of countries and organizations descended on training camps run by the Palestine Liberation Organization; in the 1980s and mid-1990s, the locus of terrorist activity shifted rst to Afghanistan to train with the Afghan mujahideen and then to al-Qaida camps. Based on her interviews with terrorists, Jessica Stern has likened these adventures to an Outward Bound experience for young men seeking challenges, excitement, and above all friendship with fellow terrorists of diverse political
120. Pedahzur, Suicide Terrorism, pp. 137138, 168. 121. Merkl, Political Violence and Terror, p. 42. 122. Gregory Johnsen, Securing Yemens Cooperation in the Second Phase of the War on alQaida, CTC Sentinel, Vol. 1, No. 1 (December 2007), p. 34. 123. Crenshaw, How Terrorists Think, p. 73; and Crenshaw, The Psychology of Political Terrorism, in M.G. Hermann, ed., Political Psychology: Contemporary Problems and Issues (San Francisco, Calif.: Jossey-Bass, 1986), p. 394. See also Shira Fishman, Perceptions of Closeness as a Function of Group Importance, University of Maryland, 2007. 124. See Richardson, What Terrorists Want, p. 301; Bloom, Dying to Kill, pp. 19, 39; and Hoffman, Inside Terrorism, pp. 7375. 125. Jacob N. Shapiro, The Terrorists Challenge: Security, Efciency, Control, paper presented at the Terrorist Organizations: Social Science Research on Terrorism conference.

Contending with Terrorism 194

backgrounds.126 First-hand accounts from these camps conrm that the terrorists often had little idea or preference where they would ght upon completing their training.127 Ninth, there is circumstantial evidence that terrorist organizations collapse when they cease to be perceived as desirable social collectivities worth joining. David Rapoports research demonstrates that throughout history terrorist organizations have disbanded when their members grew old, tired of waging the armed struggle, and their group failed to appeal to the younger generation.128 Cronins research on the decline of terrorist groups also lists generational transition failure as their leading cause of death.129 The tendency for terrorist groups to die out in the course of a human life cycleirrespective of the state of their political grievancessuggests that they appeal to new members primarily for social, not political, reasons. The research landscape is constrained by the limited reliable demographic data on terrorists, representative samples, and controlled studies to rmly establish causation. In the aggregate, however, there is mounting empirical evidence that people may participate in terrorist organizations mainly to achieve social solidarity, not their ofcial political agendas. This incentive structure is testable. The natural systems model posits that when members attach utmost importance to an organizations social benets, the organization will seek to prolong its existence, even when doing so impedes its ofcial goals. This is precisely the way terrorist organizations typically behave. the puzzles revisited The seven puzzles are perplexing for the strategic model because they demonstrate that terrorist organizations behave more as social solidarity maximizers than as political maximizers. The puzzles are easily resolved from the vantage of organization theory. The natural systems model predicts that terrorist organizations will routinely engage in actions to perpetuate and justify their exis-

126. Stern, Terror in the Name of God, p. 5. 127. Nasiri, Inside the Jihad, pp. 151, 178, 217. 128. David C. Rapoport, The Fourth Wave: September 11 in the History of Terrorism, Current History, Vol. 100, No. 650 (December 2001), pp. 419424. See also David C. Rapoport, Generations and Waves: The Keys to Understanding Rebel Terror Movements, paper presented at the Seminar on Global Affairs, Ronald W. Burkle Center for International Affairs, University of California, Los Angeles, November 7, 2003, http:/ /www.international.ucla.edu/cms/les/ David_Rapoport_Waves_of_Terrorism.pdf. 129. Cronins superb study identies seven reasons why terrorist organizations have historically gone out of business. More terrorist organizations suffered from the failure to make the generational transition than from any of the other six reasons explored. It should be noted that Cronin does not purport to categorize the universe of terrorist groups. See Cronin, How al-Qaida Ends, p. 19.

What Terrorists Really Want 195

tence, even when these undermine their ofcial political agendas. True to the model, terrorist organizations (1) prolong their existence by relying on a strategy that hardens target governments from making policy concessions; (2) ensure their continued viability by resisting opportunities to peacefully participate in the democratic process; (3) avoid disbanding by reexively rejecting negotiated settlements that offer signicant policy concessions; (4) guarantee their survival by espousing a litany of protean political goals that can never be fully satised;130 (5) avert organization-threatening reprisals by conducting anonymous attacks, even though they preclude the possibility of coercing policy concessions; (6) annihilate ideologically identical terrorist organizations that compete for members, despite the adverse effect on their stated political cause; and (7) refuse to split up after the armed struggle has proven politically unsuccessful for decades or its political rationale has become moot. None of these common tendencies of terrorist organizations advances their ofcial political agendas, but all of them help to ensure the survival of the social unit. Together, they reveal the operating decision rules of terrorist members. Whereas the strategic model locates the motives of terrorists in the ofcial goals of the terrorist organization, the trade-offs it makes provides direct insight into its members incentive structure. Just as economists measure utility functions through revealed preferences, terrorism scholars need not make comparisons among utilities.131 The seven puzzles discussed above contradict the strategic model because terrorists already make such trade-offs by regularly prioritizing the maintenance of the terrorist organization over the advancement of its ofcial political agenda as predicted by the natural systems model.132

130. The tendency for terrorist organizations to issue protean political demands may dissuade target countries from making policy concessions. See Paul Wilkinson, Security Challenges in the New Reality, paper presented at the 33d IFPA-Fletcher Conference on National Security Strategy and Policy, Washington, D.C., October 16, 2002, http:/ /www.ifpaetcherconference.com/ oldtranscripts/2002/wilkinson.htm. 131. For a similar argument unrelated to terrorist motivations, see Jeffrey Pfeffer, Usefulness of the Concept, in Goodman and Pennings, New Perspectives on Organizational Effectiveness, p. 137. On revealed preferences, see Amartya Sen, Behaviour and the Concept of Preference, in Elster, Rational Choice, pp. 61, 67. 132. In this way, the role of social solidarity is very different in terrorist organizations than in conventional armies. In the military, training is designed to foster in-group cohesion not as the end goal, but as a means to enhance battleeld performance. Unlike terrorist organizations, conventional armies therefore do not regularly sacrice their political goals for the social benet of the ghting unit. On the complementary relationship between small unit cohesion and military performance, see James Grifth, Institutional Motives for Serving in the U.S. Army National Guard, Armed Forces and Society, Vol. 20, No. 10 (May 2007), pp. 129; and Guy L. Siebold, The Essence of Military Group Cohesion, Armed Forces and Society, Vol. 33, No. 2 (January 2007), pp. 286295.

Contending with Terrorism 196

In sum, the seven puzzles for the strategic model challenge the prevailing view that terrorists are rational people who use terrorism for political ends. The preponderance of theoretical and empirical evidence is that people participate in terrorist organizations not to achieve their ofcial political platforms, but to develop strong affective ties with fellow terroristsan incentive structure reected in the trade-offs terrorist organizations typically make to maintain their survival. If terrorists generally attach greater importance to the social benets than to the political benets of using terrorism, then extant counterterrorism strategies require fundamental change.

Counterterrorism Implications
The most common counterterrorism strategies are designed to reduce terrorism by divesting it of its political utility. The predominant strategy is to deter terrorism by decreasing its political utility via a strict no concessions policy.133 Like most heads of state, President George W. Bush believes that terrorism will desist when its practitioners realize that these crimes only hurt their [political] cause.134 Although target governments rarely appease terrorists, there is also a widespread belief in the international community that they can be defused through political accommodation.135 Proponents of this second strategy urge rekindling stalled peace processes, for example, to deny prospective political benets from using terrorism. The third most common counterterrorism strategy is democracy promotion, which is intended to decrease terrorisms utility by empowering citizens to peacefully address their countrys political problems.136 All three strategies have poor track records. As I have shown, terrorist organizations often resist disbanding in the face of consistent political failure, in spite of the ending of their immediate political grievances, and even when presented with peaceful alternatives for political gain. Why does withholding political concessions, granting political concessions, or providing nonviolent political alternatives fail so often to eradicate terrorism? The strategic models premise that terrorists are political maximizers is empirically weak. Strategies to dry up the demand for terrorism by minimizing its political utility are misguided and hence unlikely to work on any sys133. See Martha Crenshaw, Terrorism, Legitimacy, and Power: The Consequences of Political Violence (Middletown, Conn.: Wesleyan University Press, 1983), p. 10. 134. Quoted in Alan M. Dershowitz, Why Terrorism Works: Understanding the Threat, Responding to the Challenge (New Haven, Conn.: Yale University Press, 2002), p. 17. 135. Laqueur, Terrorism, p. 5. 136. See George W. Bush, President Discusses War on Terror, National Defense University, Washington, D.C., March 8, 2005, http:/ /www.whitehouse.gov/news/releases/2005/03/ 20050308-3.html.

What Terrorists Really Want 197

tematic basis. The evidence is stronger that terrorists tend to think and act more as social solidarity maximizers, which requires a different counterterrorism approach. Both supply-side and demand-side counterterrorism strategies must be informed by the terrorists incentive structure. Supply-side strategies can help law enforcement identify potential terrorists, unravel covert networks, and even thwart terrorist attacks by exploiting the knowledge that people tend to participate in terrorist groups to develop strong affective ties with fellow terrorists. There is no single terrorist personality, but certain communities are prone to terrorism. Law enforcement must pay greater attention to the socially marginalized than to the politically downtrodden. This includes diaspora communities in Western countries that host large unassimilated, dislocated populations such as the Maghrebin in France; single, unemployed, Islamist men residing in comparatively secular Muslim countries such as in Pakistan; restive, youthful populations that feel estranged from the state such as in Saudi Arabia; and prison populations, which, by denition, are home to the socially isolated and dislocated. These are impossibly large groups of people to monitor. Law enforcement can tighten the noose considerably by exploiting the fact that terrorist groups are composed of networks of friends and family members, and that knowing one of them is the key scope condition for entry into the group. Governments should utilize this knowledge to aggressively boost funding of social network analysis (SNA) research. SNA is a mathematical method for mapping and studying relationships between people, with untapped counterterrorism potential. The basic idea is to trace the social relations or links emanating from known terrorists or suspects, and then connect the dots between these nodes of people, to estimate the probability of their involvement in the terrorist network. People who email, talk on the phone, or intentionally meet with terrorists or their close friends are statistically more likely to be complicit. In this way, SNA can help law enforcement identify and then surveil the inner circle. Because acquaintances can also play a critical role in the network, greater data-mining power and accuracy need to be developed to expose these weak ties without undue infringements on civil liberties.137 Demand-side strategies should focus on divesting terrorisms social utility, in two ways. First, it is vital to drive a wedge between organization members. Since the advent of modern terrorism in the late 1960s, the sole counterterrorism strategy that was a clear-cut success attacked the social bonds of the

137. For a useful primer on SNA, see Patrick Radden Keefe, Can Network Theory Thwart Terrorists? New York Times, March 12, 2006. See also Sageman, Understanding Terror Networks, pp. 163, 169, 178.

Contending with Terrorism 198

terrorist organization, not its utility as a political instrument. By commuting prison sentences in the early 1980s in exchange for actionable intelligence against their fellow Brigatisti, the Italian government inltrated the Red Brigades, bred mistrust and resentment among the members, and quickly rolled up the organization.138 Similar deals should be cut with al-Qaida in cases where detainees prior involvement in terrorism and their likelihood of rejoining the underground are minor. Greater investment in developing and seeding double agents will also go a long way toward weakening the social ties undergirding terrorist organizations and cells around the world. Second, counterterrorism strategies must reduce the demand for at-risk populations to turn to terrorist organizations in the rst place. To lessen Muslims sense of alienation from democratic societies, these societies must improve their records of cracking down on bigotry, supporting hate-crime legislation, and most crucially, encouraging moderate places of worshipan important alternative for dislocated youth to develop strong affective ties with politically moderate peers and mentors. In authoritarian countries, an abrupt transition to democracy risks empowering extremists.139 These regimes must, however, permit the development of civil society to provide opportunities for the socially disenfranchised to bond in peaceful voluntary associations. Counterterrorism operations must also redouble their efforts to minimize collateral damage, which invariably creates dislocation, social isolation, and calls for revenge. Such policies will help reduce the incentive and therefore incidence of terrorism by diminishing its social benets, which are what its practitioners apparently value most.

138. See Bruce Hoffman, Foreword, in Cindy C. Combs, Twenty-rst Century Terrorism (New York: Prentice Hall, 1996), pp. v18. 139. F. Gregory Gause III, Can Democracy Stop Terrorism? Foreign Affairs, Vol. 84, No. 5 (September/October 2005), pp. 6276.

Correspondence: What Makes Terrorists Tick

Correspondence

Erica Chenoweth, Nicholas Miller, and What Makes Terrorists Tick Elizabeth McClellan Hillel Frisch Paul Staniland Max Abrahms

To the Editors (Erica Chenoweth, Nicholas Miller, and Elizabeth McClellan write):
Max Abrahmss article What Terrorists Really Want: Terrorist Motives and Counterterrorism Strategy is a welcome critique of the many points taken for granted by rational choice interpretations of terrorist group behavior.1 His systematic review of the observable implications of rational choice perspectives on terrorism reveals some of the important shortfalls in the current literature. Abrahms overreaches, however, in rejecting strategic models of terrorism without providing ample empirical evidence or qualications to his claims. Abrahms presents seven puzzling tendencies of terrorist organizations as anomalies for the strategic model. We argue, however, that a strategic perspective can account for these anomalous behaviors when one examines the groups internal dynamicsparticularly the relationship between the groups leadership and its constituentswhich may require scholars to consider this level of analysis to explain terrorist group behavior. We consider each of Abrahmss puzzles in turn.
Erica Chenoweth is Assistant Professor of Government at Wesleyan University. Nicholas Miller is an undergraduate Government major at Wesleyan. Elizabeth McClellan is an undergraduate Comparative Government and Studio Art double major at Wesleyan. For a longer, more detailed version of this letter, see Erica Chenoweths website, http://wesfiles.wesleyan.edu/home/echenoweth/web/home.htm. Hillel Frisch is Associate Professor in the Departments of Political Studies and Middle East History at BarIlan University. He thanks Stuart Cohen, Efraim Inbar, and Gordon McCormick for their helpful comments. Paul Staniland is a Ph.D. candidate in the Department of Political Science and member of the Security Studies Program at the Massachusetts Institute of Technology, and a Research Fellow at the Belfer Center for Science and International Affairs in the Harvard Kennedy School at Harvard University. The author would like to thank Brendan Green, Vipin Narang, Joshua Rovner, and Cyrus Samii for helpful comments. Max Abrahms is a Social Science Predoctoral Fellow in the Center for International Security and Cooperation at Stanford University, and a Ph.D. candidate in the Department of Political Science at the University of California, Los Angeles. He would like to thank Mia Bloom, Martha Crenshaw, Matthew Gottfried, Peter Krause, Jessica Stern, Paul Stockton, and Robbie Totten for their helpful comments. He is especially grateful to Richard English for sharing his expertise over the years on the Irish Republican Army. 1. Max Abrahms, What Terrorists Really Want: Terrorist Motives and Counterterrorism Strategy, International Security, Vol. 32, No. 4 (Spring 2008), pp. 78105. Further references to this article appear parenthetically in the text.
International Security, Vol. 33, No. 4 (Spring 2009), pp. 180202 2009 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

199

Contending with Terrorism 200

puzzle #1: coercive ineffectiveness Abrahmss argument that terrorism is ineffective, and therefore not rational, has two main aws. First, by relying on the State Departments current list of foreign terrorist organizations, Abrahms misses the large number of terrorist groups that have ceased to exist and inherently selects groups that are still operating precisely because they have not yet achieved their goals. Seth Jones and Martin Libicki have found that since 1968, 268 terrorist groups have disbanded and an additional 136 have splintered into other violent groups.2 Indeed, Jones and Libicki found that 27 groups disbanded after achieving their goals, for a success rate of 10 percent.3 Although not an overly impressive rate, there is a critical difference between never succeeding and succeeding one out of ten times when an organization surveys its strategic options. Second, Abrahms does not explore alternative measures and perceptions of success upon which terrorist organizations may rely. There may not be a clear causal link between the terrorist campaign and the realization of the organizations goals, but if leaders perceive that terrorism can be successful relative to other alternatives, then choosing terrorism is rational. This is an issue that Robert Pape addresses in his discussion of suicide terrorism, noting that in this search for an effective strategy, coercers assessments are likely to be largely a function of estimates of the success of past efforts; for suicide terrorists, this means assessments of whether past suicide campaigns produced signicant concessions.4 Because it is often ambiguous whether a governments decisions are driven by terrorism or unrelated factors, it is rational to consider terrorism successful as long as this interpretation is shared by a signicant portion of other observers.5 For instance, Abrahms was probably right that the 2004 Madrid bombings had a questionable effect on the results of the Spanish elections and Spains subsequent withdrawal of its military forces from Iraq. But many observers have interpreted these outcomes as examples of terrorist successan opinion likely shared by many terrorists and leaders who are contemplating adopting a terrorist strategy.6 Many terrorist leaders cite prominent examples of perceived terrorist successes in explaining their tactics,

2. Seth G. Jones and Martin C. Libicki, How Terrorist Groups End: Lessons for Countering al Qaida (Washington, D.C.: RAND, 2008), p. 19. 3. Ibid. 4. Robert A. Pape, The Strategic Logic of Suicide Terrorism, American Political Science Review, Vol. 97, No. 3 (August 2003), pp. 343361, at p. 350. 5. Ibid., p. 351. 6. Max Abrahms, William Rose, and Rysia Murphy, Correspondence: Does Terrorism Ever Work? The 2004 Madrid Train Bombings, International Security, Vol. 32, No. 1 (Spring 2007), pp. 185192. p. 189.

Correspondence: What Makes Terrorists Tick 201

whether they refer to Irgun, Hezbollah, the African National Congress (ANC), the Tamil Tigers, or al-Qaida.7 puzzle #2: terrorism as the rst resort Abrahms argues that terrorist groups do not embrace terrorism as a last resort and seldom elect to abandon the armed struggle to become nonviolent political parties, and that this undermines the strategic model (p. 84). But the strategic model does not require groups to adopt a number of other alternatives before adopting terrorism. They only consider other alternatives and decide that terrorism is the optimal strategy. This was the case with the ANC, which considered four types of violent activities before judging that open revolution was inconceivable.8 That terrorists assessments are rarely correct or justiable does not mean that they have made a priori irrational choices. Furthermore, Abrahmss assertion that terrorist groups rarely transform into nonviolent political parties is empirically weak. Jones and Libicki found that the most common way for terrorist groups to end was by joining the political processeither through a peace settlement with the government and abandonment of violence or through civic action in the absence of an explicit agreement.9 Of 268 groups that have ended since 1968 without splintering, 114 (or 43 percent) did so by entering nonviolent politics in one way or another.10 Examples include the Irish Republican Army (IRA), the Farabundo Mart National Liberation Front in El Salvador, and the Mozambican National Resistance.11 Even if such empirics were absent, Abrahmss claim that terrorist groups toil alongside peaceful parties and sabotage open elections that would have yielded major political gains for the group is not a disconrmation of the strategic model (p. 85). Just because political parties with the same general ideology gain power does not suggest that extremist groups will be satised. Indeed, slight variations in political agendas may give the groups sufcient reason to remain dubious about the parties intentions and capabilities to address group grievances. For instance, Abrahms uses the Italian left-wing
7. For a collection of original terrorist documents, see Walter Laqueur, ed., Voices of Terror: Manifestos, Writings, and Manuals of al-Qaeda, Hamas, and Other Terrorists from around the World throughout the Ages (New York: Sourcebooks, 2005). See, for instance, excerpts of Ayman al-Zawahiris Knights under the Prophets Banner, pp. 426433. 8. Nelson Mandela, Long Walk to Freedom: The Autobiography of Nelson Mandela (Boston: Little, Brown, 1994), p. 240. 9. Jones and Libicki, How Terrorist Groups End, pp. 2022. 10. Ibid., p. 19. 11. Ibid., pp. 2223.

Contending with Terrorism 202

terrorists of the 1970s as an example of this phenomenon. These terrorists, however, viewed the Italian Communist Party as having sold out to the Christian Democrats to form a governing coalition, especially when grievances about labor and wages were not immediately satised.12 Once these grievances were addressed, members of the groups became introspective.13 Effective counterterrorism policies subsequently brought the terrorist groups to their demise. puzzle #3: reexively uncompromising terrorists There are at least three problems with Abrahmss claim that terrorists unwillingness to compromise with governments undermines the strategic model. First, some terrorist groups have attempted to compromise by declaring unilateral cease-res, as the Basque terrorist group ETA did in 2006. As mentioned above, since 1968, 43 percent of terrorist groups have ended by entering the political process, often due to negotiations.14 Second, failing to compromise is not in itself evidence of irrationality. Terrorist groups may have extreme aims that preclude compromise. Jones and Libickis ndings support this claim, as groups with limited aims were far more likely to renounce violence and enter politics.15 Furthermore, although Abrahms claims that many terrorist organizations profess surprisingly moderate political positions such as separatism and self-determination, he does not acknowledge that these moderate political positions often seem extreme to the governments that terrorists are targeting (p. 86). Historical instances of states willingly surrendering control over territory are rare indeed, especially in the absence of a protracted struggle. Even among terrorists with limited goals, there are strategic reasons to be uncompromising and to attack at seemingly counterproductive times. These reasons include outbidding among groups,16 suspicions about the governments credibility,17 and the number of combatant groupsin nearly every case of the uncompromising terrorists mentioned in this section, other rival

12. Donatella della Porta, Left-wing Terrorism in Italy, in Martha Crenshaw, ed., Terrorism in Context (University Park: Penn State University Press), pp. 105159. 13. Erica Chenoweth, The Inadvertent Effects of Democracy on Terrorist Group Proliferation, Ph.D. dissertation, University of Colorado, Boulder, 2007, pp. 175245. 14. Jones and Libicki, How Terrorist Groups End, p. 18. 15. Ibid., p. 20. 16. Mia Bloom, Dying to Kill: The Allure of Suicide Terror (New York: Columbia University Press, 2005). 17. Dan Reiter, Exploring the Bargaining Model of War, Perspectives on Politics, Vol. 1, No. 1 (March 2003), pp. 2743.

Correspondence: What Makes Terrorists Tick 203

terrorist groups existed. Terrorist groups may attempt to hold peace processes hostage (e.g., Real IRA Omagh bombing in 1998) because they are dubious of their competitors abilities to extract sufcient concessions rather than selling out, or because holding out for the end of the peace talks may enable the group to extract more concessions.18 Third, Abrahms does not address the role of the government in negotiations with terrorist groups. In addition to all of the standard difculties with negotiations (e.g., cheating, indivisibility, and domestic pressures), negotiating with terrorists can pose further challenges to the state. In fact, Peter Neumann notes that most governments are reexively uncompromising toward terrorists.19 As Abrahms argues previously, Target countries view the negative consequences of terrorist attacks on their societies and political systems as evidence that the terrorists want them destroyed. Target countries are understandably skeptical that making concessions will placate terrorist groups believed to be motivated by these maximalist objectives.20 If he is right, governments will hesitate to compromise with terrorists who target civilians because they infer from their tactics that their extreme goals preclude a settlement. This indeed seems plausible, but it demands qualication from Abrahmss later claim of reexively uncompromising terrorists (pp. 8587). If governments are generally unwilling to make concessions to terrorist groups, then it is inaccurate to blame terrorists solely for failing to reach compromises with the government. puzzle #4: protean political platforms Abrahms argues that the tendency of terrorist groups to change their goals over time refutes strategic models of terrorism. Changing stated platforms, however, is not a disconrmation of the strategic model, especially when one relaxes the assumption that terrorist groups are unitary actors. Terrorist groups may make up political goals to survive because, at the group level of analysis, survival may be a stable preference over time. According to Daniel Byman, Over 90 percent of terrorist groups do not survive their rst year,21 while those that do survive have exible organizational structures and are able to remain discrete. Terrorist organizations that do not place a high
18. David E. Cunningham, Veto Players and Civil War Duration, American Journal of Political Science, Vol. 50, No. 4 (September 2006), pp. 875892. 19. Peter Neumann, Negotiating with Terrorists, Foreign Affairs, Vol. 86, No. 1 (January/February 2007), pp. 128138. 20. Max Abrahms, Why Terrorism Does Not Work, International Security, Vol. 31, No. 2 (Fall 2006), p. 59. 21. Daniel Byman, Deadly Connections: States That Sponsor Terrorism (Cambridge: Cambridge University Press, 2005), p. 63.

Contending with Terrorism 204

priority on group maintenance are likely to dissolve, so leaders may sacrice some stated aims for the benets of exibility and continuity. For instance, because joining a terrorist group is often an irreversible decision, the groups leadership may invent new causes to appease their most devoted members, donors, and sponsors (pp. 8689). Indeed, remaining united may be the only alternative for some terrorists given the high costs of exit. puzzle #5: anonymous attacks Abrahms argues that terrorists often do not claim responsibility for their attacks, and that this strategy is illogical because it pre[cludes] even the possibility of successful coercion (p. 90). There are numerous strategic explanations, however, for this behavior. First, one of the major goals of terrorist violenceto provoke a disproportionate government reaction to the violenceis often best served through anonymous attacks that provoke the government into cracking down on the population indiscriminately.22 If this is part of the terrorist groups intentions, then claiming attacks is unnecessary and may even be counterproductive. A strategy of spoiling may also be effectively carried out anonymously.23 If a country seeks to bargain with moderates on terms that are unacceptable to terrorist organizations, the organizations may have strategic interests in disrupting the talks. Anonymous attacks may even be more effective at disrupting negotiations because they create suspicion that the moderates themselves may support terrorism. Anonymity may also protect terrorist organizations from loss of popular support. An irate public may be more likely to back erce government retaliation through visible counterterrorist responses.24 Mia Bloom, for instance, has documented a number of unclaimed attacks that groups avoid claiming because of negative public reactions.25 On the other hand, highly visible public actions can deliver recruits and resources. Correspondingly, sometimes multiple groups claim an attack after their constituents respond favorably to

22. Andrew H. Kydd and Barbara F. Walter, The Strategies of Terrorism, International Security, Vol. 31, No. 1 (Summer 2006), pp. 4979. 23. Ibid. 24. James D. Fearon, Domestic Political Audiences and the Escalation of International Disputes, American Political Science Review, Vol. 88, No. 3 (September 1994), pp. 577592; and Max Abrahms, Why Democracies Make Superior Counterterrorists, Security Studies, Vol. 16, No. 2 (AprilJune 2007), pp. 223253. 25. Bloom, Dying to Kill.

Correspondence: What Makes Terrorists Tick 205

it.26 Because of the uncertainty of the public response, anonymity can help terrorist groups survive unpopular miscalculations. puzzle #6: terrorist fratricide Abrahms claims that the strategic model cannot explain why organizations purporting to ght for a common cause frequently attack each other more than their mutually declared enemy (p. 90). On the contrary, liquidating groups with similar ideologies is rational given that the groups are competing for the same market share of recruits, sponsorship, sympathy, and attention.27 The Tamil Tigers viciously pursued rivals in the early years and have evolved to be the hegemonic violent nationalist movement in Tamil Eelam. Consequently, the Tamil Tigers have become synonymous with the Tamil nationalist movement, so that governments are unable to address policy issues regarding Tamil Eelam without considering the Liberation Tigers of Tamil Eelam. The National Liberation Front in Algeria also dominated its competition and eventually negotiated the terms of the French departure.28 From a strategic standpoint, establishing dominance enables a group to control the terms of negotiation if it reaches negotiations with the government. Regardless, terrorists spend less time attacking their rivals than they do their primary opponents. According to Christopher Hewitt, fratricidal attacks in the United States account for only 17 percent of terrorist activities, whereas terrorist targets are linked to stated aims more than 65 percent of the time.29 In fact, terrorist fratricide would be more puzzling from the perspective of natural systems theory, where terrorist recruits seek camaraderie and companionship through social solidarity within a violent group. Attacking colleagues with similar political platforms inside or outside the group may not necessarily serve this goal either. Such behavior can be attributed to strategic calculations of relative gain and loss and, therefore, conforms more to strategic models than Abrahms suggests. puzzle #7: never-ending terrorism Abrahms argues that the persistence of terrorist groups after their stated aims have become obsolete challenges the strategic model. Importantly, however, re-

26. Ibid. 27. Ibid. 28. Martha Crenshaw, The Effectiveness of Terrorism in the Algerian War, in Crenshaw, Terrorism in Context, pp. 473513, at pp. 511513. 29. Christopher Hewitt, Understanding Terrorism in America: From the Klan to al Qaeda (New York: Routledge, 2003), pp. 6667.

Contending with Terrorism 206

searchers at RAND found that 404 terrorist groups have ended since 1968; 136 of these splintered into other terrorist groups, and 268 have renounced violence.30 conclusion Although his article is an important, insightful, and creative critique, Abrahms goes too far in rejecting strategic models of terrorism. The puzzles that he presents are not puzzles for the strategic model. Instead, these anomalies are opportunities to delineate the levels of analysis at which rational behavior is failing. What is rational for a terrorist recruit may not be rational for the groups leadership, and vice versa. Abrahmss alternative to the strategic modelnatural systems theoryconcerns only the individuals incentives for joining a group. Natural systems theory does not describe how terrorist leaders organize, apply ideology, and select strategy, but rather deals exclusively with foot soldiers. Abrahms may be right that the strategic model does not adequately address the interests of the rank-and-le (p. 95). But it may be necessary to apply the strategic model to the internal dynamics of the groups by disaggregating foot soldiers from strategists rather than rejecting the model outright. Abrahmss main contribution may be to explain the types of individuals who are attracted to terrorist groupsin other words, how the preferences of terrorist recruits form. His article reinforces this need to reconsider the unitary actor assumption in rational choice models, which may provide many fruitful avenues for future research.31 Indeed, strategic models and natural systems theory can be complementary, particularly if they unify the different levels of analysis discussed above. But at the level of the groups strategy, the natural systems perspective does not outperform the strategic model in generating a unique set of predictions that enhance scholars theoretical understandings or policy choices. Erica Chenoweth Middletown, Connecticut Nicholas Miller Middletown, Connecticut Elizabeth McClellan Middletown, Connecticut
30. Jones and Libicki, How Terrorist Groups End, p. 19. 31. For one such application, see Wendy Pearlman, Fragmentation and Violence: Internal Inuences on Tactics in the Case of the Palestinian National Movement, 19202006, Ph.D. dissertation, Harvard University, 2007.

Correspondence: What Makes Terrorists Tick 207

To the Editors (Hillel Frisch writes):

Max Abrahmss article What Terrorists Really Want: Terrorist Motives and Counterterrorism Strategy constitutes a well-written, thought-provoking critique of the assumption that terrorist groups are strategic in their pursuit of political goals and hence amenable to political concessions.1 Nevertheless, the evidence Abrahms presents to support his argument, his alternative explanation for why terrorists pursue terrorism, and the policy implications he draws from his analysis are all open to question. Abrahms claims that members of terrorist organizations are motivated primarily by the quest for social solidarity rather than by the desire to achieve clearly dened political goals. As Gordon McCormick notes, however, the truth is that there are movements that are either strategic or motivated by social solidarity, or both: Terrorist groups, in this respectas in many others are not created equal. Nor is their behavior necessarily consistent over the course of their operational life. Some (otherwise distinctive) groups, such as the contemporary Irish Republican Army (IRA), al-Qaeda, Hamas, and the Tamil Tigers, have largely managed to subordinate their actions to their political objectives. . . . Others, such as the late November 17, the Popular Forces of 25 April, the Justice Commandos of the Armenian Genocide, or any number of todays amateur terrorists, have effectively subordinated their political objectives to their need to act.2 Albert Camus, in his intriguing play on a Russian terrorist cell in 1905, The Just Assassins, effectively demonstrates this duality down to the level of the two main characters, Stephan and Yanek.3 A more serious problem is Abrahmss assumption that the behavior of terrorist organizations is irrational and therefore a puzzle for the strategic paradigm. Take, for example, his puzzle that terrorist organizations with identical political platforms routinely attack each other more than their mutually professed enemy (p. 82). Is this indeed irrational, nonstrategic behavior? Not when one considers that most terrorist movements have two basic objectives: to wring concessions from the government but, no less, to achieve hegemony in the rebel camp. Movements such as Fatah, for example, not only strive to

1. Max Abrahms, What Terrorists Really Want: Terrorist Motives and Counterterrorism Strategy, International Security, Vol. 32, No. 4 (Spring 2008), pp. 78105. Additional references to this article appear parenthetically in the text. 2. Gordon H. McCormick, Terrorist Decision Making, Annual Review of Political Science, Vol. 6 (June 2003), p. 480. 3. See David C. Rapoport, Inside Terrorist Organizations, in Rapoport, ed., Inside Terrorist Organizations (New York: Columbia University Press, 1988), p. 2.

Contending with Terrorism 208

gain independence; they also aim to constitute the government of the future state. This is why so many wars of national liberation are also civil wars. Therefore, fratricide is often motivated by strategic thinking, as the recent struggle for dominance between Hamas and Fatah over control of the Palestinian Authority (PA) suggests. Prior to their brief civil war, the Fatah Tanzim and the al-Aqsa Martyrs Brigade undermined PA institutions to correct what activists from the inside (i.e., those who fought the Israelis while the PLO leadership was far away in Tunis) felt was an unfair share of positions of power allocated to them with the establishment of the Palestinian Authority in 1994. Regarding al-Qaida, since 2003, at least, it has almost exclusively targeted states with a military presence in Iraq, demonstrating its strategic motivations.4 The African National Congress employed terrorism only late in its struggle against the white government of South Africa, effectively refuting Abrahmss claim that terrorist organizations never use terrorism as a last resort. Whether terrorist organizations can be assumed to be nonstrategic because of changing protean political platforms can also be questioned. One wonders whether the political platforms of terrorist groups are more protean than those of regular political parties or movements. The Basque, Fatah, and PKK terrorist movements shift in their political objectives according to calculations of what is achievable, similar to the mainstream Zionist political movement, which did not engage in terrorism. Although Theodor Herzl, who founded the World Zionist Organization, wrote about Jewish statehood even before the creation of the Zionist movement, statehood was never mentioned again as the movements goal by mainstream Zionists at least until the Holocaust. The reason was because the leadership believed that disclosure of its ultimate objective of a Jewish state was not politically expedient. A terrorist movement that ratchets up its objective from achieving autonomy to statehood or ratchets down its demands from statehood to autonomy might be making a rational (re)assessment between its goals and its capabilities relative to those of the state. Yasir Arafat, for example, accepted the autonomy plan in 1993, to which he had been bitterly opposed for more than twenty years, when he signed the Declaration of Principles along with Prime Minister Yitzhak Rabin on the White House lawn, for fear that Israel might make a deal with the leadership within the territories. Hamas prefers controlling Gaza compared to the
4. Robert A. Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (New York: Random House, 2005), p. 57.

Correspondence: What Makes Terrorists Tick 209

much smaller probability that it would be allowed to build a radical Islamist Palestinian state. In short, terrorist movements can be strategic even when changing their political platforms over time.5 Abrahms also conates the motivations of terrorists with those of the movements leadership. Just as techniques used to motivate individual soldiers or small units do not necessarily reect the strategic objectives of the national and military leadership, neither can one necessarily assume correspondence between the factors motivating terrorists and the strategic objectives of the terrorist movement. Terrorists might indeed join the movement in search of companionship, as Abrahms suggests, but this hardly reects on the strategic goals of the movement itself. Equally problematic is Abrahmss proposal that the social model is more relevant than the strategic model when assessing terrorists motivations. If indeed terrorists are rational people who use terrorism primarily to develop strong affective ties with fellow terrorists, collective terrorism should be more prevalent in Western states, as alienated individuals seek friendship and social interaction (p. 80). Nor can the social model account for why, in the Palestinian political-terrorist arena, organizations such as the Abu Nidal Organization and the Popular Front for the Liberation of PalestineGeneral Command pursued violence at all costs, whereas the Fatah mainstream movement sharply reduced its terrorist activity during the Oslo years (19942000) and then returned to it with a vengeance once the al-Aqsa intifada broke out. Abrahmss explanation might account for why the Abu Nidals of the world exist (although even Abrahms concedes that they became guns for hire and thus might have been acting strategically, assuming the alternative would have meant a life sentence), but it cannot account for why Fatah, the dominant group in the Palestinian arena for decades, was strategic. Even in the life cycle of a single terrorist group, the salience of visceral psychological or social explanations might vary over time. Forty years ago, Fatah members were heavily inuenced by Franz Fanons thesis of violence as catharsis. Fanons name and his legacy nd almost no echo in more recent Fatah writings. Instead of seeking to determine whether the strategic model or the social/ psychological paradigm is correctthey are both right in the sense that some

5. Harvey Kushner, Terrorism in America: A Structured Approach to Understanding the Terrorist Threat (New York: Charles and Thomas, 1998), pp. 8183. See also Leonard Weinberg and William Lee Eubank, Cultural Differences in the Behavior of Terrorists, Terrorism and Political Violence, Vol. 6, No. 1 (Spring 1994), pp. 68.

Contending with Terrorism 210

organizations act according to the expectations of the strategic model and others because of social/psychological expectationsscholars should be asking why some organizations seem to be more strategic and others less so or not at all. An initial explanation can be derived from the works of Barry Collins and Harold Guetzkov, as well as Martha Crenshaw.6 These scholars observed that, as movements go deeper underground, the interpersonal rewards of group membership become more important than the strategic goals of their political mission. One would assume that as long as organizations are able to grow more politically salient, they will maintain or cultivate a strategy to achieve denable political goals. Conversely, failure, setback, or prolonged marginality will make organizations less strategic, as their immediate social environment overwhelms them. The seeds of a competing argument to that proposed by Abrahms may have emerged. Abrahms argues that it is the quest for affective ties that drives terrorist organizations, whereas this alternative explanation is based on the premise that it is the competitive environment between terrorists and government and the relative success achieved by the terrorists over time that predicts how strategic or nonstrategic they may become. This is only one of many potential alternative explanations, however. What is important is the need to take up the challenge to explain the puzzle of why some terrorist organizations seem to be strategic, others not, and still others seem to be both over different time periods. One can then tease out the policy implications of the ndings for conict resolution or, short of that, for successful counterinsurgency. Abrahms has made a valuable contribution to explaining why some terrorist organizations are not necessarily motivated by political goals, and many more at some time during their life cycles. The puzzle of why this is true of some organizations most of the time, or most organizations some of the time, remains a challenge that students of terrorism should endeavor to meet. Hillel Frisch Ramat Gan, Israel

6. Barry E. Collins and Harold Guetzkow, The Social Psychology of Group Processes for Decision Making (New York: John Wiley and Sons, 1964); and Martha Crenshaw, An Organizational Approach to the Analysis of Political Terrorism, Orbis, Vol. 29, No. 3 (Fall 1985), pp. 465489. These works are analyzed in McCormick, Terrorist Decision Making, p. 489.

Correspondence: What Makes Terrorists Tick 211

To the Editors (Paul Staniland writes):

In What Terrorists Really Want: Terrorist Motives and Counterterrorism Strategy,1 Max Abrahms argues that terrorists are social solidarity maximizers (p. 101), rather than strategic political actors, meaning that individuals join terrorist groups to develop strong affective ties with other terrorists (p. 96). As a result, terrorist groups routinely engage in actions to perpetuate and justify their existence, even when these undermine their ofcial political agendas (pp. 101102). Abrahms argues that the seven predictions of the natural systems model nd support in the preponderance of theoretical and empirical evidence (p. 103). Abrahms makes a valuable contribution to scholars understanding of terrorism. His argument, however, has signicant theoretical and empirical shortcomings. I focus on three issues: the diverse motivations of individual participants, armed group elites use of organizations to shape the actions of foot soldiers, and the behavior of terrorist organizations in Northern Ireland and Sri Lanka, two cases that Abrahms repeatedly refers to in support of his claims. the multiple logics of participation in militancy Research on civil wars suggests that it is impossible to offer a single answer to the question, What do terrorists want?2 This work shows that there is no dominant logic of participation in militancy. In a study of combatants in Sierra Leones civil war, Macartan Humphreys and Jeremy Weinstein nd that different logics of participation may coexist in a single war.3 Social sanction, grievances, and selective incentives all provided motivation to join. Stathis Kalyvas and Ana Arjonas survey of former combatants in the Colombian civil

1. Max Abrahms, What Terrorists Really Want: Terrorist Motives and Counterterrorism Strategy, International Security, Vol. 32, No. 4 (Spring 2008), pp. 78105. Further references to this article appear parenthetically in the text. 2. It is possible that insurgent groups differ from purely terrorist groups; indeed Abrahms draws such a distinction on page 83 of his article. Still, he repeatedly references insurgents such as the Tamil Tigers; Hezbollah; PKK; Hamas; Afghan, Iraqi, and Chechen guerrillas; and FARC in support of his argument. Thus there does not appear to be an empirically relevant distinction. There are few major organizations that do not kill both civilians and agents of the state. 3. Macartan Humphreys and Jeremy M. Weinstein, Who Fights? The Determinants of Participation in Civil War, American Journal of Political Science, Vol. 52, No. 2 (April 2008), p. 437.

Contending with Terrorism 212

war reveals that revenge, ideology, and material rewards were the dominant motivations for participation.4 Qualitative eldwork makes similar points. Lucian Pye explores the varied reasons people joined communist guerrilla groups in Malaya.5 Elisabeth Wood highlights multiple mechanisms that encourage participation, particularly oppressive state policy and the pleasure of agency.6 Weinsteins work on insurgent organization suggests at least two types of participantsinvestors and consumers.7 Roger Petersens study of insurgent mobilization reveals several mechanisms that encourage participation, including community norms, rational calculations, psychological hopes, and a lack of other options.8 elite goals and organizational control Abrahms assumes that individuals motives for joining a group determine the goals and actions of the organization. The problem with this assumption is that insurgent elites often have political goals distinct from foot soldiers varied motivations. Elites use organizational mechanisms to bring cadre behavior in line with leadership aims. For instance, recruits joined the Provisional Irish Republican Army (PIRA) for many reasons (though only after the political crisis of the late 1960s). PIRAs ranks included thugs, intellectuals, ordinary people, criminals, Marxists, Catholics, and dozens of other types of individuals.9 But throughout the conict, the PIRA leadership pursued the political goal of unifying Ireland by calibrating offensives, strategizing about how to use the Sinn Fin political party, negotiating with other parties and the British government, and nally agreeing to a political settlement and disbandment. The leadership used coercion, material rewards, persuasion, and social pressures to largely keep its cadres in line, regardless of their original motivations for joining.10
4. Stathis Kalyvas and Ana Arjona, Preliminary Results of a Survey of Demobilized Combatants in Colombia, unpublished paper, Yale University, 2006. 5. Lucian Pye, Guerrilla Communism in Malaya: Its Social and Political Meaning (Princeton, N.J.: Princeton University Press, 1956). 6. Elisabeth Jean Wood, Insurgent Collective Action and Civil War in El Salvador (Cambridge: Cambridge University Press, 2003). 7. Jeremy M. Weinstein, Inside Rebellion: The Politics of Insurgent Violence (Cambridge: Cambridge University Press, 2007). 8. Roger D. Petersen, Resistance and Rebellion: Lessons from Eastern Europe (Cambridge: Cambridge University Press, 2001). 9. Eamon Collins provides an inside picture of the range of personalities in the PIRA. See Collins with Mick McGovern, Killing Rage (London: Granta, 1998). 10. Rogelio Alonso highlights internal repression as central to PIRA operations. See Alonso, The IRA and Armed Struggle (London: Routledge, 2007), especially pp. 34.

Correspondence: What Makes Terrorists Tick 213

Foot soldiers similarly end up in the Liberation Tigers of Tamil Eelam (LTTE, or Tamil Tigers) through numerous mechanisms, but the core leadership has resolutely pursued the goal of an independent Eelam despite heterogeneous cadre-level motivations. Some foot soldiers were abducted as children; others fervently believe in the need for a Tamil state; others are in thrall to the personality cult of LTTE leader Velupillai Prabhakaran; others need a job; still others joined after seeing their mothers raped by soldiers.11 The Tigers leadership has harnessed these diverse motivations toward a consistent separatist goal. Abrahms emphasizes organization theory, but he does not deal with the vast literature on the control, indoctrination, and incentives that leaders use to create coordinated outcomes from disparate individuals.12 The Tigers and PIRA are fundamentally political, not affective. In the next section I show that both groups leaders have used their organizational control to pursue strategic and political goals.13 how do armed groups behave? Below I briey survey the accuracy of Abrahmss seven predictions of group behavior. I nd little support for his argument in Northern Ireland and Sri Lanka, where I have done detailed eld research on armed groups. The failure of the natural systems model to illuminate these cases casts doubt on its broader accuracy. I focus on the Provisional IRA and LTTE but also discuss the other major groups in each war. Abrahmss rst prediction is that armed groups prolong their existence by relying on a strategy that hardens target governments from making policy concessions (p. 102). By contrast, the LTTE has repeatedly brought the government of Sri Lanka to the negotiating table. It won some concessions from the government in 198990 and 2002, though neither set of concessions went far enough for the Tigers. The PIRA extracted concessions from the British and then disbanded. There is little evidence that either group intentionally stopped the government from making concessions that it believed would advance its

11. A knowledgeable aid worker in Sri Lanka listed six major motivations for individuals to join the Tigers, in addition to forced conscription. Interview by author, Colombo, March 2008. 12. A pioneering study is Nathan Leites and Charles Wolf Jr., Rebellion and Authority: An Analytical Essay on Insurgent Conicts (Chicago: Markham, 1970). 13. Between 1923 and 1968, the IRA was a peripheral and largely inactive organization. The LTTE was formed in 1972, the other Tamil militant groups between 1975 and 1980; major combat did not occur until 1983. It is hard to understand how the quest for affective ties explains this over-time variation, unless there were no socially alienated single males in these societies prior to the late 1960s. This seems unlikely. Abrahms is trying to explain a variable with a constant.

Contending with Terrorism 214

strategic goal. Conicts endured because of uctuating power and distrust between the state and its foe. This is also true of the other armed groups. In Northern Ireland, the Ofcial IRA (OIRA) unilaterally declared a cease-re in 1972 despite the absence of concessions, and the Ulster Volunteer Force (UVF) signed on to the 1998 Good Friday agreement. In Sri Lanka the other major Tamil militant groupsEelam Peoples Revolutionary Liberation Front (EPRLF), Tamil Eelam Liberation Organization (TELO), Eelam Revolutionary Organization of Students (EROS), and Peoples Liberation Organization of Tamil Eelam (PLOT)all agreed to the 1987 Indo-Lanka accord.14 Abrahmss second prediction is that armed groups ensure their continued viability by resisting opportunities to peacefully participate in the democratic process (p. 102). Cross-national research casts doubt on this claim.15 Northern Irelands history is similarly disconrming. The PIRAs Sinn Fin political wing has engaged in the democratic process and now shares power in the Northern Ireland Assembly. The OIRA entered mainstream politics as the Workers Party. The UVF declared a cease-re in 1994 and put forward the Progressive Unionist Party as its electoral face. In Sri Lanka, EPRLF, PLOT, and TELO all agreed to lay down their arms and entered the democratic mainstream via the 1987 Indo-Lanka accord. Even after rearming during 198790 to protect themselves, EPRLF and TELO are now unarmed parties that regularly contest elections. The armed parties keep their weapons to protect themselves from the Tigers, who have deemed them political traitors.16 Abrahmss third prediction is that armed groups avoid disbanding by reexively rejecting negotiated settlements that offer signicant policy concessions (p. 102). In Northern Ireland, the PIRA and the OIRA have disbanded, and the UVF is in the process of doing so. The Irish National Liberation Army (INLA) and Ulster Defence Association (UDA) remain intact, but have become fractious drug dealers, not seekers of social solidarity. In Sri Lanka, EPRLF, TELO, and a faction of EROS have disbanded as combatants. Groups that have not disbanded have plausible strategic rationales.
14. M.R. Narayan Swamy, Tigers of Lanka: From Boys to Guerrillas (Delhi: Konark, 1994), pp. 240 250. 15. Abrahms asserts, No peace process has transformed a major terrorist organization into a completely nonviolent political party (p. 86). A study of 648 terrorist groups between 1968 and 2006, however, reveals that a transition to the political process is the most common way in which terrorist groups ended (43 percent). Seth G. Jones and Martin C. Libicki, How Terrorist Groups End: Lessons for Countering al Qaida (Washington, D.C.: RAND, 2008), p. xiii. 16. There are three pro-state Tamil armed groups that contest elections: the PLOT, the Eelam Peoples Democratic Party (EPDP), and Tamil Makkal Viduthalaip Puligal (TMVP).

Correspondence: What Makes Terrorists Tick 215

The LTTE does not trust a Sinhalese-dominated government to deliver on the terms of a settlement, much less grant independence. PLOT, EPDP, and TMVP remain armed to protect themselves and reap rewards from extortion.17 Abrahmss fourth prediction is that armed groups guarantee their survival by espousing a litany of protean political goals that can never be fully satised (p. 102). Although protean political goals certainly are espoused by some groups around the world, in Northern Ireland and Sri Lanka, organizations have advanced clear separatist goals. The IRA wanted the unication of Irelandan unlikely goal, but given the history of 191622, the isolation of Northern Ireland from the British mainstream, and demographic trends, not unthinkable.18 The OIRA was willing to settle for democratization and popular participation, and the PIRA showed it would settle for power sharing. The Protestant paramilitaries wanted to maintain union with the United Kingdom.19 In Sri Lanka, even the hardest-line Tamil groups have sought a straightforward goalan independent Tamil homeland. As shown in 1987 and since, several Tamil groups have been willing to settle for power devolution. Abrahmss fth prediction is that armed groups avert organization-threatening reprisals by conducting anonymous attacks (p. 102). In Northern Ireland, the PIRA, OIRA, and INLA often claimed responsibility for their attacks. When they did not, it was because they did not want to lose support or be condemned internationally for an operation that went wrong. The INLA took credit for killing Member of Parliament Airey Neave, but not for the Darkley massacre. The PIRA tried to distance itself from the Enniskillen bombing, but took credit for bombing Thiepval Barracks.20 The Protestant paramilitaries claimed responsibility when it suited their purposes. In Sri Lanka in the 1980s, the competition between Tamil groups encouraged them to take credit for attacks. Since the LTTE gained primacy, it has publicized and claimed credit for many of its suicide attacks. Abrahmss sixth prediction is that armed groups annihilate ideologically identical terrorist organizations that compete for members (p. 102). It is un-

17. Human Rights Watch, Sri Lanka: Political Killings during the Ceasere, Human Rights News (New York: Human Rights Watch, August 7, 2003), http:/ /hrw.org/backgrounder/asia/ srilanka080603.htm. 18. For an analysis, see M.L.R. Smith, Fighting for Ireland? The Military Strategy of the Irish Republican Movement (London: Routledge, 1995). 19. Steve Bruce, The Red Hand: Protestant Paramilitaries in Northern Ireland (Oxford: Oxford University Press, 1992). 20. The name used to take credit for PIRA attacks was P. ONeill. After the disastrous Enniskillen bombing (which killed eleven civilians), the PIRA leadership blamed local mistakes by the Fermanagh Brigade.

Contending with Terrorism 216

clear what behavior this would refer to in either case. The PIRA was ideologically different from the OIRA, which it split from in 1969. The Irish National Liberation Army differed from the OIRA leadership, resulting in a break in 1974.21 The PIRA espoused a hard-line Irish nationalism, in contrast to the mass mobilizing, violence-averse focus of the OIRA or the far-left vision of the INLA.22 These ideological differences help to account for why the PIRA and the INLA broke away from the OIRA and why the OIRA stopped ghting. Despite feuds between these groups, there was no annihilation. Abrahmss discussion of Sri Lanka is empirically suspect. First, he incorrectly claims that the Tamil Tigers did not target the Sinhalese government in the mid-1980s (p. 90). The LTTEs initial killings in the 1970s targeted progovernment political guresthe mayor of Jaffna in 1975, two policemen in 1977, and two police inspectors and a member of Parliament in 1978. The LTTEs June 1983 ambush of a convoy in Jaffna, which killed thirteen soldiers, triggered a major escalation of violence between the Tigers and the state. The LTTE did make a bid for dominance (primarily between April 1986 and March 1987) by targeting other groups, but it was simultaneously involved in antigovernment insurgency throughout northeastern Sri Lanka. Second, there were important ideological differences among Tamil groups. The LTTEs closest ideological match was TELO, but it was an Indian proxy army. The LTTE, by contrast, distrusted the Indian state. PLOT broke away from the LTTE in part because of PLOT supporters left-wing vision of Sinhalese-Tamil cooperation in seeking to forge an island-wide revolution.23 EPRLF and EROS espoused leftist ideologies and emphasized lower-caste mobilization. These differences mattered in the negotiation and (failed) implementation of the Indo-Lanka accord of 1987: EPRLF and TELO supported it, the LTTE opposed it, and PLOT tried to remain neutral.24 The remnants of PLOT and parts of EPRLF and EROS now support the Sri Lankan government; another faction of EROS was integrated into the Tigers. Abrahmss seventh prediction is that armed groups refuse to split up after the armed struggle has proven politically unsuccessful for decades or its politi21. Jack Holland and Henry McDonald, INLA: Deadly Divisions (Dublin: Torc, 1994). 22. Peter Taylor, The Provos: The IRA and Sinn Fein (London: Bloomsbury, 1998); and Richard English, Armed Struggle: The History of the IRA (Oxford: Oxford University Press, 2003). 23. For discussion of PLOT ideology, see Mark P. Whitaker, Learning Politics from Sivaram: The Life and Death of a Revolutionary Tamil Journalist in Sri Lanka (London: Pluto, 2007). 24. Rohan Gunaratna, Indian Intervention in Sri Lanka: The Role of Indias Intelligence Agencies (Colombo: South Asian Network on Conict Research, 1993); and University Teachers for Human Rights (Jaffna), The Broken Palmyrah (Jaffna, Sri Lanka: University Teachers for Human Rights, 1990).

Correspondence: What Makes Terrorists Tick 217

cal rationale has become moot (p. 102). Contrary to Abrahmss emphasis on social solidarity, most of these organizations have suffered a variety of elite splits and internal conict over politics and personalities. The origins of PIRA and INLA lie in disagreement with the OIRA leadership over how to respond to sectarian violence in Belfast and Derry.25 In 1997 the Real IRA broke away from PIRA because of discontent with its growing politicization.26 The Loyalist Volunteer Force split from the UVF in 1996 in part over the peace process.27 Splits and clashes within the UDA and INLA have occurred because of personal and criminal rivalries. In Sri Lanka, the LTTE suffered a major split when its eastern commander, Karuna Amman, broke away in 2004. This split was driven by personal rivalry and disagreement over the desirability of armed struggle.28 Internal factional struggles within the LTTE occurred in 1980 (leading to PLOTs formation) and the early 1990s. EPRLF broke from EROS in 1980, and the EPDP in turn from the EPRLF in 1987, partly because of disagreements about politics. EROS broke in two in the late 1980s over joining the LTTE. TELO was in the midst of an internal war when targeted by the Tigers in 1986.29 Given the pervasiveness of intra-organizational splits, feuds, and factionalism, the idea that militants seek Chester Barnards condition of communion (p. 95) is bizarre. the complexity of militancy Rather than apolitical but solidaristic collections of lonely men, many militant groups bear resemblance to militaries and even statesorganizationally complex, often internally divided, driven by political goals, and sometimes willing to abandon violence. Individuals join for numerous reasons and, in turn, are shaped by mechanisms of discipline and indoctrination. This does not mean that either militants or armed groups are purely strategic or rational actors, but it does suggest that they are not simplistic solidarity maximizers. Paul Staniland Cambridge, Massachusetts
25. Patrick Bishop and Eamon Mallie, The Provisional IRA (London: Corgi Books, 1987), pp. 119 138. 26. On the Continuity split, see Robert W. White, Ruar Brdaigh: The Life and Politics of an Irish Revolutionary (Bloomington: Indiana University Press, 2006). On the Real split, see Ed Moloney, A Secret History of the IRA (New York: W.W. Norton, 2002), chap. 17; and English, Armed Struggle, pp. 316318. 27. Jim Cusack and Henry McDonald, UVF (Dublin: Poolbeg, 1997). 28. Situation Report: The Rise and Fall of Karuna, Sunday Times (Colombo), April 11, 2004. 29. This history is based on Swamy, Tigers of Lanka.

Contending with Terrorism 218

Max Abrahms Replies:

I appreciate the thoughtful comments of Erica Chenoweth and her coauthors, Paul Staniland, and Hillel Frisch to my recent article, as well as this opportunity to address them. In my article I presented a test of the strategic model, the dominant paradigm in the academic and policy communities on terrorist motives, behavior, and counterterrorism strategy. This model holds that rational people use terrorism to achieve a political return; terrorist groups hence operate as political maximizers; and governments can therefore combat terrorism by reducing the political utility for its practitioners. Contrary to the strategic model, I found that in practice terrorist groups engage in seven politically counterproductive behavioral tendencies that perpetuate the existence of the group at the expense of its ofcial political goals. According to the natural systems model in the organization literature, this behavioral trade-off reveals that members of an organization attach greater importance to its social benets than to its ofcial ones. I concluded that if terrorists likewise prioritize the social benets over the political benets of participating in terrorist groups, then extant counterterrorism strategies require reform.1 Erica Chenoweth, Nicholas Miller, and Elizabeth McClellan raise three main objections to my argument.2 First, they claim that my analysis of terrorist motives pertains only to foot soldiers, and not to their leaders. Second, they draw upon a recent RAND study to contest the empirical basis of the seven puzzling terrorist tendencies for the strategic model that I identied.3 Third, they maintain that these putative puzzles, even if empirically valid, are not evidence of irrationality given the internal dynamics of terrorist groups. Each objection is misplaced. First, demographic studies routinely nd that terrorist foot soldiers are motivated by the socialnot the politicalreturn.4 This is consistent with de1. Max Abrahms, What Terrorists Really Want: Terrorist Motives and Counterterrorism Strategy, International Security, Vol. 32, No. 4 (Spring 2008), pp. 78105. 2. Erica Chenoweth, Nicholas Miller, and Elizabeth McClellan, Correspondence: Politics, Motivation, and Organization in the Study of Terrorism, International Security, Vol. 33, No. 4 (Spring 2009), pp. 180202. 3. Seth G. Jones and Martin C. Libicki, How Terrorist Groups End: Lessons for Countering al Qaida (Washington, D.C.: RAND, 2008). 4. See, for example, Marc Sageman, Understanding Terror Networks (Philadelphia: University of Pennsylvania Press, 2004), p. 92; Olivier Roy, Terrorism and Deculturation, in Louise Richardson, ed., The Roots of Terrorism (New York: Routledge, 2006), pp. 159160; and Joseph Felter and Jarret Brachman, An Assessment of 516 Combatant Status Review Tribunal (CSRT) Unclassied Summaries, CTC Report (West Point, N.Y.: Combating Terrorism Center, July 15, 2007), pp. 2425, 34.

Correspondence: What Makes Terrorists Tick 219

mographic research on national militaries, which nds that troops are also generally motivated by the social benets of participation rather than by the missions ofcial political purpose.5 In keeping with the strategic model, however, my study ascertained the motives of terrorists by analyzing the behavioral tendencies of terrorist groups. This approach helps to account for the motives of their leaders because they, by denition, inuence the core decisions of the group.6 My research shows that the behavior of terrorist groups is fundamentally different from that of national militaries. Unlike militaries, terrorist groups tend to act in ways that keep the ghting unit intact, even when these actions undercut these groups given political rationale. This behavioral trade-off implies that terrorist members with the greatest capacity to inuence the core decisions of the group attach utmost value to perpetuating the social unit. Second, rather than undermining the seven puzzles to the strategic model, the RAND study lends empirical support to these politically irrational terrorist tendencies. In my article I demonstrated, for example, that the use of terrorism is itself an unproductive behavior for groups to achieve their political platforms, an obvious puzzle for the strategic model. Chenoweth, Miller, and McClellan cite the study to declare that terrorist groups have a political success rate of 10 percent, and hence terrorism is perhaps a rational political strategy. In fact, what Seth Jones and Martin Libicki found is that 10 percent of terrorist groups that have ended since 1968 accomplished their political platforms. An additional 244 groups still use terrorism, and another 136 groups have morphed into other terrorist groups. All told, the RAND study concluded that only 4 percent of terrorist groups since 1968 have attained their policy demands. Even more important, the study stresses that these outlying groups did not tend to accomplish these demands due to terrorism. For example, several Armenian groups achieved their political platforms only by dint of the Soviet Unions unraveling.7 The study thus afrms the empirical puzzle for the strategic model that terrorism is ineffective in achieving terrorist groups

5. See Benjamin A. Valentino, Final Solutions: Mass Killing and Genocide in the 20th Century (Ithaca, N.Y.: Cornell University Press, 2005), p. 58. 6. Inferring the motives of terrorists through their behavior makes additional sense as terrorist groups have become increasingly networked, blurring the distinction between leaders and foot soldiers. This trend applies not only to religious groups such as al-Qaida and its afliates but also to ethnonationalist groups such as Fatah and Hamas. On the increasingly networked nature of even ethnonationalist terrorist groups, see Ami Pedahzur and Arie Perliger, The Changing Nature of Suicide Attacks: A Social Network Perspective, Social Forces, Vol. 84, No. 4 (June 2006), pp. 19872008. 7. Jones and Libicki, How Terrorist Groups End, p. 15.

Contending with Terrorism 220

platforms. In the article I also demonstrated why it is politically irrational for terrorist groups to reexively eschew peace processes, elections, and other potentially effective nonviolent political avenues. To challenge the external validity of these puzzles, Chenoweth, Miller, and McClellan repeatedly reference the RAND study as claiming that since 1968, 43 percent of terrorist groups have ended by entering the political process, broadly dened. They seize on this gure to suggest that terrorist groups are rational political actors that survey their strategic options, often pursuing nonviolent alternatives when these serve their political ends. This gure, however, again excludes hundreds of terrorist groups that are hardest for the strategic model to explain, including those that have splintered for the purpose of derailing peace processes, elections, and other historically productive nonviolent political alternatives. When the universe of terrorist groups is included in the analysis, the RAND study found that since 1968 fewer than one in ve groups have embraced any type of nonviolent political path, a strikingly low percentage given terrorisms manifest political impotence. In sum, a careful analysis of the RAND study bolsters my contention that terrorist groups tend to engage in behaviors that prolong their existence, even when these impede progress on their stated political aims. This trade-off can account for the pronounced disparity between the longevity of terrorist groups and their negligible political accomplishments.8 Third, Chenoweth, Miller, and McClellan also assert that the seven behavioral tendencies I discuss in my article are not evidence of terrorist irrationality. I agree; the seven politically counterproductive terrorist tendencies are irrational only from the vantage of the strategic model. They make perfect sense if terrorist groups are driven not to coerce government compliance, but to perpetuate the social unit. Indeed, Chenoweth and her coauthors note that terrorist groups rely on anonymous attacks both to prevent the target government from eliminating the organization and to generate new recruits by provoking government overreaction; that terrorist groups abruptly change their political platforms to retain current members and to attract new ones; and that to win recruits, terrorist groups often attack other groups that share their platform. Chenoweth, Miller, and McClellan ultimately reject a social explanation for terrorism, however, claiming that these terrorist behaviors are reconcilable with what they coin as strategic models of terrorism. Yet they neither dene this nomenclature nor demonstrate that prolonging the life span of terrorist
8. See Max Abrahms, Why Terrorism Does Not Work, International Security, Vol. 31, No. 2 (Fall 2006), pp. 4278.

Correspondence: What Makes Terrorists Tick 221

groups helps to advance their policy demandsa necessary, empirically invalid assumption for their behaviors to be politically rational.9 Paul Staniland raises three other objections with my article.10 First, he draws upon the civil war and guerrilla warfare literatures to argue that many militant groups do not behave in the manner I described. This was not my unit of analysis, however. Within terrorism studies, there is a widespread belief that for maximum analytic leverage, terrorist groups, as traditionally dened, should be distinguished from other types of violent substate groups.11 This is precisely the reason we have the eld of terrorism studies. Second, Staniland states that it is impossible to offer a single answer to the question, What do terrorists want? My goal, however, was more modestto compare the power of a social explanation for terrorism to a political one by analyzing the behavioral tendencies of terrorist groups. His third point, ironically, is that the behavior of the Irish Republican Army (IRA), Tamil Tigers, and afliates reveals that their leaders used terrorism rationally for a single purposeto achieve concrete political aims. In defense of the conventional wisdom, Staniland claims to show that the seven politically counterproductive habits of terrorist groups are nowhere to be found in the histories of the IRA, Tigers, or associated groups, casting doubt on the broader accuracy of my study. His assessment is problematic for four reasons. First, Staniland compares apples to oranges by testing slightly different terrorist group behaviors than I did. He quotes my work extensively, but none of the terrorist tendencies that he tests is culled from the sections that dealt with the strategic model. In fact, Staniland does not purport to test this model, even though that was the purpose of my study. Instead, he treats the deracinated variants of the terrorist tendencies as testable predictions to falsify the natural systems model. This model makes no such predictions, however. The natural systems model is not a terrorism model at all; it is a methodological device for assessing the social basis of organizational behavior.12
9. See Abrahms, Why Terrorism Does Not Work. 10. Paul Staniland, Correspondence: Politics, Motivation, and Organization in the Study of Terrorism, International Security, Vol. 33, No. 4 (Spring 2009), pp. 189195. 11. On this distinction, see Abrahms, Why Terrorism Does Not Work, pp. 5254; Jeff Goodwin, A Theory of Categorical Terrorism, Social Forces, Vol. 84, No. 4 (June 2006), pp. 202746; Assaf Moghadam, Suicide Terrorism, Occupation, and the Globalization of Martyrdom: A Critique of Dying to Win, Studies in Conict and Terrorism, Vol. 29, No. 6 (August 2006), p. 710; and Jones and Libicki, How Terrorist Groups End, appendix. 12. On the perils of testing empirical relationships outside of their theoretical context, see David A. Lake, Fair Fights? Evaluating Theories of Democracy and Victory, International Security, Vol. 28, No. 1 (Summer 2003), p. 154.

Contending with Terrorism 222

Second, in treating the seven terrorist tendencies as discrete predictions, Staniland strongly implies that a groups failure to engage in all of them constitutes disconrming evidence of its political irrationality. The strategic model is clear, however, that a terrorist group is politically irrational whenever it engages in any of these seven tendencies. Sun-Ki Chai has pointed out, for example, that if terrorism is itself politically ineffective behavior, then it is not politically rational, regardless of whether there exist alternative methods of inducing political change.13 Conversely, if terrorist groups do not seize superior political outlets, their behavior is also politically irrational, even if it is true that a very limited number of groups that have used terrorism happened to accomplish their political platforms. A large portion of Stanilands test is thus irrelevant for demonstrating that a terrorist group did not behave as a political maximizer in accordance with the strategic model, even when its underlying rationality assumptions are relaxed. Third, Staniland uses his case studies in a methodologically unorthodox and suspect way. To demonstrate the tendency of terrorist groups to engage in the seven politically irrational behaviors, I relied on large-n studies, citing my own and those of others. Staniland charges that his two cases disconrm these terrorist tendencies. Case studies offer a number of methodological advantages over large-n studies, but generalizing empirical relationships is not considered one of them.14 This is a particular concern because several of the actions highlighted in his two cases are seen as exception[s] that are contextually specic and thus unlikely to be widely replicable.15 Fourth, Staniland is mistaken to conclude that none of the seven politically counterproductive tendencies was evident in the Irish or Tamil campaigns. In both cases, his empirical analysis suffers from problems of commission and omission. For reasons of space, I will restrict my analysis to the IRA and its republican afliates, groups that Staniland claims to know best.16 I will set aside
13. Sun-Ki Chai, An Organizational Economics Theory of Antigovernment Violence, Comparative Politics, Vol. 26, No. 1 (October 1993), p. 100. 14. See Gary King, Robert O. Keohane, and Sidney Verba, Designing Social Inquiry: Scientic Inference in Qualitative Research (Princeton, N.J.: Princeton University Press, 1994), chap. 6. 15. Donald L. Horowitz, Explaining the Northern Ireland Agreement: The Sources of an Unlikely Constitutional Consensus, British Journal of Political Science, Vol. 32, No. 2 (April 2002), p. 193; Andrew Reynolds, A Constitutional Pied Piper: The Northern Irish Good Friday Agreement, Political Science Quarterly, Vol. 114, No. 4 (Winter 19992000), p. 617; and Ami Pedahzur, Suicide Terrorism (Cambridge: Polity, 2005), p. 72. These quotations refer to the Good Friday agreement and the Tigers military capabilities, which greatly exceed those of typical groups that have used terrorism. 16. For two excellent accounts of organizational goals trumping political goals in the Tamil cam-

Correspondence: What Makes Terrorists Tick 223

the demographic studies of their members who were found to be principally motivated by social solidarity and not the political return.17 Instead, I will focus on the rst three politically irrational terrorist group behaviors that I identied, which the Irish case allegedly disconrms. Large-n studies show that terrorism is an ineffective coercive strategy.18 Staniland argues that the 1998 Good Friday power-sharing agreement is countervailing evidence. His view of this agreement (or executive) is at odds, however, with that of longtime IRA experts. Louise Richardson explains, Those who argue, for example, that the establishment of the power-sharing executive in Northern Ireland has rewarded the terrorism of the IRA are quite wrong. The IRA did not wage a terrorist campaign to share power with Protestants in Northern Ireland. . . . The IRA campaign was fought to bring about a united Ireland, which they have not succeeded in achieving.19 Other IRA experts agree with Richardson that Irish republican terrorism was a political failure. In fact, a recurrent observation in the literature is that the attacks on civilians steeled Londons resolve against making territorial concessions.20 In addition, large-n studies demonstrate that terrorist groups do not tend to use terrorism as a last resort, even when democratic political alternatives exist.21 Staniland argues that the Irish case is again disconrming. Yet the
paign, see Mia Bloom, Dying to Kill: The Allure of Suicide Terror (New York: Columbia University Press, 2005), chap. 3; and Pedahzur, Suicide Terrorism, chap. 4. These accounts detail the Tigers target selection, offering strong evidence that they did not behave as political maximizers. 17. See, for example, Robert W. White, Political Violence by the Nonaggrieved, in Donatella della Porta, ed., International Social Movement Research, Vol. 4 (Greenwich, Conn.: Jai Press, 1992), p. 92; Robert W. White, From Peaceful Protest to Guerrilla War: Micromobilization of the Provisional Irish Republican Army, American Journal of Sociology, Vol. 94, No. 6 (May 1989), pp. 1277 1302; and Edgar OBallance, Terror in Ireland: The Heritage of Hate (Novato, Calif.: Presidio, 1981), p. 138. 18. See Abrahms, Why Terrorism Does Not Work. In her comprehensive study on how terrorist campaigns end, Audrey Kurth Cronin also nds that terrorism has not paid politically. See Cronin, How Terrorism Ends: Understanding the Decline and Demise of Terrorist Campaigns (Princeton, N.J.: Princeton University Press, forthcoming). 19. Louise Richardson, What Terrorists Want: Understanding the Enemy, Containing the Threat (New York: Random House, 2006), p. 75. 20. See, for example, Paul Wilkinson, Terrorism and the Liberal State, 2d ed. (New York: New York University Press, 1986), p. 163; Rogelio Alonso, The Modernization in Irish Republican Thinking toward the Utility of Violence, Studies in Conict and Terrorism, Vol. 24, No. 2 (January 2001), pp. 136137, 141; and Caroline Kennedy-Pipe, From War to Uneasy Peace in Northern Ireland, in Michael Cox, Adrian Guelke, and Fiona Stephen, eds., A Farewell to Arms? Beyond the Good Friday Agreement, 2d ed. (New York: Manchester University Press, 2006), p. 51. 21. See Abrahms, What Terrorists Really Want, pp. 8485. For an important related study, see Maria J. Stephan and Erica Chenoweth, Why Civil Resistance Works: The Strategic Logic of Nonviolent Conict, International Security, Vol. 33, No. 1 (Summer 2008), pp. 744. See also Jones and Libicki, How Terrorist Groups End.

Contending with Terrorism 224

Sinn Fin regularly complained that republican terrorist groups were not only non-electoral but militantly anti-electoral.22 Following the Sunningdale agreement of 1973, Irish Catholics gravitated toward constitutional political parties. And yet the terrorism persisted vigorously for decades.23 As Richardson points out, It is surely reasonable to expect that the same [political] result could have been achieved through concerted peaceful political action over the past thirty years and without any signicant loss of life.24 Former IRA members go further, believing that a peaceful democratic strategy would have been politically superior.25 This was eminently knowable even at the time, in the opinion of many authorities on the subject.26 Large-n studies suggest as well that terrorist groups tend to eschew compromise at the expense of their stated political goals.27 Staniland maintains that the Irish case disproves this terrorist tendency. Yet a bevy of rejectionist splinter groups emerged after Sunningdale. Most of them developed for social reasons, not political ones.28 Bruce Hoffman has likened these spoilers to sharks in the water that kept moving for no instrumental purpose other than to remain intact and avert their demise.29 Irish leaders corroborate Hoffmans account. In an article entitled The Futile Path of Militarism, the leading republican newspaper explained that the tactic of armed struggle rather than the political objective for which it is [supposedly] carried outhas been elevated to a principle.30 Other authorities have likewise concluded that after Sunningdale, the IRA and its afliates ignored the political objective for which they claimed to be struggling and raised military actions to an end in itself.31

22. Quoted in Brian Feeney, Sinn Fin: A Hundred Turbulent Years (Madison: University of Wisconsin Press, 2003), pp. 269270. 23. See Alonso, The Modernization in Irish Republican Thinking toward the Utility of Violence, p. 135. 24. Richardson, What Terrorists Want, p. 17. 25. See, for example, Eamon Collins with Mick McGovern, Killing Rage (London: Granta, 1997), p. 4. 26. See, for example, Loren E. Lomasky, The Political Signicance of Terrorism, in R.G. Frey and Christopher W. Morris, eds., Violence, Terrorism, and Justice (New York: Cambridge University Press, 1991), p. 90. 27. See, for example, Andrew Kydd and Barbara F. Walter, Sabotaging the Peace: The Politics of Extremist Violence, International Organization, Vol. 56, No. 2 (Spring 2002), pp. 263296. See also Stephen John Stedman, Spoiler Problems in Peace Processes, International Security, Vol. 22, No. 2 (Fall 1997), pp. 553; and Jones and Libicki, How Terrorist Groups End, p. 19. 28. Richardson, What Terrorists Want, p. 211. 29. Bruce Hoffman, Inside Terrorism (New York: Columbia University Press, 1998), p. 162. 30. The Futile Path of Militarism, An Phoblacht, August 20, 1998, at http:/ /republican-news.org/ archive/1998/August20/20pol.html. 31. Ibid.

Correspondence: What Makes Terrorists Tick 225

In sum, the historical record is replete with empirical evidence that Irish republican terrorist groups did not even remotely behave as political maximizers, notwithstanding the widespread belief in the strategic model. These terrorist groups acted in ways that served not to advance their political platform, but to sustain themselves, providing insight into the incentive structure of their leaders. Hillel Frisch levels two other objections. First, he maintains that the terrorist behaviors that I identied are in fact consistent with the strategic model. As evidence, he purports to show how certain terrorist groups were acting in a politically rational manner by engaging in these seemingly puzzling behaviors for the strategic model. Yet none of his cases exemplify the behavioral tendencies that I described. To demonstrate that it is actually politically rational for terrorist groups with the same political platform to ght each other, Frisch lists several well-known cases of groups with different platforms that have done so. Similarly, to demonstrate that it is actually politically rational for terrorist groups to abruptly alter the entire nature of their political demands, he lists examples of groups that have only minimally ratchet[ed] down their original demands. His examples serve to demonstrate only that terrorist groups do not always engage in all of the behavioral tendencies that I enumerated. In highlighting cases that deviate from the puzzling terrorist tendencies, he fails to demonstrate how these puzzles conform to the strategic model. Frischs second contention is that whereas some terrorist groups are motivated by the social benets, others are motivated by the political benets.32 He asserts that the former applies to defunct groups of marginal historical importance, such as the November 17 group and the Popular Forces of 25 April, and that the latter applies to the most dangerous contemporary groups, such as alQaida and its afliates. Frisch does not explain why he believes that such group types exist or how to discern them from each other. The evidence is accumulating, however, that the terrorist groups of greatest concern to the United States and its allies are not rational political actors. Based on their behavior, al-Qaida and its afliates do not appear (1) to be motivated by relatively stable and consistent political preferences; (2) to evaluate the expected political payoffs of the available options; or (3) to adopt terrorism because of its superior political return. On the contrary, terrorism is having a deleterious impact on al-Qaidas political platform, and the likelihood of ad32. Hillel Frisch, Correspondence: Politics, Motivation, and Organization in the Study of Terrorism, International Security, Vol. 33, No. 4 (Spring 2009), pp. 180202.

Contending with Terrorism 226

vancing it with terrorism is close to zero;33 its leadership opposes elections, even when these would further its given political cause, such as in Iraq;34 alQaida is reexively averse to politically compromising;35 it possesses what Jessica Stern has dubbed protean political demands, which constantly change to the befuddlement of even its military leadership;36 al-Qaida attacks are generally anonymous, or at least we suspect this to be so;37 it targets other Muslim groups even when they share major elements of its political platform;38 and al-Qaida has resisted disbanding after consistently failing to achieve its policy demands or even when they have become moot for reasons that have nothing to do with terrorism.39 These are rational behaviors only if the endgame for al-Qaida is to stay intact. The future poses a natural experiment. Will al-Qaida and its afliates immediately disband and renounce terrorism now that they have unambiguously failed to advance their political demands? Will their leaders nally embrace the ballot? Will they offer a credible compromise in the interests of achieving at least part of their ever-changing demands? If the Irish case is at all generalizable, do not hold your breath. Max Abrahms Palo Alto, California

33. Jones and Libicki, How Terrorist Groups End, p. xvii. 34. Abrahms, What Terrorists Really Want, p. 85. 35. Ashton B. Carter, John Deutch, and Philip Zelikow, Catastrophic Terrorism: Tackling the New Danger, Foreign Affairs, Vol. 77, No. 6 (November/December 1998), p. 85. 36. Jessica Stern, The Protean Enemy, Foreign Affairs, Vol. 82, No. 4 (July/August 2003), p. 1. See also Vahid Brown, Cracks in the Foundation: Leadership Schisms in al-Qaida from 19892006, CTC Report (West Point, N.Y.: September 2007), p. 10. 37. See Assaf Moghadam, The Globalization of Martyrdom: Al Qaeda, Sala Jihad, and the Diffusion of Suicide Attacks (Baltimore, Md.: Johns Hopkins University Press, 2008), pp. 4951. 38. Ibid., p. 34. 39. See Stern, The Protean Enemy, p. 14.

Part III: Countering Terrorism

Deterring Terrorism

Deterring Terrorism

Robert F. Trager and Dessislava P. It Can Be Done Zagorcheva

an deterrence work against contemporary terrorists? Many prominent international relations scholars and analysts have argued that deterrent strategies have no signicant role to play in countering the new terrorist threat. Richard Betts, for example, writes that deterrence has limited efcacy . . . for modern counterterrorism.1 A RAND study asserts, The concept of deterrence is both too limiting and too nave to be applicable to the war on terrorism.2 And the belief that deterrence is inadequate as a counterterrorist strategy is also shared by President George W. Bush and his administration, whose National Security Strategy states, Traditional concepts of deterrence will not work against a terrorist enemy.3 The case against the use of deterrence strategies in counterterrorist campaigns appears to rest on three pillars. First, terrorists are thought to be irrational, and therefore unresponsive to the cost-benet calculation required for deterrence.4 Second, as Robert Pape argues, many terrorists are said to be so highly motivated that they are willing to die, and so not deterred by fear of punishment or of anything else.5 Third, even if terrorists were afraid of punishment, they cannot be deterred because they lack a return address against

Robert F. Trager is a Fellow in the Department of Politics and International Relations at Oxford University. Dessislava P. Zagorcheva is a Ph.D. candidate in political science at Columbia University. The authors would like to thank Richard Betts, Maria Fanis, Tanisha Fazal, Robert Jervis, Brigitte Nacos, Stephanie Neuman, William Karl Riukas, Sebastian Rosato, Anne Sartori, Sean Smeland, Jack Snyder, and two anonymous reviewers for their helpful comments and advice. They are also grateful to seminar participants at the Institute for Social and Economic Research and Policy at Columbia University, and the 2003 Summer Workshop on the Analysis of Military Operations and Strategy. 1. Richard K. Betts, The Soft Underbelly of American Primacy: Tactical Advantages of Terror, in Demetrios James Caraley, ed., September 11, Terrorist Attacks, and U.S. Foreign Policy (New York: Academy of Political Science, 2002), p. 46. 2. Paul K. Davis and Brian Michael Jenkins, Deterrence and Inuence in Counterterrorism: A Component in the War on al-Qaeda (Santa Monica, Calif.: RAND, 2002), p. xviii. 3. George W. Bush, The National Security Strategy of the United States of America (Washington, D.C.: U.S. Government Printing Ofce, September, 2002), p. 15. 4. The claim that terrorists are irrational is more commonly found in the popular press, but it is also a key argument against deterrence strategies generally, and therefore must be addressed in any discussion of the use of deterrence in counterterrorism. 5. Robert A. Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (New York: Random House, 2005), p. 5.
International Security, Vol. 30, No. 3 (Winter 2005/06), pp. 87123 2006 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

229

Contending with Terrorism 230

which retaliation can be visited.6 (The claim that terrorists are fanatical appears to represent a combination of the rst and second pillars.) If these arguments are correct, not only will deterrence prove ineffective but the worldand the United States in particularfaces a grim and unpreventable onslaught of terrorist attacks. If terrorists cannot be found, the use of force against them is ineffective. Counterterrorist strategies that attempt to address root causes, such as winning hearts and minds and economic aid and democratization, are strategies for the long run. In the meantime, religious terrorism is on the rise,7 and the rate of suicide terrorist attacks has increased signicantly: from 41 in the 1980s, to 100 in the 1990s, to 174 in 200003 alone.8 These trends are particularly dangerous because many scholars, analysts, and policymakers increasingly worry that terrorists could acquire and use mass casualty weapons, arguably the gravest threat to developed countries and to world order.9 In this article, we argue that the claim that deterrence is ineffective against terrorists is wrong. Many terrorists can be deterred from actions that harm targeted states, and deterrence should remain an important weapon in the counterterrorism arsenal. Moreover, even seemingly fanatical terrorists, intensely motivated by religious beliefs, are not irrational in a sense that makes them impossible to deter. Further, some essential elements of terrorist support systems are likely to be less motivated and therefore vulnerable to traditional forms of deterrence, particularly at early decision nodes in the lengthy process of preparation required for major attacks. Even the most highly motivated terrorists, however, can be deterred from certain courses of action by holding at risk their political goals, rather than life or liberty. We show that this is possible for two reasons: (1) terrorist-state rela-

6. Betts, The Soft Underbelly of American Primacy, p. 45. 7. According to the U.S. State Departments classication, the number of religiously motivated foreign terrorist organizations rose from nearly zero in 1980 to almost half of all such organizations by 1998. See Mark Juergensmeyer, Terror in the Mind of God: The Global Rise of Religious Violence (Berkeley: University of California Press, 2000), p. 6; and Bruce Hoffman, Inside Terrorism (New York: Columbia University Press, 1999). 8. Compiled from data in Pape, Dying to Win, app. 1, pp. 253264. From 2000 to 2003, 20 of the 174 total attacks were perpetrated by Iraqi rebels against U.S. and allied forces. 9. Jessica Stern has argued that terrorists or their state sponsors could obtain nuclear and chemical materials from poorly guarded former Soviet facilities as well as the expertise of underpaid nuclear scientists. See Stern, The Ultimate Terrorist (Cambridge, Mass.: Harvard University Press, 1999). Graham Allison states that a nuclear terrorist attack on America in the decade ahead is more likely than not. Allison, Nuclear Terrorism: The Ultimate Preventable Catastrophe (New York: Times Books, 2004).

Deterring Terrorism 231

tionships, while adversarial, are often not zero sum; and (2) although terrorists are difcult to nd, powerful states still have the ability to inuence their political aims. From a policy perspective, the ability to hold political ends at risk is a crucial point, because doing so stands by far the best chance of fracturing the global terrorist network, one of the most important objectives of counterterrorism policy. Policymakers should be sensitive to this central objective of grand strategy, namely, preventing terrorist adversaries from cooperating with one another. This article has six main sections. In the rst, we dene deterrence in the context of the interactions between states and nonstate actors and examine why critics believe that it is an ineffective means of counterterrorism. The next three sections address each of the purported impediments to deterring terrorists. In the section on terrorist motivation, we develop a framework based on terrorist goals and levels of motivation that claries the strategies available to states to associate costs and benets with courses of action of different types of groups. In the fth section, we illustrate the effectiveness of coercion even against highly motivated groups by analyzing the results of its use against elements of terrorist networks in the Southern Philippines. We argue that the current approach of the U.S. and Philippine governments vis--vis the Moro Islamic Liberation Front (MILF)accommodating some of the groups political goals and then holding that accommodation at risk to prevent the MILF from cooperating with al-Qaida and Jemaah Islamiahis the best means of achieving the principal U.S. objective of denying all forms of support to groups intent on mass casualty attacks against the United States. In the concluding section, we apply our theoretical framework to current U.S. efforts to counter global terrorism and provide policy recommendations based on our ndings.

The Meaning of Deterrence in the Context of Counterterrorism

Deterrence approaches are only one of several classes of strategies for countering terrorism. Other strategies include persuasion (or winning hearts and minds), economic aid and democratization, appeasement, and military force.10 A deterrence strategy, by contrast, consists of the following two ele10. Our focus on deterrence does not imply that other strategies do not have important roles to play as well. On the contrary, the realm of ideas will be a key battleground over the long run. On persuasion, see, for example, Helena K. Finn, The Case for Cultural Diplomacy: Engaging For-

Contending with Terrorism 232

ments: (1) a threat or action designed to increase an adversarys perceived costs of engaging in particular behavior, and (2) an implicit or explicit offer of an alternative state of affairs if the adversary refrains from that behavior.11 Additionally, to be called a deterrence strategy, this increase in the adversarys perceived costs must be the result of costs imposed, at least in some contingencies, by the deterrer itself.12 This denition of deterrence subsumes what Glenn Snyder has called deterrence by punishment and deterrence by denial.13 Generalizing these concepts so that they apply to the interactions between state and nonstate actors as well as to interactions among states, we take deterrence by punishment to
eign Audiences, Foreign Affairs, Vol. 82, No. 6 (November/December 2003), pp. 1520; Peter G. Peterson, Diplomacy and the War on Terrorism, Foreign Affairs, Vol. 81, No. 5 (September/ October 2002), pp. 7496; and Michael Mousseau, Market Civilization and Its Clash with Terror, International Security, Vol. 27, No. 3 (Winter 2002/03), pp. 529. On economic aid and democratization, see Carol Graham, Can Foreign Aid Help Stop Terrorism? Not with Magic Bullets, Brookings Review, Vol. 20, No. 3 (Summer 2002), pp. 2832; Ivo H. Daalder and James M. Lindsay, Nasty, Brutish, and Long: Americas War on Terrorism, Current History, December 2001, pp. 403 408; and Alan B. Krueger and Jitka Maleckova, The Economics and the Education of Suicide Bombers: Does Poverty Cause Terrorism? New Republic, June 20, 2002, pp. 2733. 11. This second element denes the magnitude of the political objectives sought by the coercing state. Other works that also address this issue include Barry M. Blechman and Tamara S. Cofman, Dening Moment: The Threat and Use of Force in American Foreign Policy, Political Science Quarterly, Vol. 114, No. 1 (Spring 1999), pp. 130; and Alexander L. George, Forceful Persuasion: Coercive Diplomacy as an Alternative to War (Washington, D.C.: United States Institute of Peace, 1991). 12. This last requirement distinguishes deterrence from other forms of diplomacy, such as persuasion, where no costs are imposed. Although we regret the need for yet another denition of deterrence, existing denitions are inadequate because they do not encompass both deterrence by punishment and deterrence by denial; they are too closely tied to the interstate context; or they do not clearly distinguish deterrence from persuasion. For classic writings on coercion and deterrence, see Thomas C. Schelling, The Strategy of Conict (New York: Oxford University Press, 1963); Thomas C. Schelling, Arms and Inuence (New Haven, Conn.: Yale University Press, 1966); Glenn H. Snyder, Deterrence and Defense: Toward a Theory of National Security (Princeton, N.J.: Princeton University Press, 1961); Alexander L. George and Richard Smoke, Deterrence in American Foreign Policy: Theory and Practice (New York: Columbia University Press, 1974); Patrick M. Morgan, Deterrence: A Conceptual Analysis (Beverly Hills, Calif.: Sage, 1977); Robert Jervis, Richard Ned Lebow, and Janice Gross Stein, Psychology and Deterrence (Baltimore, Md.: Johns Hopkins University Press, 1985); John J. Mearsheimer, Conventional Deterrence (Ithaca, N.Y.: Cornell University Press, 1983); and Herman Kahn, On Escalation: Metaphors and Scenarios (New York: Frederick A. Praeger, 1965). Recent works on coercion include Robert A. Pape, Bombing to Win: Air Power and Coercion in War (Ithaca, N.Y.: Cornell University Press, 1996); and Daniel Byman, The Dynamics of Coercion: American Foreign Policy and the Limits of Military Might (New York: Cambridge University Press, 2002). 13. Snyder, Deterrence and Defense, pp. 1416. Thomas Schelling, Arms and Inuence, pp. 7071, contrasts deterrence (the threat to take hostile action if the adversary acts) with compellence (the threat to take hostile action unless the adversary acts). Because of the close relationship between these two terms, much of what we say about deterrence by punishment can be applied to compellence as well. See Patrick M. Morgan, Deterrence Now (New York: Cambridge University Press, 2003), p. 3; and Robert J. Art, To What Ends Military Power? International Security, Vol. 4, No. 4 (Spring 1989), pp. 335.

Deterring Terrorism 233

refer to the threat of harming something the adversary values if it takes an undesired action. Such a threatened trigger of punishment might be a terrorist attack, but it might also be an action believed to be a precursor to an attack.14 Deterrence by denial involves hardening targets in the hope of making an attack on them too costly to be tried and convincing terrorists of the states determination not to make concessions in the face of terror tactics. Thus, it is generally true that where punishment seeks to coerce the enemy through fear, denial depends on causing hopelessness.15 For both deterrence by punishment and deterrence by denial strategies to be successful, two conditions must hold: the threatened party must understand the (implicit or explicit) threat, and decisionmaking by the adversary must be sufciently inuenced by calculations of costs and benets. Because terrorists can appear fanatical, some analysts believe that, in general, neither of these conditions can be met. We term this the problem of irrationality. Deterrence by punishment also requires that several additional conditions hold: the deterrer must be able to hold something the adversary values at risk; the adversary must value what is held at risk over the expected value of taking action; and both the threat of retaliation and the deterrers promise not to take action if its conditions are met must be credible. These conditions depend on the capabilities of the two sides; the deterrer must be able to carry out its threat. They also depend on the deterrers having an incentive to follow through; the deterrer must not be made worse off by carrying out a threat than if it had simply not responded to the provocation.16 Thus, the conditions for deterrence by punishment strategies depend on the existence of a state of the world that both sides prefer to the state in which the deterrer takes action against the adversary and the adversary responds as best it can, often by doing
14. According to Glenn Snyder, deterrence by punishment primarily affects the aggressors estimate of possible costs, rather than [the aggressors] estimate of the probability of gaining his objective. Snyder, Deterrence and Defense, p. 15. A question thus arises: How should threats to take offensive action against groups in response to precursors to terrorist acts be classied? Although such threats may primarily inuence terrorists calculations of the probability of attaining an objective later on, we call this deterrence by punishment because it involves a choice of triggers for actions whose credibility must be demonstrated. 15. David Johnson, Karl Mueller, and William Taft V, Conventional Coercion across the Spectrum of Conventional Operations: The Utility of U.S. Military Forces in the Emerging Security Environment (Santa Monica, Calif.: RAND, 2002), http:/ /www.rand.org/publications/MR/ MR1494/, pp. 1617. Defense is distinct from deterrence by denial because it merely aims at reducing . . . costs and risks in the event deterrence fails, rather than inuencing the behavior of the adversary. Snyder, Deterrence and Defense, p. 3. 16. This last condition must hold over the long run. Deterrence is consistent with, and often requires, short-term sacrice.

Contending with Terrorism 234

its worst against the deterrer. This in turn requires that there be some overlap in the preferences of both sides over states of the world. If their preferences are precisely opposed, deterrence is impossible.17 As Thomas Schelling puts it, If his pain were our greatest delight and our satisfaction his greatest woe, we would just proceed to hurt and frustrate each other.18 Viewed in this way, deterrence resembles a bargain: both sides agree to cooperate on a state of affairs that both prefer to alternatives. Deterrence, therefore, is not just about making threats; it is also about making offers. Deterrence by punishment is about nding the right combination of threat and offer that meets the conditions listed here.19 In the case of state-terrorist interaction, these conditions seem difcult or impossible to meet. Because of their ideological and religious beliefs, many terrorists place extreme value on their political objectives relative to other ends (e.g., life and property). For this reason, it appears impossible that a deterrer could hold at risk something of sufcient value to terrorists such that their behavior would be affected. Similarly, deterrence by denial strategies seem destined to fail for the same reason, because they require that terrorists prefer the status quo to taking action given the dangers. Put differently, if the terrorists motivation is high enough, then even a small probability of a successful operation and a high probability of punishment will not deter them. Further, because the interests of terrorists and states seem so opposed, it also appears impossible that the two sides could agree on a state of affairs that both prefer

17. When preferences are nearly opposite, relations are adversarial, but when preferences are precisely opposed, relations are zero sum. See Robert Axelrod, Conict of Interest: A Theory of Divergent Goals with Applications to Politics (Chicago: Markham, 1970). 18. Schelling, Arms and Inuence, p. 4. 19. Whether in a terrorist context or not, this highlights the need for the second element in our denition of deterrence. When analysts have attempted to study threats but ignored the offer implicit in any threat, they have been led astray. To take one example from the Cold War: Would additional nuclear capabilities be more likely to deter a rst strike? When we consider only the severity and credibility of the threat, the answer appears to be that at least additional capability can do no harm. When we also consider the credibility of the offer in the event the adversary complies, however, our conclusions may change radically. Increased capability may give the deterrer a greater incentive to strike rst, decreasing the credibility of the offer not to attack. This in turn may give the adversary an increased incentive to strike rst itself, reversing our original conclusion that increased capabilities would aid in deterring a rst strike. See Snyder, Deterrence and Defense, p. 10. For related literature on provocation and deterrence, see, for example, Richard Ned Lebow, Provocative Deterrence: A New Look at the Cuban Missile Crisis, Arms Control Today, Vol. 18, No. 6 (July/August 1988), pp. 1516; Janice Gross Stein, Reassurance in International Conict Management, Political Science Quarterly, Vol. 106, No. 3 (Autumn 1991), pp. 431451; and Robert Jervis, Rational Deterrence: Theory and Evidence, World Politics, Vol. 41, No. 2 (January 1989), pp. 183207.

Deterring Terrorism 235

to that in which each does its worst against the other. We call these issues the problem of terrorist motivation. Even if this problem were solved, and targets for retaliation valued by terrorists discovered, a practical problem for deterrence by punishment strategies seems to remain. The capability to impose sufcient costs such that terrorists are deterred may require the ability to nd the members of the terrorist group responsible. We call this the return address problem. Other doubts about the efcacy of deterrence strategies are still subject to ongoing debate. Even during the Cold War, analysts pointed out numerous difculties with deterrence strategies, as well as a variety of factors that could lead to deterrence failure.20 Although we disagree with the stronger claims against the efcacy of conventional deterrence, we do not try to resolve these debates here.21 Instead, we focus on the three arguments that pertain particularly to deterrence in the terrorist context.22

The Problem of Irrationality

The assertion that terrorists are highly irrational is contradicted by a growing body of literature that shows that terrorist groups (though not necessarily every individual who engages in terrorist activities) usually have a set of hier20. As Edward Rhodes wrote in his review of empirical studies of conventional deterrence, The nature of conventional deterrence is that it will regularly fail, even in cases where commitments to respond are clearly dened, repeatedly publicized, and defensible, and the committed state [gives] every indication of its intention to defend them by force if necessary. Richard Ned Lebow, Conclusions, in Jervis, Lebow, and Stein, Psychology and Deterrence, p. 211, quoted in Rhodes, Can the United States Deter Iraqi Aggression? The Problems of Conventional Deterrence, 2002, Rutgers University, p. 4. See also Lebow, Miscalculation in the South Atlantic: The Origins of the Falklands War, in Jervis, Lebow, and Stein, Psychology and Deterrence, pp. 89124; Lebow, The Deterrence Deadlock: Is There a Way Out? in Jervis, Lebow, and Stein, Psychology and Deterrence, pp. 180202; and Joseph Lepgold, Hypotheses on Vulnerability: Are Terrorists and Drug Trafckers Coercible? in Lawrence Freedman, ed., Strategic Coercion: Concepts and Cases (New York: Oxford University Press, 1998). 21. For a review of methodological difculties that have beset past studies of deterrence, see Christopher Achen and Duncan Snidal, Rational Deterrence Theory and Comparative Case Studies, World Politics, Vol. 41, No. 2 (January 1989), pp. 143169; James Fearon, Signaling versus the Balance of Power and Interests, Journal of Conict Resolution, Vol. 38, No. 2 (June 1994), pp. 236269; Curtis S. Signorino and Ahmer Tarar, A Unied Theory and Test of Extended Immediate Deterrence, working paper, University of Rochester, December 7, 2001, http:/ /www .rochester.edu/college/psc/signorino/papers/SigDeter.pdf; and Bear F. Braumoeller, Causal Complexity and the Study of Politics, Political Analysis, Vol. 11, No. 3 (Summer 2003), pp. 209233. 22. Throughout we also focus attention on deterrence strategies that target classes of individuals who are essential for the functioning of a terrorist group as a whole. Deterring specic individuals retards the capacity of groups only when the supply of individuals willing to ll their roles is a constraining factor. In contemporary terrorist networks, this is often true to only a limited degree.

Contending with Terrorism 236

archically ordered goals and choose strategies that best advance them.23 The resort to terror tactics is itself a strategic choice of weaker actors with no other means of furthering their cause.24 Suicide tactics in particular, as Pape shows, are practiced in the context of coercive campaigns and were adopted because they proved to be remarkably successful for coercing liberal democracies.25 In addition, terrorist groups have often put interest ahead of strictly interpreted ideology, for instance, in cooperating with groups and states with opposed beliefs.26 Therefore, even though terrorist decisionmaking processes are certain to consist of both rational and nonrational elements, this is neither peculiar to terrorists nor precludes deterrence. Deterrence requires only that terrorists be sufciently inuenced by cost-benet calculations. As Robert Jervis argued more than a quarter century ago, Much less than full rationality is needed for the main lines of [deterrence] theory to be valid.27

The Problem of Terrorist Motivation

The issue of terrorist motivation is the most serious difculty facing powerful states attempting to implement deterrence strategies. To address this issue, we have developed a framework that species the types of deterrence strategies that can be effective against particular classes of groups and elements of terrorist networks.28 This framework is represented in Figure 1, where the intensity of the terrorists motivation is on the vertical axis.29 We dene motivation as
23. See Martha Crenshaw, The Causes of Terrorism, in Charles W. Kegley, ed., International Terrorism: Characteristics, Causes, Controls (New York: St. Martins, 1990), p. 117; Betts, The Soft Underbelly of American Primacy; Robert A. Pape, The Strategic Logic of Suicide Terrorism, American Political Science Review, Vol. 97, No. 3 (August 2003), pp. 343361; Jonathan Schachter, The Eye of the Believer: Psychological Inuences on Counter-terrorism Policy-Making (Santa Monica, Calif.: RAND, 2002), p. 96; and Andrew Kydd and Barbara Walter, The Politics of Extremist Violence, International Organization, Vol. 56, No. 2 (Spring 2002), pp. 279289. 24. Betts, The Soft Underbelly of American Primacy. 25. Pape, Dying to Win, pp. 4445. 26. Consider, for instance, al-Qaidas decision before the 2003 Iraq war to cooperate with Iraqi Baathists against the United States. 27. Robert Jervis, Deterrence Theory Revisited, World Politics, Vol. 31, No. 2 (January 1979), p. 299. 28. Other authors have suggested a variety of taxonomies of terrorist groups, based on different criteria, that may prove useful for particular applications. In general, experts have identied the following types of terrorism: nationalist, religious, anarchist, state sponsored, left wing, and right wing. 29. We have drawn four categories, but each axis is better thought of as representing a continuum. The more motivated the group, for instance, the less susceptible it will be to the deterrence strategies listed in quadrants 1 and 2.

Deterring Terrorism 237

Figure 1.

Potential Deterrence Strategies Based on the Intensity of Terrorist Motivation and the Similarity of Preferences over Outcomes

Some terrorist groups have objectives that could be at least partially accommodated either by the deterring state or by actors over whom the deterring state has leverage. In this sense, the relationship is not zero sum. b Temporary deterrence implies that groups can be influenced to refrain from taking action while they build capability for larger strikes. This is sometimes to the advantage of the deterrer because it provides a greater window of opportunity for the use of offensive strategies against the group.

the extent to which terrorists value their political goals over nonpolitical ends. Examples of the latter may include life, liberty, property, and social standing (when not derived directly from terrorist activity). The degree to which terrorist groups have some political goals that could be accommodated by deterring states is on the horizontal axis. We rst discuss the potential of deterrence by punishment strategies targeted at less motivated elements of terrorist networks at stages in the terror process when they are most susceptible to inuence. This corresponds to quadrants 1 and 2 in Figure 1. We show that by considering only the possibility of deterring a suicide terrorist the moment before he (or, more rarely, she) commits the act, analysts have overlooked deterrence strategies that could prove effective. We then consider the possibility of deterring the most highly motivated terrorists, that is, those willing to run any risk in pursuit of their

Contending with Terrorism 238

goals (quadrants 3 and 4). We show that by holding at risk political ends, states can deter such terrorists from certain courses of action. We then argue that despite the problem of terrorist motivation, deterrence by denial strategies can be effective against all classes of terrorist groups, and conclude the section by applying the theoretical framework developed here to the issue of deterring multiple groups at once. deterring less motivated elements of terrorist networks The higher the value terrorists place on what is not gained through terror relative to what is, the more compelling is a threat against the former, and the more likely deterrence will succeed. It is less obvious that, through careful attention to the many different elements of terrorist systems and an understanding of the processes that lead to attacks, even apparently highly motivated groups may be susceptible to the deterrence strategies listed in quadrants 1 and 2. To produce a large-scale attack, terrorists must constitute a system of actors fullling specic functional roles. As Paul Davis and Brian Jenkins characterize it, such a system comprises leaders, lieutenants, nanciers, logisticians and other facilitators, foot soldiers, supporting population segments, and religious or otherwise ideological gures.30 Some elements of the terrorist system are more difcult to deter than others. Financiers, for example, are sometimes less fanatically motivated than other elements of the system and thus easier to deter. Although states often have difculty tracking them down,31 they have had success when making this a priority.32 The greatest difculties are often political rather than a matter of nding perpetrators, and the resolution of political difculties usually depends on the level of diplomatic resources that states are willing to commit. Although nding all nanciers of all groups prior to terrorist attacks is impossible, once large-scale attacks are committed, it is sometimes possible to con-

30. Davis and Jenkins, Deterrence and Inuence in Counterterrorism, p. xi. 31. See, for example, Louise Branson, Cutting Off Terrorists Funds, Straits Times, September 26, 2001. 32. For instance, the U.S. government has closed down several important sources of terrorist nancing since the attacks of September 11, 2001; and in November 2002, the CIA together with nancial investigators announced that they had traced tens of millions of dollars owing to alQaida, mostly from Saudi sources. See, for example, U.S.: Al Qaida Funded by Only 12 Individuals, Most Saudis, WorldTribune.com, October 20, 2002, http:/ /www.hvk.org/articles/1002/194 .html. See also The 9/11 Commission Report: Final Report of the National Commission on Terrorist Attacks upon the United States (Washington, D.C.: U.S. Government Printing Ofce, July 2004), p. 382.

Deterring Terrorism 239

centrate investigative resources and uncover the source of nancing in particular cases. A demonstrated policy of committing signicant resources to nd and punish nanciers may therefore deter an essential part of the system from engaging in terrorist activity. This example highlights an underrecognized contrast between brute force and deterrence strategies. When a government weighs the benets of using force against the costs of diplomatic and material concessions to states whose assistance is required to punish terrorist militants and nanciers, it may decide that the price is too high. The possibility of deterring future terrorists, however, provides a strong additional incentive. Thus, a deterrence approach implies an even more aggressive policy than a brute force approach if deterrence is unsuccessful in a particular case. State sponsors represent another element of terrorist systems that many view as less motivated and easier to nd, and therefore susceptible to deterrence.33 Scholars and policymakers who are skeptical of using deterrence against terrorists often believe that, on the contrary, their state sponsors are deterrable. The Bush administrations National Strategy for Combating Terrorism contains a long discussion of the administrations policy of deterring state sponsors of terrorism, though it makes no other explicit reference to a deterrence approach.34 Other scholars argue, however, that failing states may be highly motivated to sell their capabilities and provide other assistance for nancial gain. Nevertheless, because the response of a powerful state to a terrorist attack will likely be proportional to the scale of the attack, even highly motivated potential state sponsors with advanced capabilities and other countercoercive instruments will be forced to exercise restraint. The capabilities of state sponsors may enable them to avoid being deterred from supporting smaller-scale international terrorism, but powerful states will likely retain the ability to deter would-be state sponsors from supporting larger-scale attacks.35 In addition to paying attention to the diverse elements that make up terrorist systems, as Michael Powers has argued in the context of WMD terrorism,
33. Currently Cuba, Iran, Libya, North Korea, Sudan, and Syria are on the U.S. list of states that sponsor terrorism. Ofce of the Coordinator for Counterterrorism, U.S. Department of State, Country Reports on Terrorism, 2004 (Washington, D.C.: U.S. Government Printing Ofce, 2005), pp. 8890. 34. National Strategy for Combating Terrorism, (Washington, D.C.: U.S. Government Printing Ofce, February 14, 2003), http:/ /www.whitehouse.gov/news/releases/2003/02/counter_terrorism/ counter_terrorism_strategy.pdf. 35. We are not implying that all terrorist groups have state sponsors or that the Bush administrations focus on states over nonstate actors is appropriate.

Contending with Terrorism 240

deterrers should think of terrorist activity as a process, or series of actions culminating in violence, rather than a single act or event such as the terrorist attacks of September 11, 2001.36 Consider, for example, the process that culminated in those attacks. As early as 1996, Mohammed Atta began planning and recruiting for them in Hamburg. Over the next ve years, he and his associates arranged for nancing, visas, accommodation, and ight lessons; selected targets; continued to recruit; and ultimately carried out the attacks.37 Each stage presents an opportunity for detection by intelligence networks and law enforcement or for a military response. When, at a particular stage in the process leading to an attack, the risks of detection and punishment outweigh the benets of a successful attack multiplied by the probability of success, the terrorist or terrorist supporter will be deterred. Deterrence is possible when the benets of a successful attack are not too high, as in quadrants 1 and 2 of Figure 1. In analyzing the decision calculus of elements of terrorist systems at each point in the process, we can see the usefulness of developing a deterrence strategy that sets triggers for action that occur early in that process. Even though the early stages are often the most difcult to detect, the punishment need not be as severe, nor the probability of detection as high, to have an equivalent deterrent effect. This is because, in the early stages, the prospects of achieving a successful attack are more uncertain. Even those willing to give their lives when the success of an attack is assured may be unwilling to begin a process that may not, in the end, advance their cause. Thus, to deter terrorists, states should search for less motivated or more visible elements of terrorist networks and threaten retaliation at early or more easily detectable stages of the terror process. deterring terrorists whose goals can be partially accommodated In some cases, terrorists are so motivated that deterrence by punishment strategies that target the nonpolitical ends of terrorists are insufcient. Nevertheless, overlap in the preferences of terrorists and deterring states can still create a range of agreements that deterrence strategies can enforce (see quadrant 3). This is a type of deterrence by punishment where political ends are held at risk. Whereas the previous discussion emphasized the potential of the deterrer
36. Michael J. Powers, Deterring Terrorism with CBRN Weapons: Developing a Conceptual Framework, occasional paper no. 2, CBACI (February 2001), http:/ /knxas1.hsdl.org/homesec/ docs/nonprof/nps08-091704-01.pdf, p. 5. 37. See, for example, Al-Jazeera Offers Accounts of 9/11 Planning, CNN.com, September 12, 2002, http:/ /www.cnn.com/2002/WORLD/meast/09/12/alqaeda.911.claim.

Deterring Terrorism 241

to harm members of the terrorist group, here we also consider the alternative to the threat, that is, the deterring states offer made to the group in the event it refrains from the undesired behavior. Terrorists usually have a range of objectives, some dearer than others. States also have preferences over these same objectives. When the preference orderings of terrorists and states are precisely opposed, deterrence is impossible.38 Even a small overlap in the preferences of the two sides, however, can be exploited in the cause of deterrence. We highlight this possibility by focusing on two areas of common ground. First, many terrorist organizations with global objectives have local concerns even closer to heart. In fact, some organizations merely advance the agenda of other groups in return for resources and expertise they can apply locally. For instance, three members of the Irish Republican Army (IRA) arrested in Colombia in 2001 were suspected of training the Revolutionary Armed Forces of Colombia (FARC) in how to conduct an urban bombing campaign.39 It is unlikely that the IRA had taken up the cause of Marxist insurgency in Colombia. Rather, some trade must have occurred between the IRA and the FARC, such that the interests of both were furthered. In the jargon of the U.S. National Strategy for Combating Terrorism, the two must have achieved synergies. Terrorist preferences of this sort are represented in the upper half of Figure 2. Such terrorists would like to cooperate with groups intent on striking the deterring state, or they might like to strike the deterring state themselves. There are several reasons why this might be the case: a strike might further the goals of groups directly; the response of the deterring state might be thought likely to galvanize support for their cause; or assistance on the part of one group might have been traded for another form of assistance from the other. The group whose preferences are represented in Figure 2, however, is even more interested in advancing a local agenda through acts targeted at a domestic audience. Of the forty-two foreign terrorist organizations (FTOs) currently designated by the U.S. Department of State, the vast majority fall into this category to a greater or lesser degree.40 The terrorists local agenda may have nothing in common with the foreign policy goals of the deterring state, but if some aspects of this agenda can be

38. We discuss one partial exception to this rule below, which we term temporary deterrence. 39. National Strategy for Combatting Terrorism, p. 8. 40. See Ofce of Counterterrorism, U.S. Department of State, Fact Sheet: Foreign Terrorist Organizations, March 23, 2005, http:/ /www.state.gov/s/ct/rls/fs/37191.htm.

Figure 2. An Example of Overlapping Preferences between Terrorists and a Deterring State

Contending with Terrorism 242

Deterring Terrorism 243

even partially accommodated, there is overlap between the preferences of the two sides. This presents the state with strategic opportunity. Consider, for instance, the interests of the United States vis--vis terrorist groups such as Sendero Luminoso (Shining Path), the Liberation Tigers of Tamil Eelam, Hamas, and the Basque Fatherland and Liberty. While the interaction between these groups and the local states may be zero sumones pain is strictly the others gaintheir relationship with the United States may not be.41 Each of these groups may be able to achieve synergies through cooperation with other transnational groups, but they may prefer not to cooperate with such groups if that induces the U.S. government to refrain from devoting signicant resources to intervening in their local conicts. The United States, in turn, may prefer not to devote signicant resources to targeting groups that do not cooperate with those it considers most threatening. Thus, if the local agenda does not sufciently conict with the interests of the deterring state, the local interests of the terrorist group can serve as an effective hostage for a policy of deterrence. Often, the deterring state can threaten to tip the scales in the local conict. Terrorist groups whose primary concern is the local theater may be willing to refrain from certain actions (e.g., cooperating with groups considered more dangerous by the deterring state) in return for less interference by the deterring state in the local conict. In such cases, terrorists would be coerced into courses of action not just out of fear for their lives and property, but also out of fear for their cause. This in turn implies that designing an effective strategy for combating terrorism requires an indepth understanding of terrorist adversaries, not only their capabilities and intentions toward a particular state but also their stakes in other congurations. If, instead, force is actually used by a state against the terrorist group, this can create a harmony of interest between the group and more dangerous terrorist organizations or even change the preferences of the group such that local concerns seem less important. If the local agenda is put out of reach, members of the group may even turn their focus to international terrorism. More important, by using force against terrorist groups, states give these groups every incentive to cooperate with other groups and organizations whose interests are similarly opposed by the target state. This is a dangerous possibility because local terrorist groups can provide intelligence, personnel, expertise, resources, and safe havens to groups and individuals who may be even more
41. For a discussion of the bargaining relationship between terrorist groups and local states, see Paul Pillar, Terrorism and U.S. Foreign Policy (Washington, D.C.: Brookings, 2001), pp. 145148.

Contending with Terrorism 244

threatening.42 A policy that makes action by the deterring state contingent on the creation of such links would have the opposite effect. Rather than creating a harmony of interests among terrorist groups opposed to the deterrer, such a policy would help to fracture the global terrorist network. In fact, in addition to being deterred from cooperating, some groups might even be coerced into providing local intelligence on other groups, as in the case of the MILF discussed below. A second example of overlapping preferences that can produce a deterrence equilibrium occurs when both sides prefer bounding the scope of violence to the state of affairs when each side does its worst against the other. Sometimes, by tacitly permitting smaller-scale attacks, or those of a particular type, a state can deter those of larger scale, or of an alternative variety. Suppose a state can credibly threaten to inict substantial damage on the capabilities of the terrorist group, in the event the group carries out an attack. If the state makes such a threat, so that its actions are contingent on the actions of the terrorist group, what would be the likely response of the group? If the threat is sufciently credible, and if lower levels of violence also advance the groups aims but do not elicit such a strong response from the target, then the groups best option would be to moderate the destruction it causes. The group would be partially deterred, not for fear of losing life or property, but for fear of losing the ability to prosecute its cause altogether.43 The conict between the Israeli Defense Forces (IDF) and the South Lebanon Army (SLA), on the one hand, and the Lebanese liberation movement Hezbollah, on the other, may provide an example of these dynamics.44 Throughout the rst half of the 1990s, both sides retaliated following successful attacks on their forces by targeting civilians identied with the opposing

42. National Strategy for Combatting Terrorism, p. 9. 43. This is similar to the argument in the literature on intrawar deterrence and limited war that, even in intense conicts, states may limit the severity of their attacks to avoid provoking retaliation in kind. On the abstention of both sides from using chemical weapons in World War II, see, for example, Jeffrey W. Legro, Cooperation under Fire: Anglo-German Restraint during World War II (Ithaca, N.Y.: Cornell University Press, 1995), chap. 4, especially p. 201. On the geographical limits on ghting during the Korean War, see Stephen J. Cimbala, Military Persuasion in War and Policy: The Power of Soft (New York: Praeger, 2002), pp. 100101. 44. See Clive Jones, Israeli Counter-insurgency Strategy and the War in South Lebanon, 1985 97, Small Wars and Insurgencies, Vol. 8, No. 3 (Winter 1997), pp. 82108; Daniel Sobelman, Hizbollah Two Years after the Withdrawal: A Compromise between Ideology, Interests, and Exigencies, Strategic Assessment, Vol. 5, No. 2 (August 2002), http:/ /www.tau.ac.il/jcss/sa/ v5n2p4Sob.html; and Augustus Richard Norton, Hizballah and the Israeli Withdrawal from Southern Lebanon, Journal of Palestinian Studies, Vol. 30, No. 1 (Autumn 2000), pp. 2235.

Deterring Terrorism 245

side. In 1994 Hezbollah even attacked Jewish targets in London and Buenos Aires. In 1996 Israel hoped to use force to break Hezbollah, in the words of an Israeli general, in an extensive operation code-named Grapes of Wrath.45 As a result of Israeli air and artillery bombardment, 400,000 Lebanese ed north, creating a massive refugee problem for the Lebanese government. Many Lebanese civilians were also killed, including 102 sheltered in a United Nations compound. Israels attacks put pressure on Hezbollah through several channels: group members were killed during the operation; refugees created difculties for the Lebanese government, which in turn pressured Hezbollah to make concessions to end the violence (or pressured Damascus to pressure Hezbollah); there was also a danger that the Lebanese population would cease its support of Hezbollah to halt the violence. As a strategy of brute force, the operation was a failure, as hundreds of Katyusha rockets launched from Southern Lebanon continued to fall on Northern Israel throughout the conict. The operation also failed to deter Hezbollah from future actions against Israeli interests in Southern Lebanon. Thus, because of Hezbollahs high level of motivation, Israeli coercion was insufcient to compel Hezbollah to give up its objectives entirely. The escalating cycle of violence was broken, however, by an agreement in 1996, brokered by former U.S. Secretary of State Warren Christopher. In it, Hezbollah agreed to refrain from attacking targets inside Israel, and the IDF and SLA committed to refrain from attacking Lebanese civilians. This agreement appears to have been generally successful in bounding the scope of the conict to the present day. Thus, although Hezbollah could not be deterred from pressing its interests by violent means, it was deterred from terrorist activities within Israel. A limited strategy of this sort may be optimal for states that have the capability to retaliate effectively but also have other pressing uses for intelligence and operational and other material resources, or whose retaliation would be associated with extreme costs, such as inaming resentment against the state. This approach might be the best strategy against some of the terrorist groups that Ian Lustick has labeled solipsistic.46 Such groups use terror partly or primarily as a means of affecting the behavior of groups and individuals with whom the perpetrators identify. Thus, because their use of terror is not meant
45. Eitam Rabin, Interview with IDF Major-General Amiram Levine, OC Northern Command, Haaretz, August 7, 1996. 46. Ian Lustick, Terrorism in the Arab-Israeli Conict: Targets and Audiences, in Martha Crenshaw, ed., Terrorism in Context (State College: Pennsylvania State University Press, 1995), pp. 514552.

Contending with Terrorism 246

to coerce, lower levels of violence may sufciently advance their objectives, and they may not wish to incite the retaliation that a large-scale attack would bring.47 The sort of preference overlap represented in Figure 2 is also a plausible description of the state of affairs for the United States and other nations vis--vis certain terrorist groups prior to September 11. Consistent with this interpretation, the U.S. response to the 2,400 anti-U.S. terrorist incidents from 1983 to 1998 was fairly moderate: the United States retaliated militarily only three times.48 In the 1990s, the United States experienced a series of terrorist attacks, some of which were allegedly tied to Osama bin Laden and al-Qaida. Among these were the attack on the USS Cole off the coast of Yemen in 2000, the Khobar Towers bombing in Saudi Arabia in 1996, and the rst World Trade Center bombing in 1993. The U.S. response focused primarily on pursuing the individuals directly involved in the attacks rather than the group responsible. Although U.S. actions may have been consistent with a policy of deterring more serious threats, they may also have been interpreted as a sign of unwillingness to bear the costs of a more vigorous response. Bin Laden may have interpreted U.S. withdrawals from Beirut in 1983, following the bombings of the U.S. embassy and marine barracks, and from Somalia in 1993, six months after eighteen U.S. soldiers were killed in a erce reght in Mogadishu, in this way.49 If this interpretation is correct, it may be that deterrers must demonstrate capability and resolve before deterrence can function effectively. The implications of an adversary developing a perception that a deterrer lacks the willingness to respond are serious because, once established, such reputations are difcult to change.50 From a theoretical perspective, the discussion here highlights the importance of the offer made to the challenging state or terrorist group if it complies with the deterrers demands, and the possibility of using leverage over one po47. Determining which groups should be classied as solipsistic requires intensive analysis. Ian Lustick, Terrorism and the Arab-Israeli Conict, pp. 514552, convincingly argues that Palestinian violence in the 1970s and Zionist violence in the 1940s deserve this title. 48. Michele Malvesti, Explaining the United States Decision to Strike Back at Terrorists, Terrorism and Political Violence, Vol. 13, No. 2 (Summer 2001), pp. 85106. 49. See Osama bin Laden, Declaration of War against the Americans Occupying the Land of the Two Holy Places, August 23, 1996; and John Miller, To Terrors Source, interview with Osama bin Laden, Afghanistan, ABC News, May 28, 1998. 50. As Robert Jervis argues, One of the basic ndings of cognitive psychology is that images change only slowly and are maintained in the face of discrepant information. This implies that trying to change a reputation of low resolve will be especially costly. Jervis, Deterrence and Perception, International Security, Vol. 7, No. 3 (Winter 1982/83), p. 9.

Deterring Terrorism 247

litical end for leverage over the actions a group takes in pursuit of another. By altering the offer, such as by refraining from intervening in local conicts or even tacitly permitting lower-level violence, states can wholly or partially deter even highly motivated groups. At times, therefore, deterring states should consider limiting their demands. Furthermore, certain sorts of accommodations are likely to be more effective than others. In particular, accommodations that can be held at risk serve the cause of deterrence. Those that do not may reduce terrorist grievances and therefore motivation, but they do not serve deterrence directly. As an example, consider a states decision to release captured militants. On the one hand, this strengthens the capabilities of the militants organization but does not address their core concerns. On the other hand, if the group is highly motivated, it is by denition undeterred by fear of capture. Thus, in many cases, the release of militants will not provide any leverage over the organizations behavior because this accommodation, unlike the two other examples given above, cannot be held at risk. Any accommodation also carries the risk that it will encourage other groups to demand similar treatment. The deterring state must therefore be clear in signaling the different approach it intends to use with different groups. Groups that fall into quadrant 1 in Figure 1 should not be treated like those that fall into quadrant 3, and the groups must be made to understand this. States will have to weigh the risks associated with (publicly observable) accommodation in individual cases.51 deterring terrorists with precisely opposed preferences Highly motivated terrorists whose preferences are precisely opposed to those of the deterring state cannot be deterred, though they can be inuenced.52 They cannot be deterred because their high level of motivation means that no matter what threat is leveled against them, they will always pursue their objectives, and because their preferences are precisely opposed, no bargaining space exists. Such terrorists cannot be made to refrain from taking hostile action. Different groups fall into this category for different states, but in general few groups belong in quadrant 4. From the perspective of the United States, because the vast majority of terrorist groups are primarily concerned with lo-

51. On the trade-offs between clarity and ambiguity in signaling, see Snyder, Deterrence and Defense, pp. 246249. 52. Davis and Jenkins make a similar point in Deterrence and Inuence in Counterterrorism, p. 9.

Contending with Terrorism 248

cal conicts, only certain parts of the al-Qaida network seem to t this classication. When states attempt to deter such groups, there is a signicant danger that deterrence may appear to succeed in the short run, lulling the state into complacency. If the state has even a chance of retaliating effectively, the terrorist group may have incentive to bide its time, building its capabilities in preparation for a more massive strike.53 The radicalization of terrorist discourse in the 1990s may therefore explain both the higher number of casualties per attack and the decrease in the number of individual attacks.54 Still, in some cases, such temporary deterrence can be useful if it provides the deterring state with time to apply offensive strategies. Unfortunately, the temporary lull in attacks is likely to be followed by attacks of greater severity. using deterrence by denial Deterrence by denial strategies have the potential to be effective against all four types of terrorist groups. By hardening targets (e.g., fortifying embassies, reinforcing cockpit doors, upgrading border security, and tightening immigration controls) and demonstrating resolve not to make concessions, the deterring state can lessen terrorists motivation by reducing the benets of terror tactics. Thus, although defensive strategies cannot protect every target, they can minimize the terrorists power to hurt, thereby lessening the coercive power of terrorist action. This in turn reduces terrorist motivation, increasing the effectiveness of many of the strategies described above. As some analysts have pointed out, one of the reasons for the 1998 attacks on U.S. embassies in Tanzania and Kenya was that terrorists believed that U.S. assets in Africa were easier targets compared with better-secured facilities in

53. Terrorists are unlikely to wait too long before striking because the process of building greater capability runs the risk of being discovered before it is utilized. A failed deterrence strategy, however, is consistent with low levels of terrorist activity in the near term. 54. On the radicalization of terrorist groups, see Rex A. Hudson, The Sociology and Psychology of Terrorism: Who Becomes a Terrorist and Why? (Washington, D.C.: Federal Research Division, Library of Congress, 1999), pp. 13; and Ian Lesser, Bruce Hoffman, John Arquilla, David Rondfeldt, Michelle Zanini, and Brian Michael Jenkins, Countering the New Terrorism (Santa Monica, Calif.: RAND, 1999), http:/ /www.rand.org/publications/MR/MR989/. On trends in international terrorism, see U.S. Department of State, Patterns of Global Terrorism: 2001, app. 1: Statistical Review, http:/ /www.state.gov/s/ct/rls/pgtrpt/2001/html/10265.htm; and Bruce Hoffman, The Conuence of International and Domestic Trends in Terrorism, working paper (Edinburgh: Center for the Study of Terrorism and Political Violence, St. Andrews University, 1996), http:/ / www.ciaonet.org/wps/hob01/.

Deterring Terrorism 249

the Middle East and elsewhere.55 This has been taken as a sign of the futility of defensive strategies, but in some ways it is just the opposite. Terrorist motivation, and with it terrorist actions, would likely increase if the terrorists were able to strike higher-value targets more easily. Thus, although defensive strategies are inadequate in themselves, they form an important component of a deterrence approach. deterring multiple groups at once During the Cold War, deterrence was mainly considered in the context of the interaction between the United States and the Soviet Union. Today, however, governments are faced with the challenge of deterring multiple groups at once. This presents both difculties and strategic opportunities. Investigating and targeting multiple groups in disparate areas of the globe that use different languages and operating procedures will severely tax the intelligence, diplomatic, administrative, and military resources of deterring states. States activities in different countries require separate negotiations with local authorities. These factors limit the ability of the deterrer to threaten focused retaliation against terrorist groups. In fact, a deterrence by punishment policy that would be successful against a single group may not be credible against multiple groups because of the resource constraints of the deterring state.56 Rather than attempting or threatening to use force against all terrorist groups at once, a more credible policy might produce the desired result by committing to focus resources only on those considered the most dangerous. Suppose there are two groups that a state wishes to deter. Both are highly motivated, but they do see intermediate levels of violence as furthering their cause. The deterring state has several options. First, it might threaten both of them, but the groups might realize that if they both attack, the states resources would not be sufcient to retaliate effectively against the two at once. Second, the state could threaten only one, in which case the other would be completely undeterred. A third option is more attractive, though not without practical difculties. The deterring state could threaten to concentrate its resources against whichever group shows itself to be more dangerous. Each group
55. Richard A. Falkenrath and Philip B. Heymann made a similar argument that soft targets in U.S. cities would be attacked if embassies were fortied. Falkenrath and Heymann, Wed Better Be Ready for an Escalation, Boston Globe, August 27, 1998. 56. For a discussion of the problem of deterring multiple adversaries, see Daniel S. Treisman, Rational Appeasement, International Organization, Vol. 58, No. 2 (April 2004), pp. 345373.

Contending with Terrorism 250

would then have an incentive to be slightly less violent than the other. To plan operations that the deterrer would perceive as less worthy of retaliation, each would have to guess the actions that the other group is likely to take. If the two groups thought of coordinating, they would have no reason to trust each other. Therefore, because each would be evaluating the likely actions of the other, and expecting to be evaluated in turn, both would be forced to signicantly moderate their behavior. In practice, policymakers resource allocation decisions require considerable judgment, and different terrorist groups must be dealt with in different ways. In attempting to deter multiple groups at once, the deterring state might decide to signal an intention to focus resources on several of the most dangerous groups. There may also be relatively low-cost actions, such as blocking assets and restricting the travel of individuals, that could be taken against many terrorist groups. But in other cases, effective measures against groups may require signicant intelligence, diplomatic, and special operations resources. When signaling dynamics allow, these resources should be reserved for a smaller set of groups, as described here. Inclusion on this short list should be as uid as possible so that even groups that are not on it are still under threat.

The Return Address Problem

Terrorists are usually difcult to nd, which reduces the degree of coercive leverage of certain sorts of threats. It does not follow from this, however, that attempts at deterrence by punishment will fail. Rather, this observation highlights the importance of matching the demands that states make of particular terrorist groups with the level of threat that can credibly be brought to bear against them. The question then is whether signicant demands can be made. We submit that they can be, despite terrorists lack of a return address. First, when states devote sufcient resources, they can nd members of terrorist organizations. Many terrorist groups operate partly or wholly out of known base areasfor instance, al-Qaida in Afghanistan (before the 2001 war), Abu Sayyaf on the Philippine islands of Basilan and Jolo, and the Mujahedin-e Khalq in Iraq. The armed forces of the Philippines, assisted by the United States, were able to nd and kill many Abu Sayyaf Group members. Other terrorist organizations, however, do not have large and identiable bases. Some have broad areas of operation, or they may be dispersed in particular populations. Examples include Aum Shinrikyo, Sendero Luminoso, and

Deterring Terrorism 251

al-Qaida after the Afghan war. These groups tend to be smaller and, because they are dispersed among the population, are more difcult to nd.57 Still, members of such groups have been found in the past and punished for their activities when states have made this a priority.58 A deterrence approach that reserves intelligence and other resources for use in cases of deterrence failure would only increase the ability of states to nd group members. Second, even though targeted states cannot nd every terrorist and terrorist facilitator, they almost always have the ability to increase the costs to terrorist groups of achieving their political goals.59 States therefore can put these political goals further from reach, decreasing the likelihood that the group will achieve any of its objectives. Some mechanisms for frustrating a groups political goals require that its members be found; others do not. Counterinsurgency and law enforcement operations are principal examples of the former. Examples of the latter include a deterring state providing economic and military aid to governments targeted by insurgents, pressuring targeted states not to make concessions to terrorists, aiding other groups with goals that are opposed to those of the terrorist group, and imposing travel and fund-raising restrictions on terrorist group members.60 Furthermore, uses of force such as counterinsurgency operations need not result in the capture of every terrorist to seriously harm both the political and
57. Al-Qaida is an exception to the rule that baseless groups are relatively small, partly because it did have a base in Afghanistan until the 2001 Afghan war. 58. The U.S. response following the hijacking of the Achille Lauro cruise ship in 1985 is instructive. Although the hijackers were found and captured, Italian authorities released the leader of the group responsible because the U.S. government was unwilling to pay the diplomatic costs of prevailing upon the Italians to act differently. Following the September 11 attacks, however, the United States began to dedicate more resources to this area and, as a result, has achieved some successes in the war on terror. See Federal Bureau of Investigation, War on Terrorism, http:/ / www.fbi.gov/terrorinfo/counterrorism/waronterrororhome.htm (accessed April 2003); and Al Qaedas Most Wanted, USAToday.com, March 2, 2003, http:/ /www.usatoday.com/news/world/ 2003-03-02-alqaeda-list_x.htm. According to President Bush, nearly half of al-Qaidas leadership had been captured or killed by May 1, 2003. See President Bush Announces Combat Operations in Iraq Have Ended, speech delivered on the USS Abraham Lincoln, May 1, 2003, http:/ /www .state.gov/p/nea/rls/rm/20203.htm. 59. Scholars who have doubted the efcacy of deterrence strategies have concurred in this. Robert Pape, for example, notes that military action can disrupt a terrorist groups activities temporarily. Pape, Dying to Win, p. 239. 60. Because terrorists depend on support from the communities they live in, it might also be possible to target the community as a whole, which would also not require nding individual terrorists. Individuals who terrorists hold dear, such as family members, might also be threatened. We do not explore these options further except to note that, in many cases, they stand a signicant chance of galvanizing support for the terrorists cause.

Contending with Terrorism 252

nonpolitical interests of the terrorist group. In fact, uses of force that fail to obtain a states political objectives may still hurt the interests of the group and its members. Once again, the Balikatan Operation against the Abu Sayyaf Group is a telling example. This point is particularly important because it is in cases where other strategies such as force cannot achieve the ends of states that a bargaining range exists. In these situations, the sides may agree to refrain from doing their worst against the other because terrorist-state interaction is not zero sum, making a deterrence equilibrium possible. Conversely, if deterrence strategies are ineffective in particular cases because terrorists cannot be found, the use of force will be ineffective as well. Therefore, deterrence strategies that threaten the use of force are productive when the effectiveness of force occupies a middle range, when force cannot easily achieve the ends of states but can at least disrupt terrorist group operations. Third, the difculty of nding terrorists poses a problem for deterrence only when their motivation is high, and high levels of motivation often make terrorists more susceptible to deterrence strategies that target political ends. Less motivated potential terrorists will be deterred just as less motivated potential criminals are deterred, even though the police cannot catch every one. Highly motivated terrorists, because they hold their political goals so dearly, are loath to run even lower-level risks to these goals. This magnies the coercive leverage of strategies that target political ends. A related argument made by some scholars is that because terrorists easily blend into the local population, collateral damage caused by attempts at retaliation against them inames hatred of the retaliating state and galvanizes support for the terrorists cause. Indeed, inciting such retaliation may be an explicit terrorist objective, so threatened retaliation would hardly deter.61 States have a variety of retaliatory options, however, and some of these are more narrowly focused on terrorists than others. Although lessons from the IsraeliPalestinian conict are difcult to draw, it seems intuitively clear that tactics that harm or kill large numbers of noncombatants have radicalized more mod61. For a similar argument, see David Lake, Rational Extremism: Understanding Terrorism in the 21st Century, Dialogue-IO, Spring 2002, pp. 1529. Also, as Andrew Kydd and Barbara Walter show, extremist Palestinian violence against Israelis often occurred when the prospects of peace seemed highest. One possible interpretation is that Palestinian extremists actually desired Israeli retaliation (either direct or indirect through halting the peace process) to win Palestinian moderates to their side. Kydd and Walters interpretation is slightly different: in their view, extremist violence reveals information about the trustworthiness of the other side. Kydd and Walter, The Politics of Extremist Violence.

Deterring Terrorism 253

erates than tactics that focus squarely on the perpetrators of violence. As an example of a more focused approach, consider the Israelis reaction to their athletes being taken hostage by the Palestinian Liberation Organization at the 1972 Summer Olympics in Munich: two of three surviving PLO members were assassinated without signicant collateral damage.62 Interestingly, it is often resource constraints that prevent states from adopting a more focused response, and critics have argued that the Israeli response to the Munich attack came with considerable costs in terms of manpower, [and] resources.63 The danger that, in some instances, punishment could be counterproductive applies equally to the use of force. Further, deterrence does not require that retaliation be in the near-term interests of the side that undertakes it. Rather, as discussed above, retaliation must hurt the individuals to be deterred, and the threat of retaliation must be credible. Credibility requires that the benets of a successful deterrence policy (postretaliation), in addition to the direct benets of retaliating, outweigh the near-term costs of retaliating. Thus, if Israeli retaliation hurts the perpetrators of terrorist acts, this can serve a policy of deterrence even if near-term Israeli interests are also hurt.

Fracturing the Global Terrorist Network in the Southern Philippines

In this section, we test our central argument that important elements of the global terrorist network can be deterred from actions that harm states. In particular, groups that have provided essential training and other assistance to the most dangerous terrorists can be deterred from doing so in the future. We accomplish this by examining one case where coercion is being attempted against the Moro Islamic Liberation Frontand one case where force was triedin the 2002 Operation Balikatan against the Abu Sayyaf Group. These cases offer useful lessons for U.S. counterterrorism policy because, as one study suggests, the Philippines has become the model for additional fronts in the war on terrorism, and George W. Bushs administration intends to use similar strategies in Indonesia and elsewhere.64 By comparing the outcomes in
62. The other kidnappers were killed by German police at the Munich airport. 63. Schachter, The Eye of the Believer, p. 109; and Dan Raviv and Yossi Melman, Every Spy a Prince: The Complete History of Israels Intelligence Community (Boston: Houghton Mifin, 1990), pp. 184 194. 64. Gaye Christofferson, The War on Terrorism in South East Asia: Searching for Partners, Delimiting Targets, Strategic Insight, Vol. 1, No. 1 (March 2002), http:/ /www.ccc.nps.navy.mil/ rsepResources/si/mar02/eastAsia.asp; John Gershman, Is South East Asia the Second Front?

Contending with Terrorism 254

the MILF and Abu Sayyaf Group cases, we illustrate some of the critical dynamics of the policy choice. The MILF case avoids methodological problems commonly associated with empirical tests of deterrence theories. Testing theories of general deterrence (where the actions taken by the deterrer are not in the context of an ongoing crisis) is difcult because adversaries that seem to have been deterred may not have wanted to take action in the rst place. Examining cases of immediate deterrence (where a deterrent threat is made in the context of a crisis) ameliorates the problem of evaluating the intentions of the adversary but creates a new difculty: actual adversaries are unlikely to be representative of the class of potential adversaries. Thus, inferences drawn from immediate deterrence cases may not apply to cases of general deterrence. Similarly, the very existence of a terrorist group marks the group as different from the class of potential terrorists. So the experience of deterring existing groups that have already demonstrated a willingness to carry out attacks may not provide lessons that are immediately transferable to questions concerning the whole class of terrorists and potential terrorists. Because the MILF is known to have cooperated with al-Qaida and Jemaah Islamiah in the past, we can infer that, under some circumstances, it would do so again. The case thus avoids the problem of testing for general deterrence because we know that the action to be deterred is desirable in some instances. At the same time, the U.S. and Philippine governments began an aggressive campaign to coerce the MILF to sever its ties to Jemaah Islamiah and al-Qaida only after September 11 and the string of attacks in 2002 linked to these two other groups.65 Thus, the case also avoids the problem of testing for immediate deterrence because of both the unanticipated change in policy by the deterring governments and the increased resources these governments could credibly threaten to deploy to enforce compliance. The MILF case demonstrates each aspect of our argument. First, governments often have the ability to impose costs on terrorist groups or elements of terrorist support networks, even those that are highly motivated. Second, such groups/elements respond to these incentives. Third, states can achieve important goals, such as preventing cooperation among terrorist groups, through deForeign Affairs, Vol. 81, No. 4 (July/August 2002), pp. 6074; and Walden Bello, A Second Front in the Philippines, Nation, March 18, 2002, pp. 1822. 65. Among the 2002 attacks by Jemaah Islamiah and al-Qaida in Asia are the Bali bombing, which killed more than 200 people, the Zamboanga bombings (in the Southern Philippines), and the Metro Manila bombings.

Deterring Terrorism 255

terrence. The Abu Sayyaf Group case illustrates both the capacity of states to harm terrorist groups, and the limitations of force in achieving the true ends of states that employ it. deterrence of the moro islamic liberation front In 1977 Hashim Salamat challenged Nur Misuari for leadership of the Moro National Liberation Front (MNLF), a group that views itself as carrying on a several-hundred-year struggle for the rights of Muslims in the Southern Philippines.66 Although Salamats bid for leadership was unsuccessful, several thousand ghters remained loyal to him, calling themselves the new MNLF. In 1984 this group renamed itself the Moro Islamic Liberation Front. In the 1980s and early 1990s, the MNLF negotiated with the Philippine government as the representative of the Moro people, while the MILF concentrated on building its capabilities and support at the grassroots level. In 1996 the government and the MNLF signed a peace agreement, still in place, that promised greater autonomy for Muslim regions. Within a month, however, ghting broke out between the MILF and government forces.67 Battles of varying intensities between the two sides, punctuated by several cease-re agreements, have continued almost to the present day.68 In March 2000, President Joseph Estrada ordered all-out military action against the MILF that culminated in July when government forces overran the groups main base, Camp Abubakar. Despite this use of massive military force, the group continued to pursue its objectives (principally greater autonomy for Moros), indicating a high level of motivation. Thus, if the MILF is to be coerced
66. See Guide to the Philippines Conict, BBC News, http:/ /newsvote.bbc.co.uk/mpapps/ pagetools/print/news.bbc.co.uk/2/hi/asia-pacic/1695576.stm; Anthony Davis, Attention Shifts to Moro Islamic Liberation Front, Janes Intelligence Review, April 2002, pp. 2023; Kristina Gaerlan and Mara Stankovitch, eds., Rebels, Warlords, and Ulama (Quezon City, Philippines: Institute for Popular Democracy, 2000); Alonto Norodin Lucman, Moro Archives: A History of Armed Conicts in Mindanao and East Asia (Quezon City, Philippines: FLC Press, 2000); Abaton Macapado Muslim, The Moro Armed Struggle in the Philippines: The Nonviolent Autonomy Alternative (Marawi City, Philippines: Mindanao State University, 1994); and Hilario M. Gomez, The Moro Rebellion and the Search for Peace: A Study on Christian-Muslim Relations in the Philippines (Zamboanga City, Philippines: Silsilah, 2000). 67. See, for example, Chronology for Moros in the Philippines, Minorities at Risk Project (College Park: Center for International Development and Conict Management, University of Maryland, 2004), http:/ /www.cidcm.umd.edu/inscr/mar/chronology.asp?groupId?84003. 68. After a brief suspension of the peace talks in 2005, the Philippine government and the MILF have started work on a preliminary draft of a peace settlement. The nal draft is expected by mid2006. See, for example, Barbara Mae Decanay, Negotiators Begin Work on Initial Peace Pact, Gulf News, September 24, 2005. Negotiations between the Philippine government and the MILF have been ongoing since November 2001, when the two entered into a cease-re agreement.

Contending with Terrorism 256

into ending its rebellion, either greater force than the Philippine government can muster alone must be brought to bear, or some political demands of the group must be accommodated and the accommodation held at risk. The Philippine government has repeatedly accused the MILF of terrorist activities,69 including bus and airport bombings and numerous hostage takings.70 But the validity of these reports is difcult to determine because the government also has incentive to undermine the legitimacy of the group. The rebels themselves have repeatedly denied these charges.71 Still, independent sources have also accused the group of using terror tactics, mainly against the Christian community and businesspeople refusing to pay tolls for travel in areas controlled by the rebels.72 Therefore, it appears likely that the MILF or some afliated elements have engaged in some level of terrorism. There is less doubt that elements of the group have cooperated with other terrorist organizations.73 Reports of MILF cooperation with al-Qaida and Jemaah Islamiah come not just from Philippine intelligence sources,74 but also from many other analysts.75 According to Western and Asian intelligence sources, Al Qaedas relationship with the Moro Islamic Liberation Front was . . . fruitful. At Mr. bin Ladens request, the front opened its Camp Abubakar to foreign jihadists, which meant they did not all have to go to Afghanistan.
69. Most of the violence perpetrated by the MILF has been against government forces. 70. See, for example, Rajiv Chandrasekaran, Military Finds 2 Beheaded by Philippine Rebels, Washington Post, May 7, 2000; Philippine Ofcials Link Bombings to Muslim Rebels, as Third Hostage-Taking Unfolds, CNN, World News, May 3, 2000; A Hostage Crisis Confronts Estrada, Economist, May 6, 2000, pp. 4243; and Ellen Nakashima, Five Detained in Philippine Bombing, Washington Post, March 6, 2003. 71. See MILF press releases from May 13, May 8, April 3, and February 15, all in 2003; June 21 and May 3, 2002; June 9, 2001; and July 12, 2000, http:/ /www.luwaran.com/press.htm. The case of Koronadal City is one of many examples of the differing accounts of specic incidents. See, for example. Government Nearing Decision on Whether to Tag MILF as Terrorist, Says GMA, http:/ / www.news.ops.gov.ph/archives 2003/may13.htm#MILF 72. See, for example, Muslim Separatist Movements in the Philippines and Thailand, in Angel Rabasa and Peter Chalk, Indonesias Transformation and the Stability of Southeast Asia (Santa Monica, Calif.: RAND, 2001). 73. See, for example, Daljit Singh, The Terrorist Threat in Southeast Asia, in Russell Heng and Denis Hew, eds., Regional Outlook, 20032004 (Singapore: Institute of Southeast Asian Studies, 2003), p. 4. 74. Philippine sources have in fact reversed themselves on the question of MILF links with other Southeast Asian Islamic groups, seemingly as Manilas political needs change. See Patrick Goodenough, Links between Al Qaeda and Filipino Militants Probed, September 19, 2002, http:/ /www.cnsnews.com/ForeignBureaus/Archive/200209/FOR20020919a.html. 75. For a detailed discussion of links among these groups, see Barry Desker and Kumar Ramakrishna, Forging an Indirect Strategy in Southeast Asia, Washington Quarterly, Vol. 25, No. 2 (Spring 2002), pp. 161176.

Deterring Terrorism 257

Three other camps for foreigners were opened in the 1990sCamp Palestine, primarily for Arabs; Camp Vietnam and Camp Hudaibie, for Malaysians and Indonesians.76 MILF ofcials have themselves acknowledged that members of their group trained in Afghanistan and fought against the Soviets there.77 Singaporean ofcials have detailed kinship ties between MILF and Jemaah Islamiah members,78 as well as an attempt by Jemaah Islamiah operations chief, Riduan Isamuddin (also known as Hambali), to establish a coalition of Southeast Asian Islamic groups that included the MILF. According to these ofcials, The alliance sought to promote cooperation among the separate militant groups in obtaining arms, training and nancial support, as well as conducting terrorist attacks.79 In 2003 Fathur Rohman al-Ghozi, a member of Jemaah Islamiah known to have trained terrorists from all over the Islamic world in bomb-making at a camp run by the Moro Islamic Liberation Front, was captured and killed by the Philippine military.80 Interrogations of captured Jemaah Islamiah terrorists, such as Hambali in 2003, though unreliable in themselves, conrm the evidence of cooperation between the two groups.81 Differences between groups such as the MILF, on the one hand, and Jemaah Islamiah and al-Qaida, on the other, make the former more susceptible to certain kinds of deterrence strategies. Despite MILF cooperation with Jemaah Islamiah and al-Qaida in the past, the global objectives of these two groups are not part of the MILFs core agenda. The MILF explicitly rejected the Talibans call for a jihad against the United States and its allies after the terrorist attacks on the World Trade Center and the Pentagon.82 It specically condemned the attacks, as well as the Abu Sayyaf Group and the other terrorists in the Southern Philippines.83 Further, the Philippine government appears willing to
76. Raymond Bonner, Threats and Responses: Foreign Correspondents; Southeast Asia Remains Fertile for Al Qaeda, New York Times, October 28, 2002. 77. See Philippines: Island under Siege, PBS, Frontline, http:/ /www.pbs.org/frontlineworld/ stories/philippines/thestory.html. 78. Helmi Yusof, Nabbed: Two Groomed to Head Jemaah Islamiah Group, Straits Times, December 19, 2003. 79. Alan Sipress and Ellen Nakashima, Militant Alliance in Asia Is Said to Seek Regional Islamic State, Washington Post, September 20, 2002. 80. Richard C. Paddock and Al Jacinto, Bomb Maker Gunned Down in Philippines, Los Angeles Times, October 13, 2003. 81. Sol Jose Vanzi, Jamaah Islamiyah Still Working with MILF, Philippine Headline News Online, October 7, 2003, http:/ /www.newsash.org/2003/05/ht/ht003782.htm. 82. The Taliban, which was in close alliance with Osama bin Laden, ruled Afghanistan from 1996 to 2001, when it was overthrown by U.S. troops in response to the September 11 attacks. 83. See Christina Mendez, MILF Rejects Holy War vs. US, Philippine Star, September 17, 2001,

Contending with Terrorism 258

accommodate many of the goals the MILF is seeking in the ongoing negotiations, and the groups goals can certainly be accommodated by the United States, which has little stake in the conict. Thus, despite the relatively high level of motivation of MILF members, there appears to be an opportunity to deter the group from particular courses of action, such as pursuing an independent state and using terror tactics. Even if an ultimate resolution of the conict cannot be reached, however, the threat of U.S. involvement could deter MILF cooperation with al-Qaida, Jemaah Islamiah, and the Abu Sayyaf Group. Following the September 11 attacks, the U.S. and Philippine governments used the threat of inclusion on the U.S. FTO list as a means of coercing the MILF.84 In early May 2003, the United States explicitly linked this threat to the cessation of MILF violence against civilians. As the U.S. ambassador to the Philippines, Francis Ricciardone, stated, If they continue with acts of terrorism, everybody will consider them terrorists.85 He further warned the rebels that they would lose $30 million earmarked by the U.S. Congress for their area if they did not cut their links to Jemaah Islamiah.86 Other threats were likely communicated to the MILF through channels that have not yet been documented or made public, and the coordinated U.S.-Philippine use of force against the Abu Sayyaf Group must have been at least implicitly threatening. The MILF has probably been under threat (implicitly and explicitly) from the United States and the European Union for several years. The positive turn that peace talks appear to have taken in the last couple of years may be the result of a combination of promised rewards and the increased severity of the threats that Philippine negotiators can credibly bring to bear in an environment of heightened U.S. concerns about terrorist activity. However, because sporadic peace talks began several years prior to the Sep-

cited in Larry Niksch, Abu Sayyaf: Target of Philippine-U.S. Anti-terrorism Cooperation, CRS report (Washington, D.C.: Congressional Research Service, January 25, 2002), p. 5; and MILF press release, February 14, 2003. 84. GMA Inclined to Tag MILF Terrorist, Philippine Headline News Online, May 11, 2003, http:/ /www.newsash.org/2003/05/pe/pe002597.htm. As discussed above, inclusion on the U.S. FTO list places restrictions on the ability of a group to raise funds abroad. 85. Norman Bordadora and Carlito Pablo, Ricciardone: U.S. May Declare MILF a Terrorist Group, Inquirer news service, May 6, 2003, http:/ /www.inq7.net/nat/2003/may/07/nat_2-1.htm; and U.S. Wanted MILF on Terror List, Sun.Star Network Online, November 6, 2002,http:/ /www .sunstar.com.ph/static/net/2002/11/06/us.wanted.milf.on.terror.list.ofcial.html; and Cartilo Pablo, Norman Bordadora, and Armand N. Nocum, RP Wont Ask U.S. to Recall Terrorist Tag, Inquirer news service, January 28, 2003, http:/ /ebalita.net/go/news/news.php?id=861. 86. U.S. to MILF: Cut Terror Ties or Lose Aid, October 2, 2003, Sun.Star Network Online, http:/ / www.sunstar.com.ph/static/net/2003/10/02/us.to.milf.cut.terror.ties.or.lose.aid.html.

Deterring Terrorism 259

tember 11 attacks, we can only speculate about this. But on the issue of deterring cooperation between the MILF and Jemaah Islamiah, the Abu Sayyaf Group, and al-Qaida, a signicant change in the MILFs position does appear to have occurred, with its causal roots in the changed postSeptember 11 environment. In November 2002, as a result of negotiations between the U.S. and Philippine governments and the MILF, the latter promised to help local authorities arrest about 100 suspected al-Qaida and Jemaah Islamiah operatives. As one of the negotiators explained, the MILF is more than willing to provide concrete proof that its not a terrorist group by helping us root out terrorists in the country.87 The group also agreed to assist the Philippine government in ghting the Abu Sayyaf Group. To this end, it warned the groups members against entering the territories the MILF controls and directed its armed forces to go after bandits and other criminal elements in these areas.88 Cooperation against the Abu Sayyaf Group continues, and according to Philippine Maj. Gen. Raul Relano, the government, will not stop tracking [the Abu Sayyaf Group] down with the help of our MILF friends.89 The Philippine government and the MILF have also coordinated in strikes against the Abu Sayyaf Group and the Pentagon Gang (a Filipino terrorist group that broke away from the MILF in 2001 and has continued its involvement in kidnappings and extortion and is currently on the U.S. State Departments Terrorist Exclusion List90). What is signicant about these latest examples of cooperation with the government is not only that the targeted area is an MILF stronghold, but that the rebels provided the Philippine military with critical intelligence, including information on former MILF members, and even with operational support.91 Antiterrorist cooperation between the Philippine government and the MILF was formalized in May 2002 with the creation of the Ad Hoc Joint Action

87. Quoted in Carlito Pablo, More Rebels to Hunt Jemaah Islamiah Agents, Inquirer news service, November 5, 2002, http:/ /www.inq7.net/nat/2002/nov/06/nat_1-1.htm. 88. Keith Bacongo, MILF to Help Government Crush Abu Sayyaf, MindaNews.com, June 12, 2002, http:/ /www.mindanews.com/2002/06/3rd/nws12abu.html; and Florante Solmerin, AFP, MILF Team Up vs. Sayyaf, Manila Standard, September 16, 2005. 89. GMA Adviser Sees No Obstacle to Pact with MILF, Philippine Daily Inquirer, December 1, 2004. 90. U.S. Department of State, Patterns of Global Terrorism2003, http:/ /www.state.gov/s/ct/rls/ pgtrpt/2003/31747.htm, app. D. 91. Philippine Military Chief: Abu Sayyaf Leader Hiding in Moro Rebel Area, BBC, December 6, 2004; and BBC, Philippine MORO Group Conrms Support for Operation against Kidnap Gang, August 15, 2004. See also Raymond Bonner and Carlos H. Conde, U.S. and the Philippines Join Forces to Pursue Terrorist Leader, New York Times, July 22, 2005.

Contending with Terrorism 260

Group, which is tasked with carrying out a joint operation to isolate criminal syndicates, kidnap-for-ransom groups and other criminal elements within MILF-controlled areas.92 At that time, negotiations failed to establish guidelines for action, but in late December 2004, this hurdle was overcome. According to MILF spokesman, Eid Kabalu, The joint effort to ght terrorism now includes comprehensive coverage of the Southern Philippines.93 This case illustrates the potential of coercion even against nonstate actors and highly motivated groups that have engaged in terrorist activities. The case also demonstrates the importance of tailoring the coercive approach to the goals and situation of particular groups. It is therefore essential that faraway powers understand local conicts intimately before becoming involved. joint military action against the abu sayyaf group To fully understand the dynamics of deterrence in the MILF case, it is helpful to consider both sides evaluations of a counterfactual, namely, what the state of affairs would be if the U.S. and the Philippine governments used force against the MILF. As discussed above, for a deterrence equilibrium to exist, both sides must view this outcome as less preferable to the terms of a negotiated solution. Such counterfactuals are always difcult to know. We can gain some insight into these questions, however, by briey considering a case known to all parties to the negotiations: the joint U.S.-Philippine military operation against the Abu Sayyaf Group. Like the MILF, the Abu Sayyaf Group also split from the MNLF, and the region of its basing areas is known. Denitive information is difcult to obtain, but the group appears to number in the hundreds and to be motivated primarily by the prot it gains from kidnapping ransoms.94 Its ties to bin Laden date back to ghting in Afghanistan in the 1980s. From January to July of 2002, the U.S. and Philippine armed forces conducted Operation Balikatan against the Abu Sayyaf Group. Approximately 1,300 U.S. troops, including 160 special forces, and more than 3,000 Philippine soldiers participated in the operation,

92. See Joint Communique between the Government of the Republic of the Philippines and the Moro Liberation Army, signed on May 6, 2002, http:/ /www.mindanews.com/peprcs/milf/ may6.shtml. 93. Barbara Mae Dacanay, Manila and Moros Step Up Campaign against Terrorists, Gulf News, December 31, 2004. 94. U.S. Department of State, Patterns of Global Terrorism2003, app. B.

Deterring Terrorism 261

whose main goal was to neutralize the group and to free three hostages (two Americans and one Filipino).95 This operation exemplies some of the inherent limitations of using force against terrorists, even though states can harm terrorist groups. Although several hundred Abu Sayyaf Group members were killed, its leadership remained largely intact and capable of planning and conducting new attacks. One of the American hostages was rescued, but the other two (an American missionary and a Filipino nurse) were killed. Basilan island, a stronghold of the group, was pacied, but the Abu Sayyaf Group moved to Jolo island, which became the new center of violence.96 Following the operation, Abu Sayyaf Group activities included a series of bombings, one of which killed a U.S. Green Beret.97 The presence of U.S. troops in the Philippines also caused widespread resentment and apprehensions that the U.S. presence may become permanent as it was before 1992.98 This presence may radicalize some moderate Muslims in the area, who have historically been supportive of the United States.99 For the past several years, the U.S. and Philippine governments have been considering further joint operations against the Abu Sayyaf Group, undertaking one in July 2005.100 Like previous operations, these are likely to kill some militants, but not all. Those remaining may become more radicalized, and they may seek to join global terrorist groups when their local objectives are put out of reach. Were coordinated U.S.-Philippine military action to be taken against the MILF, the results would likely be similar, though on a much larger scale given its greater size. When President Estrada declared all out war on the group in January 1999, 90,000 civilians lost their homes; the operational capabilities of

95. For more details, see C.S. Kuppuswamy, Abu Sayyaf: The Cause for the Return of U.S. Troops to Philippines? South Asia Analysis Group, No. 417 (February 2002); and C.S. Kuppuswamy, Philippines: The U.S. Campaign against Abu Sayyaf, South Asia Analysis Group, paper no. 498, July 2002. 96. Eric Schmitt with Carlos H. Conde, U.S. and the Philippines May Start New Training Mission, New York Times, December 1, 2002. 97. Ellen Nakashima, Island in Philippines Poses Counterterrorism Challenge, Washington Post, December 21, 2002; and Abu Sayyaf Launches New Attacks, Inquirer news service, July 4, 2003, http:/ /www.inq7.net/reg/2003/jul/04/reg_9-1.htm. 98. Kuppuswamy, Abu Sayyaf. 99. See, for example, Mark Landler, A Nation Challenged: The Philippines; The Temperatures a Lot Warmer but the Missions the Same: Hunting Down Terrorists, New York Times, November 4, 2001. 100. Bonner and Conde, U.S. and the Philippines Join Force to Pursue Terrorist Leader, p. 4.

Contending with Terrorism 262

the group were retarded but not destroyed; and it was encouraged to adopt more extreme tactics, including an apparent alliance with another insurgent group, the New Philippines Army, in Southern Mindanao. (The two groups agreed to conduct joint attacks and training exchanges and to share weapons.101) The use of military force thus has both benets and costs. Its periodic use may be necessary to make threats credible, but in particular situations, deterrence appears to be a preferred alternative.

Conclusions and Recommendations for U.S. Counterterrorism Policy

Our analysis leads to several conclusions for U.S. counterterrorism policy. First, when adequate resources are devoted to deterrence, traditional targeting of nonpolitical ends can sometimes deter critical elements of terrorist networks from participating in terrorist enterprises. Signicant resources should therefore be devoted to pursuing all elements of terrorist systems responsible for attacks after the fact to demonstrate the capability and will to do so and thereby increase the likelihood of future deterrence success. This implies a higher level of resource commitment than would be the case if the policy objective were merely to bring individuals responsible to justice.102 Particular emphasis should be placed on terrorist nanciers because they have targetable assets (nonpolitical ends) that stand a reasonable chance of being found. Second, even the most highly motivated terrorist groups can be deterred from certain courses of action. Of principal importance to the U.S. campaign against al-Qaida and like-minded groups is the ability to prevent them from cooperating with each other to achieve synergies. As in the case of the MILF, groups that are primarily focused on local concerns can be coerced into denying sanctuary (and other assistance) to members of more dangerous groups.103 When the United States moves beyond a deterrence posture and becomes even more deeply involved in local conicts, it will confront a number of important costs and risks. As Robert Pape and others have shown, U.S. presence abroad can promote the spread of extremist ideologies.104 The use of military
101. Antonio Lopez, Mindanaos Chance, AsiaWeek, March 5, 1999, p. 1. 102. On the effects of demonstrations of force on conventional deterrence, see Snyder, Deterrence and Defense, pp. 254258. 103. Denying terrorists sanctuary is a key recommendation of The 9/11 Commission Report, pp. 365 367. 104. Pape, Dying to Win. See also Evan Eland, Does U.S. Intervention Overseas Breed Terrorism? The Historical Record, Cato Institute Foreign Policy Brieng no. 50, December 17, 1998, http:/ / www.cato.org/pubs/fpbriefs/fpb50.pdf; U.S. Department of State, Patterns of Global Terrorism:

Deterring Terrorism 263

force, in addition to carrying direct costs in lives and resources, can become a critical source of disagreement between allies, as the lead-up to the 2003 Iraq war showed. Similar disagreements in the future could jeopardize critical U.S. efforts to maintain a broad antiterrorist coalition. Further, the use of force against terrorists and insurgencies often fails to achieve political objectives (as the Abu Sayyaf case shows), and as Martha Crenshaw warns, may radicalize the whole movement or some splinter faction.105 The application of force, and other aggressive policies, against a set of adversaries can also create powerful common interests, driving them to cooperate.106 For instance, in apparent reaction to being branded a terrorist group and having its foreign assets frozen by Western governments, the communist New Peoples Army of the Philippines announced it would combine forces with the MILF.107 In fact, the very effectiveness of local antiterrorism efforts may even turn a local movement into a global one. When primary local goals are put out of reach, militants may shift their focus to secondary global goals. Thus, Egypts effectiveness in eliminating the threat posed by Islamic Jihad may have been a reason militants such as Ayman al-Zawahiri refocused their efforts on new targets,108 linking up with bin Laden and al-Qaida.109 In choosing among policy options, decisionmakers must bear these costs in mind. This is not to say that the United States should not consider strong measures, such as the use of military force, in the war on terror. By holding at risk the local agendas of local groups, however, the United States can often more effectively achieve its ends of preventing cooperation between groups and denying sanctuary to those against which force will have to be used. Because this sort of deterrence strategy is also less resource intensive, and less likely to

1997, Department of State Publication 10535 (Washington, D.C.: Ofce of the Coordinator for Counterterrorism, Ofce of the Secretary of State, April 1998), http:/ /www.state.gov/www/ global/terrorism/1997Report/1997index.html, p. 2; Professor Chaos: Consultant Brian Jenkins Deconstructs Terrorisms Big Picture, Washington Post, June 1, 2003; and Christopher Marquis, Worlds View of U.S. Sours after Iraq War, Poll Finds, New York Times, June 4, 2003. 105. How Terrorism Ends, special report no. 48 (Washington, D.C.: United States Institute of Peace, May 25, 1999). See also Ehud Sprinzak, The Great Superterrorism Scare, Foreign Policy, No. 112 (Fall 1988), pp. 110124. 106. For the opposite argument, see James S. Robbins, Freedom Eagle, National Review, January 18, 2002, http:/ /www.nationalreview.com/contributors/robbins011802.shtml. 107. Erin Prelypchan, Manilas Twin Nightmare, Newsweek, March 24, 2003, p. 36. 108. The founder of Egyptian Islamic Jihad, al-Zawahiri allegedly played a key role in the 1988 bombings of the U.S. embassies in Tanzania and Kenya. He is believed to continue to serve as a doctor and close adviser to Osama bin Laden. 109. Nimrod Raphaeli, Ayman Muhammad Rabi Al-Zawahiri: The Making of an Arch-Terrorist, Terrorism and Political Violence, Vol. 14, No. 4 (Winter 2002), pp. 122.

Contending with Terrorism 264

cause disagreements among U.S. allies, spread extremism, and drive terrorist groups together, it is often likely to prove more effective. Third, most terrorist groups can be deterred from cooperating with al-Qaida because it is not the archetypal terrorist group. The breadth of its reach, the fanaticism of its members, and the sweeping nature of its goals make it the exception rather than the rule. Policymakers should not assume that the experience of the ght against al-Qaida is transferable to other groups. Fourth, the focus of applied resources in the antiterrorist campaign should be narrowly on al-Qaida and its few current and potential allies whose ideological afnity is so strong or whose gains from cooperation so great that they cannot be deterred. The threat that groups that target the United States directly will develop or acquire the means of causing mass casualties far outweighs all other terrorist threats the United States faces. Because of the magnitude of the resource commitment required to achieve U.S. objectives against these groups, and the gravity of the threat they pose, no resources should be unnecessarily squandered on less essential tasks (except perhaps for purposes of demonstrating capability and resolve in the event of deterrence failure). By deterring other groups from cooperating with those judged most dangerous, the United States can signicantly decrease the capacity of groups such as al-Qaida, while still preserving its resources for use against al-Qaida directly. Fifth, deterrence by denial strategies decrease the coercive leverage of terrorist tactics and therefore the motivation to carry out attacks. The United States should apply such strategies against groups that directly target it. Soft terrorist targets should be hardened, and resolve not to back down in the face of antiU.S. terrorism should be demonstrated whenever possible.110 These conclusions and recommendations are particularly timely because of the debate within the Bush administration regarding the appropriate scope of the global war on terrorism. In the spring of 2005, the administration started a high-level review of its overall counterterrorism policy.111 Taking into account the lessons from the last three years and changes in the nature of al-Qaida it110. Recent policy statements have recognized the deterrent potential of defensive measures. See, for example, Department of Defense, The National Defense Strategy of the United States of America (Washington, D.C.: U.S. Government Printing Ofce, March 2005), http:/ /www.defenselink.mil/ news/Mar2005/d20050318nds2.pdf. 111. See, for example, Susan B. Glasser, Review May Shift Terror Policies; U.S. Is Expected to Look beyond Al Qaeda, Washington Post, May 29, 2005; Jim Hogland, A Shifting Focus on Terrorism, Washington Post, April 24, 2005; and Linda Robinson, Plan of Attack, U.S. News and World Report, August 1, 2005, p. 26.

Deterring Terrorism 265

self, the U.S. government is considering comprehensive changes in its counterterrorism strategy. Although it is too early to know for certain, early reports suggest the likelihood of several key shifts.112 The new policy will emphasize the need to broaden the tools used to ght terrorism to more fully include the use of all instruments of powerdiplomatic, economic, and political. This is all to the good, particularly as it will include resources to more directly confront the spread of extremist ideas. At the same time, however, a debate continues within the administration on the optimal scope of the antiterrorist campaign, especially once fewer military and intelligence resources are focused on Iraq. A signicant broadening of the scope of the campaign to include more aggressive policies targeted at a wider range of terrorist groups will run the risks of high levels of U.S. involvement in local conicts described above. To the extent the United States is able to minimize the resources required to achieve these additional objectives by working indirectly through partner nations, the risks will be reduced but not eliminated. Given that the central policy objective in targeting these groups is to prevent them from increasing al-Qaidas global reach, a deterrence policy may be a better option.

112. Important bureaucratic changes will also be made. See, for example, Hogland, A Shifting Focus on Terrorism.

Friends Like These

Friends Like These

Counterinsurgency and the War on Terrorism

Daniel L. Byman

hroughout the 1980s, the United States poured money into El Salvador to check communist expansion in Central America. Although at that time the Salvador conict was the costliest U.S. military effort since Vietnam, at the end of the decade the United States found itself spinning its wheels. Despite almost a decade of training, aid, and high-level pushes for reform, the Salvador armed forces still suffered basic aws such as a mediocre and disengaged ofcer corps, widespread corruption, a poor promotion system, and conscripts who did not want to ght. These military weaknesses were only part of a broader problem. The armed forces perpetrated or supported blatant and brutal oppression such as the killing of moderate political opponents and Catholic Church ofcials, including priests and nuns. The military also was a strong voice against much-needed economic, political, and social reforms that, had they been implemented, would have hindered the insurgents ability to recruit and operate. Not surprisingly, as the decade ended, U.S. military ofcials concluded that an outright military victory over the communist insurgents was unlikely and that a political settlement was required. In his landmark study of El Salvador, Benjamin Schwarz found that the problem was not that the United States was ghting the wrong war or otherwise repeating Vietnam-era mistakes of using conventional military power to ght an unconventional war. Rather, the United States did not understand its own allies. El Salvadors military mirrored the country as a whole, complete with the same fractures, weaknesses, and pathologies. Indeed, U.S. attempts to initiate reform often failed because they relied on the Salvadoran military and government even though they had interests quite distinct from the U.S. agenda.1 Nor is El Salvador an exception. In his study of various Cold War
Daniel L. Byman is Director of the Center for Peace and Security Studies and Associate Professor in the Edmund A. Walsh School of Foreign Service at Georgetown University. He is also a nonresident Senior Fellow in the Saban Center for Middle East Policy at the Brookings Institution. The author would like to thank Sunil Dasgupta, David Edelstein, Alan Kuperman, Kathleen McNamara, Sara Bjerg Moller, Jeremy Shapiro, Benjamin Valentino, and two anonymous reviewers. The piece draws in part on his 2005 monograph Going to War with the Allies You Have: Allies, Counterinsurgency, and the War on Terror (Carlisle, Pa.: U.S. Army War College). 1. For a superb overview of the U.S. counterinsurgency program and its many problems, see Benjamin C. Schwarz, American Counterinsurgency Doctrine and El Salvador: The Frustrations of ReInternational Security, Vol. 31, No. 2 (Fall 2006), pp. 79115 2006 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

266

Friends Like These 267

insurgencies, Douglas Blaufarb found that the United States consistently had little leverage with its allies, which frequently pursued policies that both hindered counterinsurgency (COIN) and went contrary to U.S. interests.2 The El Salvador and other Cold War experiences are relevant to policymakers today as well as to historians, for the September 11, 2001, terrorist attacks on the United States not only ushered in a new era of counterterrorism, but they also forced the return of the counterinsurgency era. The global effort against al-Qaida has meant closer ties with a number of governments involved in ghting Islamist insurgents that, to different degrees, have ties to al-Qaida. Since the attacks, the United States has forged closer relations with Algeria, Pakistan, and other countries ghting insurgent groups that have relations with the global Sunni jihadist movement that al-Qaida champions, as well as having become embroiled in a costly counterinsurgency campaign in Iraq that has become linked to the global jihad. In most of these struggles, local security forces are the tip of the spear. And in Iraq, one of the most important U.S. problems is that Iraqi security forces are too inept to take over from the United States. Despite this central importance, thinking and scholarship on counterinsurgency tends to ignore the role of allies. Analyses are typically divided into two kinds: those that focus on the insurgents, and those that examine COIN forces, with the latter including both the United States and the host government and thus wrongly assuming they both share the same interests. Even the U.S. counterinsurgency doctrine issued in 2004 mentions the role of the host nation only in passing, without any serious discussion given to problems that may be encountered.3 What obstacles do potential U.S. allies face when ghting insurgents? What are the sources of these challenges? What inuence can the United States exert to compensate for its allies weaknesses? These are the questions this article seeks to answer.
form and the Illusions of Nation Building (Santa Monica, Calif.: RAND, 1991). Another valuable study of the U.S. problems in El Salvador with regard to reform is Andrew J. Bacevich, James D. Hallums, Richard H. White, and Thomas F. Young, American Military Policy in Small Wars: The Case of El Salvador (Washington, D.C.: Pergamon-Brasseys, 1988). 2. Douglas Blaufarb, The Counterinsurgency Era: U.S. Doctrine and Performance (New York: Free Press, 1977). This proved a particular problem for the United States in Vietnam where, as in Iraq today, the goal was to stand the Vietnamese up while U.S. forces stood down. For a review, see Scott Sigmund Gartner, Differing Evaluations of Vietnamization, Journal of Interdisciplinary History, Vol. 29, No. 2 (Autumn 1998), pp. 243262. 3. U.S. Army, Counterinsurgency Operations, FMI3-07.22 (Washington, D.C.: U.S. Army Headquarters, 2004), section 2-16, notes that the United States seeks to improve host-nation security forces, but it does not discuss their common problems and weaknesses.

Contending with Terrorism 268

This article draws on several bodies of research to better understand why allies often perform poorly, including principal-agent theory, work on moral hazards, and the vast body of research on the general problems facing alliances.4 The principal-agent approach and work on alliances suggest that allies often have highly divergent objectives and that they can effectively manipulate information to exploit the United States. When trying to exert control over its allies, the United States frequently faces problems identied by scholars working on moral hazards: the very words Washington uses to demonstrate its commitment to an ally have the unintended effect of making the ally more reckless and less likely to heed the wishes of the United States. U.S. allies that are ghting al-Qaida-linked insurgencies often have four categories of structural problems that explain some of their distinct interests and lead to particular challenges against insurgents: illegitimate (and often repressive) regimes; civil-military tension manifested by fears of a coup; economic backwardness; and discriminatory societies.5 Because of these problems, allies frequently stray far from the counterinsurgency ideal, both militarily and politically. Their military culture often is characterized by poor intelligence; a lack of initiative; little integration of forces across units; soldiers who do not want to ght; bad ofcers and noncommissioned ofcers (NCOs); and difculties with training, learning, and creativity. In addition, the structural weaknesses have a direct political effect that can aid an insurgency by hindering the development and implementation of a national strategy, encouraging widespread corruption, alienating the military from the overall population,
4. This article does not claim to systematically test these literatures. Rather, it selectively draws on their insights to shed light on the problems the United States faces when confronting insurgencies. 5. The forces examined include not only regular military forces but also those involved in ghting the insurgency, such as special police and intelligence units. In Saudi Arabia, for example, the regular army is excluded from many sensitive duties linked to counterterrorism and counterinsurgency, with the Saudi Arabian National Guard playing the key role. The special security forces and the special emergency forces play particularly important roles in the effort against al-Qaida. See Anthony H. Cordesman and Nawaf Obaid, Saudi Internal Security: A Risk Assessment, Working Draft (Washington, D.C.: Center for Strategic and International Studies, May 2004), p. 18. In Algeria, initial failures led to the formation of elite COIN units that over time became quite large. Luis Martinez, The Algerian Civil War, 19901998, trans. Jonathan Derrick (New York: Columbia University Press, 2000), p. 149. India has numerous forces involved in ghting insurgents in Kashmir, including the regular and special police, village defense committees, the central reserve police force, the Indian reserve police force, the central industrial security force, and the border security force. For a review, see Thomas A. Marks, At the Frontlines of the GWOT: State Response to Insurgency in Jammu, Journal of Counterterrorism & Homeland Security International, Vol. 10, No. 1 (Fall 2003), pp. 3846. In many countries, the paramilitary forces are as large or larger than the regular army forces. In Uzbekistan, the paramilitary forces number approximately 20,000, and those in the regular army and air force account for another 55,000. Roger N. McDermott, The Armed Forces of the Republic of Uzbekistan, 19922002: Threats, Inuence, and Reform, Journal of Slavic Military Studies, Vol. 16, No. 2 (June 2003), p. 29.

Friends Like These 269

and offering the insurgents opportunities to penetrate the military. Taken together, these problems and divergent interests explain in part why allies often have security services that are poorly postured to ght insurgencies and why allies do not make the necessary reforms to improve their performance. The United States is on the horns of a dilemma when working with allies to ght insurgents. Allies experience insurgencies because of the weakness of the state, as well as other factors such as discrimination and corruption.6 These problems create tremendous difculties when the United States expects allied militaries to ght on its behalfthe structural problems that cause the insurgencies also shape how well allies ght them. The implications of these weaknesses go beyond the ability (or lack thereof) of local forces to ght the insurgents and shape the relationship between the regime and the United States. Washington must recognize that its allies, including those in allied militaries, are often ineffective at ghting insurgents and at times can make the problem worse. U.S. COIN doctrine, no matter how well thought out, cannot succeed without the appropriate political and other reforms from the host nation, but these regimes are likely to subvert the reforms that threaten the existing power structure. The inuence of the United States is often limited, as the allies recognize that its vital interests with regard to ghting al-Qaida-linked groups are likely to outweigh any temporary disgust or anger of an allys brutality or failure to institute reforms. These conicting interests are painfully relevant to Iraq today. The United States seeks to shift the primary burden for ghting the Iraqi insurgency to Iraqi forces. Should this happen, the broader problems in Iraqi society and the distinct interests of those who make up the regime are likely to cause severe problems for the Iraqi forces ghting the insurgency, suggesting that the U.S. strategy to hand off the ght to local allies may fail. To help overcome these problems, the United States should try to increase its intelligence on allied security forces so that it can better understand the true nature of their activities. To reduce its vulnerability to manipulation, the United States should also try to diversify its intelligence sources to ensure that it does not rely exclusively on the local ally for information. At times, Washington should try to act more like a third party to a conict rather than an open and strong ally of government forces. In doing so, it can better exert lev-

6. James D. Fearon and David D. Laitin believe that factors that indicate a country is vulnerable to insurgency include poverty (which suggests states that are poor and bureaucratically weak), instability, inaccessible terrain, and a large population. See Fearon and Laitin, Ethnicity, Insurgency, and Civil War, American Political Science Review, Vol. 97, No. 1 (February 2003), pp. 7590.

Contending with Terrorism 270

erage over the government to make useful reforms and other concessions that might help solve fundamental problems. In all of its COIN programs, the United States must have realistic expectations. Training, military-to-military contacts, education programs, and other efforts to bolster its COIN capabilities all can be benecial. Indeed, these capabilities are often the best available options for the United States to shape and help allied efforts. Washington must recognize, however, that the effects of various COIN programs are likely to be limited at best because of the structural problems and divergent interests noted above: even massive amounts of aid cannot create a healthy economy where the preconditions of the rule of law and social stability do not exist, and U.S. programs for developing honest government can at best help around the margins. At times, the brutality and incompetence of the local regime may simply be too much to overcome. In these cases, the United States may damage its image profoundly yet achieve little on the ground, which suggests that the United States should carefully pick its battles and not embrace every counterinsurgency effort uncritically. The article has seven main sections. The rst section discusses the overlap between counterinsurgency and counterterrorism. The second section describes the role of allies in counterinsurgency, with a particular focus on countries ghting al-Qaida-linked insurgencies. To identify likely problems, this section draws upon research on alliances made during conventional wars and by looking at work on principal-agent problems that have plagued many businesses and international organizations. The third section offers an ideal type COIN force and then assesses how allied militaries involved in the struggle against al-Qaida fare against this hypothetical force. The fourth section discusses the structural factors that shape allies military cultures and their political proles and explains how these general factors in turn affect the politics of counterinsurgency and the military cultures of the countries in question as they are relevant to COIN. The fth examines the implications for the United States and discusses the limits of U.S. inuence over its allies. The sixth section discusses the implications of many of these problems for Iraq, should the United States transfer more of the burden of ghting the insurgency to the Iraqi forces. The article concludes with recommendations for how to mitigate some of the problems.

Counterterrorism, Counterinsurgency, and al-Qaida

Counterinsurgency and counterterrorism overlap considerably both in theory and in practice. This article uses the denition of insurgency provided in the

Friends Like These 271

Central Intelligence Agency pamphlet Guide to the Analysis of Insurgency, which reads as follows: Insurgency is a protracted political-military activity directed toward completely or partially controlling the resources of a country through the use of irregular military forces and illegal political organizations. Insurgent activity including guerrilla warfare, terrorism, and political mobilization, for example, propaganda, recruitment, front and covert party organization, and international activityis designed to weaken government control and legitimacy while increasing insurgent control and legitimacy. The common denominator of most insurgent groups is their desire to control a particular area. This objective differentiates insurgent groups from purely terrorist organizations, whose objectives do not include the creation of an alternative government capable of controlling a given area or country.7 Insurgents primary method is usually guerrilla war, which is often supplemented by terrorismdened as the use of violence or the threat of violence against civilians to achieve a political purpose and have a psychological effect. Terrorism offers insurgents another method of weakening a state and can give more narrow advantages in the broader struggle. For example, killing civilians may lead a rival ethnic group to ee a contested area, demonstrate that the government cannot impose order and protect its people, and convince ofcials and the populace as a whole to collaborate out of fear.8 The relationship between counterterrorism and counterinsurgency is not new. Many of the state-supported terrorist groups are also insurgenciesthere is no clear dividing line, and in fact tremendous overlap exists. Although the exact percentage depends heavily on coding decisions, in my judgment approximately half of the groups listed by the U.S. Department of State as foreign terrorist organizations in 2005 were insurgencies as well as terrorist groups.9 Even more important, the majority of the most worrisome terrorist groups in the world today are also insurgencies. The Liberation Tigers of Tamil Eelam (LTTE), the Kurdish Workers Party, the Lebanese Hezbollah, and the Revolu7. Central Intelligence Agency, Guide to the Analysis of Insurgency (Washington, D.C.: Central Intelligence Agency), p. 2. The pamphlet was published in the 1980s. For an alternative denition that emphasizes the importance of guerrilla war and the spirit of peasant rebellion, see Raj M. Desai and Harry Eckstein, Insurgency: The Transformation of Peasant Rebellion, World Politics, Vol. 42, No. 4 (July 1990), p. 442. Fearon and Laitin dene insurgency as involving guerrilla warfare from rural base areas, whereas the U.S. Army denition emphasizes that political power is the central issue in an insurgency. Fearon and Laitin, Ethnicity, Insurgency, and Civil War; and U.S. Army, Counterinsurgency Operations, sec. 1-1. 8. See Daniel L. Byman, The Logic of Ethnic Terrorism, Studies in Conict and Terrorism, Vol. 21, No. 2 (April/June 1998), pp. 149169. 9. U.S. Department of State, Country Reports on Terrorism, 2005 (Washington, D.C.: U.S. Department of State, April 2006), pp. 183228.

Contending with Terrorism 272

tionary Armed Forces of Colombia all use guerrilla war as a major component in their struggles, just as the Palestine Liberation Organization attempted to do in the 1960s and 1970s. Indeed, many terrorist groups that did not use guerrilla warfare, such as the Provisional Irish Republican Army and Hamas, had attempted to do so but found they were not strong enough. Other insurgent groups often expand their activities to include terrorism. Lashkar-e-Taiba, an insurgent group ghting in Kashmir, also began to conduct terrorist attacks in the rest of India as its campaign wore on.10 Al-Qaida recognizes the importance of insurgencies. Michael Scheuer, a prominent counterterrorism expert, argues that its leader, Osama bin Laden, has promoted (and at times directed) a worldwide, religiously inspired, and professionally guided Islamist insurgency.11 Support for insurgencies in Chechnya, Afghanistan, Iraq, and elsewhere has long been an al-Qaida priority as is shown in its rhetoric, recruitment, and spending.12 Understandably, the United States focuses on terrorist attacks, but with regard to both body counts and destabilization, these lower-prole insurgencies are causing much greater suffering. Insurgencies serve several vital organizational functions for al-Qaida beyond its broader ambition of wanting them to triumph and replace local governments. Insurgent veterans are often at the core of the terrorist organization. The long struggle against the Soviet Union in Afghanistan, of course, was a uniting experience for much of the al-Qaida leadership. Many members also came together and were vetted in struggles in the Balkans, Chechnya, Kashmir, and now Iraq.13 Because al-Qaida can tap into these insurgencies for recruits and its logistics network, it is able to conduct operations far beyond where its narrow core is located and can replenish cadre as they are lost.14 Insurgencies also add legitimacy to al-Qaida as Muslims around the world support many of these struggles, even though they might otherwise oppose al-Qaidas ideological agenda and use of terrorism. Thus, defeating al-Qaida
10. Thomas A. Marks, India: State Response to Insurgency in Jammu & KashmirThe Jammu Case, Low Intensity Conict & Law Enforcement, Vol. 12, No. 3 (Autumn 2004), p. 137. 11. See Anonymous, Through Our Enemies Eyes: Osama bin Laden, Radical Islam, and the Future of America (Washington, D.C.: Brasseys, 2002), p. xviii. Anonymous was subsequently revealed as Michael Scheuer. 12. Ibid., p. 207. 13. National Commission on Terrorist Attacks Upon the United States, The 9/11 Commission Report (New York: W.W. Norton, 2004), pp. 4770; and Marc Sageman, Understanding Terror Networks (Philadelphia: University of Pennsylvania Press, 2004). 14. The ability to recruit and replace lost cadre is vital for successful terrorist organizations. Kim Cragin and Sara A. Daly, The Dynamic Terrorist Threat: An Assessment of Group Motivations and Capabilities in a Changing World (Santa Monica, Calif.: RAND, 2003), pp. 3436.

Friends Like These 273

requires defeating, or at least inhibiting, its ability to tap into insurgencies around the world. Al-Qaida maintains links to several insurgencies and proto-insurgencies around the globe. Insurgent ghters in Algeria (the Salast Group for Preaching and Combat), Iraq (various Sunni groups), Afghanistan (the Taliban), India (groups ghting in Kashmir such as Lashkar-e-Taiba), Chechnya, Pakistan (former Taliban and their sympathizers among Pakistani domestic groups), Somalia (various Islamist ghters), and Uzbekistan (the Islamic Movement of Uzbekistan) all have varying degrees of ties to al-Qaida and the movement it supports. Size is a key criteria for an insurgency, and several groups examined in this study are not yet strong enough to be deemed insurgencies, but may move in that direction. They include the anti-Saudi al-Qaida on the Arabian Peninsula, the Jemaah Islamiyah in Southeast Asia, the Abu Sayyaf Group in the Philippines, and the Islamic Group and Egyptian Islamic Jihad in Egypt.15 To be clear, al-Qaida did not cause any of these insurgencies, and in almost all the cases the insurgents have agendas that are in many ways distinct from al-Qaidas. Nevertheless, individuals afliated with these groups are often members of al-Qaida, and the terrorist organization in turn exploits these groups networks and personnel for its own purposes. It is thus plausible that a war on terrorism might lead to greater U.S. involvement with these countries, and in most cases it already has.16 As Steven Metz and Raymond Millen note, The United States is more likely to assist regimes threatened by insurgents linked to al-Qaeda or its afliates.17 Iraq is excluded from the analysis section of the article because the focus is on allied militaries and their relations to their societies, which in Iraq is complicated by the U.S. occupation and heavy role in the post-occupation. Since the insurgency began in 2003, U.S. forces have done the bulk of the ghting, not local Iraqis.18 The arguments in this article, however, have tremendous implications for
15. Somalia is excluded from this study because it lacks a true government. The United States, however, appears to have allied with local militias ghting those tied to jihadists. See Emily Wax and Karen DeYoung, United States Secretly Backing Warlords in Somalia, Washington Post, May 17, 2006. 16. Bruce Hoffman notes that the United States repeatedly fails to detect the development of insurgencies until it is too late. Hoffman, Insurgency and Counterinsurgency in Iraq, Occasional Paper OP-127-IPC/CMEPP (Santa Monica, Calif.: RAND, June 2004), pp. 14. 17. Steven Metz and Raymond Millen, Insurgency and Counterinsurgency in the 21st Century: Reconceptualizing Threat and Response (Carlisle Barracks, Pa.: Strategic Studies Institute, U.S. Army War College, November 2004), p. 18. 18. Using preliberation forces, however, would skew the data as well. For a superb review of the

Contending with Terrorism 274

the future of Iraq once U.S. forces depart or draw down: objectives that both the George W. Bush administration and its critics seek. As the Bush administrations National Strategy for Victory in Iraq makes clear, the United States aims to build up the capabilities of the Iraqi security forces and steadily shift responsibility for the ght to them.19 Thus, the ndings from other struggles are applied to Iraq briey in the articles conclusion. Although counterterrorism and counterinsurgency overlap, they are far from identical. Some components of the U.S. global counterterrorism campaign have little value for counterinsurgency. The United States tries to inhibit the international travel of terrorists, track cells operating in otherwise peaceful countries, render suspects from one country to another, and construct barriers around national icons in the United States. All of these are done primarily against the relatively small number of international terrorists (or those they inspire) and have little to do with guerrilla war or the broader popular struggles noted above. They require the United States to develop ties to many countries in Europe and Asia, use its intelligence services to monitor radicals around the world, and construct effective defenses. All of these steps require capabilities different from those needed for counterterrorism. Similarly, many insurgents, including some Islamist ones, have a local focus and do not see their insurgency as inherently involving the United States. The LTTE, for example, is a vicious terrorist group and a highly capable insurgency. Its actions, though violent, are primarily directed against the government of Sri Lanka and do not involve U.S. interests. Similarly, the United States can participate in or avoid struggles against Maoist rebels in Nepal or even Islamist rebels such as those found in Hamas, who have shunned alQaida, but these are not part of the struggle against bin Laden and his allies.

Allies Centrality in Counterinsurgency

Successful counterinsurgency against anti-U.S. groups linked to al-Qaida depends heavily on allies military forces. In addition to establishing government control and eliminating insurgent combatants, military forces act to secure an area so that political and other reforms can be carried out.20 Allies are also vital
Iraqi military performance through the 1991 Gulf War, see Kenneth M. Pollack, Arabs at War: Military Effectiveness, 19481991 (Lincoln: University of Nebraska Press, 2002), pp. 155266. 19. National Security Council, National Strategy for Victory in Iraq (Washington, D.C.: National Security Council, November 2005), pp. 1821, http:/ /www.whitehouse.gov/infocus/iraq/iraq_ national_strategy_20051130.pdf. 20. Thomas A. Marks, Insurgency in a Time of Terrorism, Journal of Counterterrorism & Homeland Security, Vol. 11, No. 2 (Spring 2005), pp. 4653.

Friends Like These 275

in part for political reasons at home. The American people naturally prefer that others ght and die in their stead, particularly when the conict so obviously involves a third countrys vital interests.21 Equally important, allies should be better able to carry out most aspects of counterinsurgency. Their forces speak the language and know the culture, so they are better able to gather intelligence and avoid actions that gratuitously offend the population.22 Even the best-behaved foreigners may generate a nationalistic backlash among local citizens who feel little sympathy for the insurgents. By most accounts, the U.S. military has tried to respect Afghanistans culture and empower local ofcials there. But resistance has grown since the 2001 invasion, and the United States welcome has worn thin in parts of the country.23 Finally, perhaps the greatest factor affecting the insurgents success or failure is the response of the regime: a clumsy or foolish response can be the insurgents greatest source of recruits.24 Allies, however, have their own distinct interests and approaches to counterinsurgency and counterterrorism, as they do with all serious security issues: a point well developed in the scholarship on international relations. Historically, major powers have differed from their allies in such basic questions as, What is the overall level of risk? Who is the primary enemy? How much should different allies contribute? And what is the best means to ght the insurgents?25 Many of these differences hold true for counterinsurgency as well. Like those of the United States, the allys interests also include defeating the insurgents: but here the commonality often ends. As discussed below, allied security ser-

21. Stephen T. Hosmer, Constraints on U.S. Strategy in Third World Conicts (New York: Crane Russak and Company, 1987), p. 128. 22. U.S. Army, Counterinsurgency Operations, sec. 1-10. 23. Ahmed Rashid, Afghanistan: On the Brink, New York Review of Books, June 22, 2006, http:/ / www.nybooks.com/articles/19098. For a discussion on the nationalistic backlash that outside occupiers face, see David Edelstein, Occupational Hazards: Why Military Occupations Succeed or Fail, International Security, Vol. 29, No. 1 (Summer 2004), pp. 4991; and Bard E. ONeill, Insurgency and Terrorism: Inside Modern Revolutionary Warfare (Dulles, Va.: Brasseys, 1990), p. 137. 24. ONeill, Insurgency and Terrorism, p. 125. Some experts argue that this has happened already in Uzbekistan. See Prepared Statement of Martha Brill Olcott, reprinted in Central Asia: Terrorism, Religious Extremism, and Regional Instability, Hearing before the Subcommittee on the Middle East and Central Asia of the House Committee on International Relations, 108th Cong., 1st sess., October 29, 2003, p. 57. 25. The literature on alliances among states is vast. Leading works include Mancur Olson and Richard Zeckhauser, An Economic Theory of Alliances, Review of Economics and Statistics, Vol. 48, No. 3 (August 1966), pp. 266279; Thomas J. Christensen and Jack Snyder, Chain Gangs and Passed Bucks: Predicting Alliance Patterns in Multipolarity, International Organization, Vol. 44, No. 2 (Spring 1990), pp. 137168; Glenn H. Snyder, Alliance Politics (Ithaca, N.Y.: Cornell University Press, 1997); and Stephen M. Walt, The Origins of Alliances (Ithaca, N.Y.: Cornell University Press, 1987).

Contending with Terrorism 276

vices also seek other goalsfor example, propping up an authoritarian regime or maintaining a favored ethnic group in poweror otherwise have ambitions and problems that undermine the effectiveness of counterinsurgency campaigns. In such circumstances, the ally may not act as the United States wants (e.g., refusing democratic reforms) or may act in a counterproductive way (e.g., engaging in atrocities against rival ethnic groups) because it perceives such measures to service its other interests, even at the cost of counterinsurgency. Thus, while many of the allies problems involve incompetence and ineffectiveness, these are not a product of ignorance or chance. Rather, they stem from the very nature of allied regimes and societies. Indeed, allies that suffer from these structural problems and have ineffective militaries are more likely to experience insurgencies. The basic problem of divergent interests inherent in most alliances is compounded by dilemmas similar to a principal-agent problem in which the United States cannot rely on its allies to serve its interests faithfully.26 For counterinsurgency more than other forms of warfare, the United States relies heavily on others to act on its behalf. A classic principal-agent problem occurs when information asymmetries make it hard for an employer to monitor the action of an employee, allowing the employee to act in a way that meets his or her needs, not those of the employer. The United States, like other principals, cannot control its agents completely. Allies are sovereign states, and as such enjoy at least de jure independence. As a superpower, the United States has considerable inuence. But as noted below, many of the issues linked to coun-

26. The similarity is far from exact: foreign governments are not employees of the United States, no matter how dependent they are on U.S. aid or nancial support. Nor are the governments created bodies such as the European Commission, which other scholars have pointed to as an example of an agent that at times acts contrary to the interests of its principals. Mark Pollack, Delegation, Agency, and Agenda Setting in the European Community, International Organization, Vol. 51, No. 1 (Winter 1997), p. 108. Nevertheless, the problems are similar, making the comparison worthwhile. As Joseph Stiglitz contends, The employer cannot travel on the road with his salesman to monitor precisely the effort he puts into his salesmanship. In each of these situations, the agents . . . action affects the principal. Stiglitz, Principal and Agent (ii), in Josh Eatwell, Murray Milgate, and Peter Newman, eds., New York Palgrave Dictionary of Economics, Vol. 3 (London: Macmillan, 1989), p. 967. For other important works on the principal-agent problem, see David. E.M. Sappington, Incentives in Principal-Agent Relationships, Journal of Economic Perspectives, Vol. 5, No. 2 (Spring 1991), pp. 4566. See also Kenneth Arrow, The Economics of Agency, in John W. Pratt and Richard J. Zeckhauser, eds., Principals and Agents: The Structure of Business (Boston: Harvard Business School Press, 1985), pp. 3751. Principals can try to monitor their agents through a variety of measures, such as threatening to cut budgets or impose other sanctions and providing incentives to follow the principals desires, but these in turn are costly and often imperfect. Pollack, Delegation, Agency, and Agenda Setting in the European Community, pp. 116129. See also Roderick Kiewiet and Matthew D. McCubbins, The Logic of Delegation: Congressional Parties and the Appropriation Process (Chicago: University of Chicago Press, 1991), p. 27.

Friends Like These 277

terinsurgency involve a regimes very hold on power, a concern that at best limits outside leverage that is exercised in the form of aid or threats of sanctions.27 Information disparities compound this problem further. U.S. forces are not present in every village or neighborhood where insurgents are fought, and indeed the United States often does not have a robust ofcial presence in most of an allys major cities. Moreover, the security services that ght the insurgency are also the ones that gather much of the relevant information. They may inadvertently lter the information according to their own biases and deliberately distort the information to reinforce their views (e.g., that the rival ethnic groups village was a hotbed of al-Qaida-linked unrest). Evaluating the allys true efforts is difcult for reasons that go beyond the information disparity and involve the inherent complexity of the problem. Many of the key issues involvedfor example, the pace of democratic reform, the level of professionalism in a military, and the level of success in wooing local supportare exceptionally difcult to measure.

Counterinsurgency Ideals and Realities

Counterinsurgency is difcult for even the best militaries. It requires not only remarkable military skill but also a deft political touch. The soldier on patrol must be a ghter, a policeman, an intelligence ofcer, a diplomat, and an aid worker. Not surprisingly, even well-trained, well-led, and well-funded militaries such as those of the United States, Britain, and Israel have foundered when facing insurgent movements. COIN problems only multiply if the government in question seeks to avoid extreme brutality, which is typically a U.S. goal. Iraq under Saddam Hussein, and the Algerian government in the 1990s, both defeated insurgencies, but they did so in part by widespread slaughter. Embracing such a brutal regime has tremendous normative problems for the United States. Even if these are put aside, the problem of being tarred with an allys brush grows tremendously as a result. The problems for many militaries confronting insurgents linked to al-Qaida fall into two categories. First, at the tactical and organizational levels, they are often not prepared for counterinsurgency operations. Second, as a political en27. Daniel Byman and Matthew Waxman, The Dynamics of Coercion: American Foreign Policy and the Limits of Military Might (New York: Cambridge University Press, 2002), pp. 3745.

Contending with Terrorism 278

tity the military frequently contributes to popular anger or other problems that aid the insurgents cause.28 tactical and organizational concerns The characteristics of an ideal COIN military at the tactical and organizational levels are vast,29 but several of the most important include a high level of initiative, excellent intelligence, integration across units and services, leadership, motivated soldiers, and learning and creativity. The reality for many militaries involved in ghting al-Qaida-linked insurgencies, however, is often far from the ideal.30 initiative. Counterinsurgency ghting rarely involves a set-piece battle, which outgunned and outmanned insurgents typically shun. Soldiers try to ush out hidden insurgents and defeat them through aggressive patrolling and ambushes. To do this successfully, COIN forces must also operate out of garrison and at nighttime. Part of their job is to convince the population that they will be protected; insurgents will exploit the absence of COIN forces to intimidate locals and sow fear.31 Fighting occurs either because the insurgents choose to engage or ideally when government forces leave them no choice. In both cases, planning is difcult: the terrain can vary, the number of forces involved is hard to predict, and so on. Much of the effort is done at the smallunit level, as larger units are far easier for insurgents to avoid. Many developing world militaries, however, are garrison militaries that
28. For a general overview of insurgency and counterinsurgency issues that pays particular attention to the problem of allies, see D. Michael Shafer, Deadly Paradigms: The Failure of U.S. Counterinsurgency Policy (Princeton, N.J.: Princeton University Press, 1988). Other excellent works are Blaufarb, The Counterinsurgency Era; George K. Tanham and Dennis J. Duncanson, Some Dilemmas of Counterinsurgency, Foreign Affairs, Vol. 48, No. 2 (January 1970), pp. 113122; William E. Odom, On Internal War: American and Soviet Approaches to Third World Clients and Insurgents (Durham, N.C.: Duke University Press, 1992); Harry Eckstein, ed., Internal War: Problems and Approaches (New York: Free Press, 1964); Nathan Leites and Charles Wolf Jr., Rebellion and Authority: An Analytic Essay on Insurgent Conicts (Chicago: Markham, 1970); Ian F.W. Beckett, Modern Insurgencies and Counterinsurgencies: Guerrillas and Their Opponents since 1750 (New York: Routledge, 2003); Bruce Hoffman, Jennifer M. Taw, and David W. Arnold, Lessons for Contemporary Counterinsurgencies: The Rhodesian Experience (Santa Monica, Calif.: RAND, 1991); and Marks, Insurgency in a Time of Terrorism. 29. For one useful list, see Central Intelligence Agency, Guide to the Analysis of Insurgency, p. 13. 30. Most of these militaries also suffer similar problems in conventional military operations. The Saudis lack of initiative in counterinsurgency, for example, is mirrored in their problems in conventional warfare. This article focuses on the characteristics useful for effective counterinsurgency, but many of these problems apply to overall military operations. 31. ONeill, Insurgency and Terrorism, p. 130; and Central Intelligence Agency, Guide to the Analysis of Insurgency, p. 10.

Friends Like These 279

ght, when they do, from 9:00 to 5:00. They often operate poorly as small units, with junior ofcers and NCOs reluctant to exercise initiative. In Uzbekistan, insurgents own the night.32 In Algeria, units often feared to leave their garrison and patrol in dangerous parts of cities where insurgents enjoyed sympathy. The Egyptian and Saudi armed forces historically exercised little initiative, a problem that is more profound farther down the command chain. Even the smallest details had to be spelled out in advance.33 COIN suffered as a result in all these cases. intelligence. Intelligence is the sina qua non of counterinsurgency. The insurgents primary advantage is their stealth: if they can be found, they can usually be killed or captured. A corollary is that insurgents advantage is their superior knowledge of the local population and conditions: denying or minimizing this advantage is therefore vital.34 To be useful, intelligence must not only be collected but also be analyzed, disseminated, and integrated into the overall strategy.35 Many U.S. allies, however, collect, analyze, and disseminate information poorly. Intelligence analysis benets from superb information sharing and from the proper storage and dissemination of datageneral weaknesses in developing societies. Although many allies do a ne job collecting information, they often do not integrate it well and have at best a limited picture of their adversary.36 The major intelligence agencies in the Philippines frequently do not share information and do not coordinate their activities. Indeed, intelligence money is often paid directly to ofcials as a form of graft.37 In Egypt, key information is compartmentalized, and failures are not brought to the attention of senior ofcers.38 Often, information is not shared because commanders and units do not trust each other. In the Sikh insurgency, Indian units rarely shared information with local forces because they were perceived as sympathetic to
32. McDermott, The Armed Forces of the Republic of Uzbekistan, 19902002, p. 32. 33. Kenneth M. Pollack, The Sphinx and the Eagle, 1998, chap. 4. 34. See Fearon and Laitin, Ethnicity, Insurgency, and Civil War. 35. ONeill, Insurgency and Terrorism, p. 144; and Bruce Hoffman and Jennifer Morrison Taw, A Strategic Framework for Countering Terrorism and Insurgency (Santa Monica, Calif.: RAND, 1992), p. 9. 36. For a more complete overview of this problem, see Kenneth M. Pollack, The Inuence of Arab Culture on Arab Military Effectiveness, Ph.D. dissertation, Massachusetts Institute of Technology, 1996, pp. 564566. 37. Emil P. Jurado, Need for Coordinated Intelligence Network, Manila Standard, February 23, 2005; and Fe.V. Maragay, Safeguarding the Spy Fund, Manila Standard, January 31, 2005. 38. Pollack, The Sphinx and the Eagle.

Contending with Terrorism 280

(and penetrated by) Sikh insurgents.39 The International Crisis Group reports that Indonesian intelligence is marked by blurred lines of authority, interagency rivalry, lack of coordination, unnecessary duplication, lack of adequate oversight, and the legacy of an authoritarian past.40 integration across units and services. All military operations benet from synergies, and this holds true for counterinsurgency as well. At a most basic level, units must work together to ensure proper coverage of a territory and to prevent insurgents from slipping between the seams of different units. Also, if unexpectedly heavy resistance is found, units must reinforce their beleaguered comrades, particularly when forces operate as small units. Information sharing and coordination across services are often exceptionally poor for allies ghting insurgencies linked to al-Qaida. In Egypt and Saudi Arabia, commanders of different services and units frequently do not communicate with each other. In Afghanistan, the United States has worked not only with the Afghan National Army but also with numerous regional warlords, several of whom owe little loyalty to the central government.41 At times a military may have multiple groups within it vying for power. In Algeria, the army has numerous divisions based on region and tribe.42 The division of labor between the police and the military is not clear in Indonesia, and the militarys own coordination with regard to counterterrorism and counterinsurgency is poor.43 leadership at all levels. Ofcers must be creative, take the initiative, inspire the soldiers who follow them, and perform other essential military functions. In addition, ofcers doing COIN must play a major role in ensuring the gathering of intelligence and reassuring the populationboth difcult tasks that go beyond traditional training for conventional military operations. Because much of counterinsurgency is done by small units, having strong NCOs is also vital. Military leadership in many of the countries ghting al-Qaida, however,
39. C. Christine Fair, Urban Battle Fields of South Asia: Lessons Learned from Sri Lanka, India, and Pakistan (Santa Monica, Calif.: RAND, 2004), p. 94. 40. International Crisis Group, Indonesia: Rethinking Internal Security Strategy, Asia Report, No. 90 (December 2004), p. 13. 41. Raymond A. Millen, Afghanistan: Reconstituting a Collapsed State (Carlisle Barracks, Pa.: U.S. Army War College, 2005), p. 3. 42. International Crisis Group, The Algerian Crisis: Not Over Yet, Africa Report, No. 24 (October 2000), p. 12. 43. International Crisis Group, Indonesia: Rethinking Internal Security Strategy, p. 5.

Friends Like These 281

is quite poor. In most Arab militaries, junior ofcers and NCOs are given little responsibility. In Egypt, for example, colonels have responsibilities similar to those of U.S. captains.44 In Uzbekistan, ofcers have often performed poorly when facing insurgents.45 Afghan army leaders appear better than most of those in this study, but even here the NCOs are not given appropriate responsibility.46 motivated soldiers. Soldiers who believe in their government and their ofcers are more likely to brave the hazards of COIN warfare. They will confront rather than ee from the enemy and take the necessary initiative to ensure victory. Many developing world countries facing al-Qaida, however, have poorly motivated soldiers. Afghan recruits often prefer to work for local warlords than for the national government, and many trainees and recent recruits desert their posts. As one Afghan sergeant commented, Everyone wants to run away.47 Uzbek soldiers suffer from low morale, and mass desertions are common.48 In Egypt, many soldiers do not embrace their profession.49 In India, the emphasis on caste creates problems for cohesion, as soldiers often speak different languages and the caste system creates a hierarchy among them.50 training, learning, and creativity. Counterinsurgency requires a high degree of skill and constant renement. In addition, a successful military must learn from its mistakes and be able to go beyond its standard procedures when confronted with a new situation. Moreover, both successful and unsuccessful insurgencies go through stages, and the mix of conventional and unconventional operations needed to defeat them will vary as a result.51 COIN is also

44. Pollack, The Sphinx and the Eagle. 45. Farkhad Tolipov and Roger N. McDermott, Uzbekistan and the U.S.: Partners against Terrorism, Review of International Affairs, Vol. 2, No. 4 (Summer 2003), p. 2. 46. Charles DiLeonardo, Training the Afghan National Army, Infantry, Vol. 94, No. 2 (March/ April 2005), p. 32. 47. Quoted in Tom Coghlan, Toughing It Out in the Afghan Army, BBC News, June 15, 2005, http:/ /www.news.bbc.co.uk. 48. Ahmed Rashid, A Long, Hot Summer Ahead, Far Eastern Economic Review, April 19, 2001, p. 29. 49. Many Egyptian air force ofcers, for example, do not enjoy ying. Pollack, The Sphinx and the Eagle. 50. Stephen Peter Rosen, Societies and Military Power: India and Its Armies (Ithaca, N.Y.: Cornell University Press, 1996), p. 243. 51. ONeill, Insurgency and Terrorism, p. 131; and Metz and Millen, Insurgency and Counterinsurgency in the 21st Century, p. 25.

Contending with Terrorism 282

more art than science: creativity is vital. Helping this process is the free ow of information and an institutional culture of honest criticism.52 Many countries do not emphasize training for a variety of reasons. Most important, training still focuses on conventional military operations: the traditional emphasis of most militaries. In Uzbekistan, the military was structured from Soviet-era forces intended to ght conventional wars. Although Uzbek leaders have initiated some reforms, Roger McDermott notes that these are only a modest beginning and are focused on a few elite forces.53 (Indeed, as a sign of how bad training is, an individual who was picked to lead special forces teams had not had basic infantry training.)54 Some of the paramilitary forces involved in COIN are expected to be around only temporarily, leaving ofcials reluctant to invest in long-term training.55 Egypts military is huge, making it hard to do more than rudimentary training for many of the forces. Live-re exercises or other forms of realistic training are rare.56 Many of these militaries do not learn from their mistakes. The Egyptian military has institutionalized some practices that U.S. trainers see as disastrous, in part because changing them would require an embarrassing admission of failure.57 U.S. trainers spent years working with the Saudi air force, only to watch it steadily decline. In part, this problem occurs because professional military education is weak, and the institutions that do exist focus on perpetuating existing doctrine rather than on seeking to correct mistakes in current operations. political concerns In counterinsurgency, the military is a political actor as well as a ghting unit. Several other, more political criteria include proper civil-military integration, a lack of corruption, a lack of insurgent penetration, and a sense that the army can win over the population. civil-military integration. Defeating an insurgent movement is as much (if not more) a political effort as a military one. A national approach that incorporates all dimensions of power is essential. If political and military leaders
52. John A. Nagl, Counterinsurgency Lessons from Malaya and Vietnam: Learning to Eat Soup with a Knife (Westport, Conn.: Praeger, 2002), pp. 191226. 53. McDermott, The Armed Forces of the Republic of Uzbekistan, 19922002, p. 30. 54. Tolipov and McDermott, Uzbekistan and the U.S., p. 14. 55. Sunil Dasgupta, Rise of Paramilitary Forces: Military Reorganization in Developing Societies, Working Draft (Washington, D.C.: Brookings, 2005), p. 23. 56. See Pollack, The Sphinx and the Eagle. 57. Ibid.

Friends Like These 283

are in harmony, military and political measures to defeat the insurgents are more likely to be as well.58 The two cannot be done in isolation: the military methods used affect the overall perception of the government, and the perception of the regime inuences the ability of the military to operate. In many of the countries in question, however, civil-military relations are poor. In India, civilian leaders historically saw the military as a vestige of the British imperial mentality and at odds with their nationalistic (and more socialist) vision of the country. In Algeria and Pakistan, military leaders have seized power from civilian ofcials, while in the Philippines and Indonesia civilian leaders have feared military interference in their control of the country. In Egypt, the government has long been unsure of the reliability of the military to protect the regime: a well-founded perception given that Islamist militants penetrated the military to kill Egyptian President Anwar Sadat in 1981 and that mass riots involving 17,000 conscripts broke out in 1986.59 In some countries, the government is divided on issues related to counterinsurgency, making it exceptionally difcult to produce a coherent strategy. The Algerian regime was long split between the conciliators and the eradicators, leading to a policy that was at times incoherent, with olive branches suddenly withdrawn and attempts to intimidate offset by surprising concessions.60 honesty and corruption. The military is more likely to gain the respect of the population if it is not corrupt or otherwise engaged in illicit activities. A lack of corruption sends the message that the military is indeed ghting for the country, not just for the personal interests of a few individuals. This in turn inspires soldiers to ght harder and makes it more difcult for the insurgents to penetrate the armed forces. Corruption is rampant in many of the countries in question, and the military is no exception. Of the countries surveyed in this article, all were in the lower half of Transparency Internationals 2005 Corruption Perception Index, with the exceptions of Saudi Arabia and Egypt, which scored in the middle. Indonesia, Pakistan, and Uzbekistan were among the most corrupt countries in the world.61 Uzbekistans military leaders often will exempt an individual
58. ONeill, Insurgency and Terrorism, p. 146; and Hoffman and Taw, A Strategic Framework for Countering Terrorism and Insurgency, p. v. 59. Lawrence E. Kline, Egyptian and Algerian Insurgencies: A Comparison, Small Wars and Insurgencies, Vol. 9, No. 2 (Autumn 1998), p. 127. 60. Ibid., p. 129. 61. See Transparency International, The 2005 Transparency International Corruption Perceptions Index, http:/ /www.infoplease.com/ipa/A0781359.html.

Contending with Terrorism 284

from military service for the right price.62 The Abu Sayyaf Group buys weapons and immunity freely from government and military ofcialsseveral leading terrorists simply walked out of the heavily secured national police headquarters in Manila with the aid of local ofcers.63 In Egypt and Saudi Arabia, it is assumed that senior military leaders will skim off the top of military contracts: an assumption that is duplicated in lesser ways down the chain of command. In Algeria, business rivalries prevent the different military leaders from cooperating.64 In Indonesia, corruption is rampant in the buying of equipment and other supplies.65 All of these examples only scratch the surface of the myriad ways corruption undermines military effectiveness. insurgent penetration. Ensuring that the military remains free from insurgent penetration is vital. Successful penetration allows the insurgents to avoid regime attempts to arrest or kill insurgent cadre. In addition, it gives the insurgents inside information that greatly increases their effectiveness in planning attacks. Many of the regimes ghting al-Qaida have been penetrated by insurgents. In Pakistan and Saudi Arabia, al-Qaida has made a conscious effort to cultivate military and government ofcials.66 U.S. ofcials working with the Afghan National Army are prohibited from sharing intelligence, as they fear it will fall into the hands of ex-Taliban.67 In Algeria, many of the early insurgent successes involved members of the security forces who collaborated with them, and the regime often hesitated to use the army because it feared that many soldiers would desert.68 military support from a population. Famously, counterinsurgency involves winning the hearts and minds of the population at large and denying such support to the guerrillas. As the military plans and conducts its operations against insurgents, it must also think about how to win over the general population, as one of its most important roles is to serve as the shield for carrying out reform.69 The military needs the active support of the population to
62. McDermott, The Armed Forces of the Republic of Uzbekistan, 19922002, p. 36. 63. Steven Rogers, Beyond the Abu Sayyaf, Foreign Affairs, Vol. 83, No. 1 (January/February 2004), pp. 1520. 64. International Crisis Group, The Algerian Crisis, p. 12. 65. International Crisis Group, Indonesia: Next Steps in Military Reform, Asia Report, No. 24 (October 2001), p. 13. 66. Anonymous, Through Our Enemies Eyes, p. 22. 67. DiLeonardo, Training the Afghan National Army, p. 38. 68. Michael Willes, The Islamist Challenge in Algeria: A Political History (New York: New York University Press, 1997), p. 295; Martinez, The Algerian Civil War, 19901998, p. 161; and Kline, Egyptian and Algerian Insurgencies, p. 129. 69. Marks, Insurgency in a Time of Terrorism, p. 52.

Friends Like These 285

gain informationa disadvantage, as mere passivity often allows insurgents to operate effectively without being vulnerable to government intelligence efforts.70 In addition to using the population to collect intelligence, greater popular support gives guerrillas more eager recruits and more money on which to draw. To gain active support, it is helpful if the military is ghting for a system that offers political, economic, and other opportunities to all concerned: something that is often beyond its control. In addition, a military is more likely to win over the population in general if it is seen as ghting for more than just a political or social clique. If the armed forces are viewed as representing all the diverse communities of any state, they are more likely to be seen as trustworthy and not provoke a backlash. Finally, successful COIN is characterized by restraint as well as by violence. Too much destruction can alienate a population rather than reassure it and unwittingly create disincentives to ght for and cooperate with the government.71 Such benign characteristics are often lacking among militaries ghting alQaida-linked groups, making it harder for them to capture popular support. In India, for example, the armys outlook is more akin to the British imperial army. As a result, it sees itself more as occupier than as part of the local population, with the result that relations are often poor.72 When suppressing the Punjab insurgency, the Indian forces saw themselves as protecting the local Hindu population from Sikh militants, and as a result it alienated local services that sought to balance Sikh and Hindu concerns.73 Indonesian soldiers often take sides in local disputes according to whether the soldiers are Muslim or Christian.74 If the military lacks popular support, basic counterinsurgency functions such as gathering intelligence and denying information to the insurgents are
70. Christopher M. Ford, Speak No Evil: Targeting a Populations Neutrality to Defeat an Insurgency, Parameters, Vol. 35, No. 2 (Summer 2005), p. 53. T.E. Lawrence noted that the guerrillas needed only a limited amount of active support, but that much of the population must be passive or inactive backers. Beckett, Modern Insurgencies and Counterinsurgencies, p. 20. See also Central Intelligence Agency, Guide to the Analysis of Insurgency, p. 8. For an alternative view, see Chalmers Johnson, Civilian Loyalties and Guerrilla Conict, World Politics, Vol. 14, No. 4 (July 1962), pp. 646661. 71. For an assessment of how incentives affect the success of rebellions, see Mark I. Lichback, What Makes Rational Peasants Revolutionary? Dilemma, Paradox, and Irony in Peasant Collective Action, World Politics, Vol. 46 No. 3 (April 1994), pp. 383418; and Jeffrey Berejikian, Revolutionary Collective Action and the Agent-Structure Problem, American Political Science Review, Vol. 86, No. 3 (September 1992), pp. 647657. 72. Rosen, Societies and Military Power, p. 241. 73. Fair, Urban Battle Fields of South Asia, p. 90. 74. International Crisis Group, Indonesia: Next Steps in Military Reform, p. 4.

Contending with Terrorism 286

more difcult to fulll. In addition, the lack of popular support facilitates the ability of insurgents to gain recruits and resources. Brutality is a particular problem. The Algerian military is notorious for its atrocities against civilians. In Uzbekistan, torture is widespread,75 and in June 2005, the military red on a peaceful political opposition rally. The United States and Uzbekistan have dramatically reduced military cooperation because of these problems. The Indonesian military was linked to numerous human rights abuses in Aceh, Papua, Central Sulawesi, and Maluku.76 Such actions create sympathy for the guerrillas, particularly when the governments control is weak and people have an option of siding against it with less risk.77 As the above discussion suggests, most potential U.S. allies against alQaida-linked insurgencies do not do well according to these criteria. Indeed, it is no exaggeration to say that several range from poor to abysmal. Tables 1 and 2 provide an overview of militaries ghting al-Qaida-linked insurgencies, using the above criteria to indicate problems they have.

Structural Causes of Tactical, Organizational, and Political Problems

A look at the countries in question indicates that they suffer from several structural weaknesses: illegitimate and repressive regimes, civil-military tensions (particularly with regard to suspicions of a coup), economic backwardness, and social exclusion. Even the democracies among the lot suffer from several of these problems. Table 3 displays the extent of the difculties. These problems, of course, are not universal and, as discussed further, the military culture and broader political problems of counterinsurgency vary accordingly.78 These structural problems should not be surprising. Although the causes of insurgency can be vast, common ones include a lack of popular legitimacy caused by an exclusive government, discrimination, and economic discontent.
75. Jim Nichol, Uzbekistan: Current Developments and U.S. Interests (Washington, D.C.: Congressional Research Service, updated April 21, 2004), p. 6, Order Code RS21238. 76. Aleksius Jemadu, Intelligence Agencies Must Be Held to Account, Jakarta Post, June 20, 2005. 77. See Leites and Wolf, Rebellion and Authority. 78. These structural problems raise the interesting question of why the insurgencies have not won outright victories. The regimes continued survival stems from several sources. First, the insurgents are often poorly led, brutal, and unpopular. Second, they face a much harder collective action problem. Because of the insurgent groups smaller size, it is difcult to push uncommitted individuals to actively side with them. Third, not all of the regimes suffer from every problem identied here. Fourth, the insurgents very structure makes eliminating them hard, but at the same time, this makes it difcult for them to mass forces to engage in mid-intensity combat. Thus there is often a stalemate or low-level strife, with neither side able to gain a decisive advantage.

Table 1. Allied Militaries and COIN: Tactical and Organizational Characteristics Algeria X X X X X X X X X X X X X X X X X X X X X X X X X X X X X X* X X X X X X X X X Egypt India Indonesia Pakistan Philippines Saudi Arabia Uzbekistan

Afghanistan

Poor intelligence Poor integration across units Garrison mentality/low level of initiative Soldiers who do not want to fight Bad officers Bad noncommissioned officers Training, learning, and creativity problems

*The Indian army historically did not integrate well across services in its counterinsurgency operations. In recent years, however, the police, military, and paramilitary forces have worked better together. See C. Christine Fair, Urban Battle Fields of South Asia: Lessons Learned from Sri Lanka, India, and Pakistan (Santa Monica, Calif.: RAND, 2004), p. 70.

Friends Like These 287

Table 2. Allied Militaries and COIN: Political Characteristics Algeria X X X X X X X X X X X X X X X X Egypt India Indonesia Pakistan X X X X Philippines X X X X Saudi Arabia X X X X X X X Uzbekistan

Afghanistan

Poor civil-military integration Bad rapport with outside patrons (United States, etc.) Corruption Military not fighting for a system that can win over the population Vulnerable to insurgent penetration

X X

Table 3. Structural Problems of Counterinsurgency Allies Algeria X X X X X X X X X X X X X X X X Egypt India Indonesia Pakistan Philippines Saudi Arabia Uzbekistan X X X

Structural Problems

Afghanistan

Contending with Terrorism 288

Illegitimate regimes Coup suspicions Economic backwardness Social exclusion

X X X

Friends Like These 289

A weak and dysfunctional government is particularly important, because in such cases even small groups can effectively use violence without being shut down by the government.79 Indeed, the very shortcomings of the regimes often are major reasons for the existence of an insurgency. Thus, at times the very causes of insurgency also create problems for ghting it. The structural characteristics produce the distinct interests for the security forces ghting insurgents. Thus, one of the problems common to alliances in general (as well as inherent in a principal-agent relationship)systematically distinct preferencesis common to U.S. relations with allied security forces. The allies distinct interests can cause many of the above-listed problems with ghting an insurgency. In addition, the regime often does not make the reforms that would improve its ability to ght the insurgents. illegitimate and repressive regimes Many if not most of the regimes facing al-Qaida-linked insurgencies have a legitimacy problem. Of the countries surveyed, only India, the Philippines, and to a lesser degree Afghanistan and Indonesia qualify as democracies. Freedom House reports that Algeria, Egypt, Pakistan, Saudi Arabia, and Uzbekistan are not free, and these countries scored sixes and sevens on their scale, with seven being the least free.80 Being not free does not inherently mean a government is illegitimate, but it is suggestive of a problem. Algerias military government took power so as to overturn elections that Islamists were poised to win peacefully. President Pervez Musharraf of Pakistan also took power in a military coup. The coup had considerable support at the time, but he has subsequently alienated middle-class Pakistanis, while many Islamists view him with suspicion. Uzbekistan is governed by a brutal dictator, and all opposition political activity is banned.81 Egypts leader has held power for a quarter century, with only the trapping of democracy. Saudi Arabia is a monarchy that has some legitimacy even though it is not free, but the regimes corruption and exclusiveness have bred considerable cynicism.82

79. Fearon and Laitin, Ethnicity, Insurgency, and Civil War. In such circumstances, the military becomes vital. As Morris Janowitz notes, The absence of or the failure to develop more effective patterns of political and social control leads military regimes or military-based regimes to rely more heavily on internal police control. Janowitz, Military Institutions and Coercion in the Developing Nations (Chicago: University of Chicago Press, 1988), p. 7. 80. Freedom House, Table of Independent Countries Comparative Measures of Freedom, Freedom in the World, 2005, http:/ /www.freedomhouse.org/uploads/chart33File36.pdf. 81. Jim Nichol, Uzbekistan, p. 5. 82. A lack of democracy does not inherently make a regime illegitimate in the eyes of its people. The al-Saud royal family, for example, has long enjoyed legitimacy (though some argue this is fad-

Contending with Terrorism 290

An illegitimate and repressive regime has several pernicious effects on tactical and organizational aspects of counterinsurgency. The ow of information in authoritarian states is limited, particularly if the information may be perceived as critical of the regime. In such an environment, information is deliberately compartmentalized. Nor are mistakes critically examined or even identied. As a result, the overall quality of intelligence is poor, either because intelligence ofcers lack all the necessary information or because many plausible ndings (e.g., that people are rebelling because the regime is brutal and illegitimate) are suppressed because they are unwelcome at senior levels. In Uzbekistan, the regime has resisted intelligence reform that would enhance the information gained by the military, as it wants to ensure that intelligence is concentrated in the hands of regime loyalists.83 Military regimes such as those in Algeria and Pakistan face particular problems. In contrast to militaries whose senior ofcers must ultimately report to civilian ofcials, military regimes face less pressure to change their procedures and methods in response to problems.84 Militaries that are not accountable to elected leaders and the public in general are less likely to correct mistakes or undertake bureaucratically painful changes. Particular problems may include poor integration across units and services and a lack of creativity when standard procedures fail or when new situations arise. Politically, such a military may be more prone to human rights abuses, as it can cover up any problems and not risk broader censure. Corruption is also a problem, as military gures in power use their positions to enrich themselves at the publics expense, even if it hinders overall military effectiveness. A corrupt military is also a less popular one. Soldiers will not be inspired by their ofcers, and the people in general will see the military more as a parasite than as a savior. Uzbekistans soldiers oppose military service, in part because corruption is widespread, which enables many to buy their way out of serving.85 Not surprisingly, ofcers enriching themselves through their military positions are likely to resist any reforms that increase accountability and oversight or otherwise hinder opportunities for graft.
ing) due to its provision of economic goods, conquest of rivals, and efforts to uphold the kingdoms religious credentials. For the purposes of this article, Saudi Arabia is coded as a legitimate regime, though it is a borderline case. See Mamoun Fandy, Saudi Arabia and the Politics of Dissent (New York: Palgrave Macmillan, 2001); and F. Gregory Gause III, Be Careful What You Wish For: The Future of U.S.-Saudi Relations, World Policy Journal, Vol. 19, No. 1 (Spring 2002), pp. 3750. 83. McDermott, The Armed Forces of the Republic of Uzbekistan, 19922002, pp. 3637. 84. Barry R. Posen, The Sources of Military Doctrine: France, Britain, and Germany between the World Wars (Ithaca, N.Y.: Cornell, 1984), pp. 5354, 225226. 85. McDermott, The Armed Forces of the Republic of Uzbekistan, 19922002, p. 36.

Friends Like These 291

The political problems dwarf the tactical and organizational ones. At the most basic level, the lack of regime legitimacy complicates the ability of the military to portray itself as ghting for a system that its citizens should embrace. The population is less likely to provide intelligence, offer willing recruits, or otherwise support the military; and many soldiers may be sympathetic to the rebels. The regime may rely primarily on repression to stay in power, as it does in Uzbekistan, leading the military to become involved in human rights abuses. In Algeria, this process reached its acme in the mid-1990s, as the military regime there regularly used its forces to commit atrocities. Military leaders may also oppose reforms because they are instruments of the regime and thus they stand to lose power, opportunities for graft, or other benets. In Algeria, the military feared that an Islamist victory at the polls would lead them to lose their power in the country and their nancial inuencea fear that led them to disrupt elections through a coup.86 Another difcult issue is that the military is often opposed to reforms that would take the wind out of the insurgencys sails, such as land reform, greater democracy and accountability, or the ending of discriminatory policies. The military leadership may see such reforms as a threat to their political and social position and thus not worth the potential benets against the insurgents. The lack of legitimacy also poses a difculty for cooperation with a foreign power. Close cooperation with a foreign government can iname nationalism and lead to questions about a governments competence: a particular problem if the government lacks broad support. The widespread unpopularity of the United States in the world today worsens this problem.87 suspicion of a coup In many developing world countries, the military is viewed as a threat as well as an ally of a regime. As a result, governments go to great lengths to coup proof their regimes, emasculating the military in a variety of ways to ensure its political loyalty.88 Egypt, India, Indonesia, Pakistan, the Philippines, Saudi

86. Willes, The Islamist Challenge in Algeria, p. 245. 87. In none of the Arab countries surveyed by the Pew Foundation in July 2005 did a majority have a favorable view of the United States, though the poll did show an increase in U.S. popularity from past rock-bottom levels. See Pew Global Attitudes Project, Islamic Extremism: Common Concern for Muslim and Western Publics, July 14, 2005, http:/ /pewglobal.org/reports/ display.php?PageID 811. 88. Risa Brooks, Civil-Military Relations in the Middle East, in Nora Bensahel and Daniel Byman, eds., The Future Security Environment in the Middle East (Santa Monica, Calif.: RAND, 2003), pp. 129162; and James T. Quinlivan, Coup-proong: Its Practice and Consequences in the Middle East, International Security, Vol. 24, No. 2 (Fall 1999), pp. 131165.

Contending with Terrorism 292

Arabia, and Uzbekistan all have taken steps to control their militaries, even at the cost of their overall effectiveness.89 Coup-proong shapes a military culture in several negative ways. Most important, the senior ofcers are chosen primarily for their loyalty to the regime, not for their competence. In Saudi Arabia, for example, many senior ofcers are royal family members, while others have close ties by marriage and other relationships. As a result, many important skills such as leadership, creativity, and knowledge of military affairs are in short supply. Indeed, in some militaries charismatic and capable leaders are viewed as a threat rather than as an asset. Finally, governments also use corruption as a way to placate military leaders.90 Coup-proong also inhibits the ow of information. The regime discourages leaders from communicating with one another, an effective means of preventing antiregime plotting but one that also inhibits coordination and learning best practices. Training can also suffer. In Egypt, for example, troops are given little independence (or ammunition) when doing training, and exercises are unrealisticin part to prevent a training mission from turning into an attempt to topple the regime. Without the ow of information, integrating forces becomes much harder, as does designing or redesigning procedures in a creative way to handle persistent problems. Poor training and learning structures can stem from coup-proong measures (and a political system where information is guarded) and have a severe impact on COIN effectiveness. Integration will suffer if units cannot train for it. Without training for COIN in particular, integration may prove particularly hard for soldiers who are given standard training for conventional operations to work in small groups, exercise low-level initiative, be discriminate in their use of repower, or otherwise carry out tasks that differ from conventional operations. Many ofcers and NCOs will lack the skills to ght insurgents properly. Without institutions to disseminate knowledge on the best techniques (and to critically appraise what is going wrong), the military will be less likely to adapt new and creative solutions to the problems it encounters.

89. In India, many state parties fear a strong army and oppose a heavy role for it, believing it would be used to weaken their autonomy at the local level. At the federal level, however, there was historically both disdain and suspicion from civilian leaders. Stephen Cohen, India: Emerging Power (Washington, D.C.: Brookings, 2001), p. 110; and Rosen, Societies and Military Power, p. 208. In the Philippines, corruption and politics both shape promotion. See Yet Another Coup in the Philippines? Janes Foreign Report, June 23, 2005. 90. Brooks, Civil-Military Relations in the Middle East.

Friends Like These 293

economic backwardness Many of the countries ghting al-Qaida-linked insurgencies are poor, while others are at best in the middle-income range. Algeria, Egypt, India, Indonesia, Pakistan, the Philippines, Saudi Arabia, and Uzbekistan suffer a range of economic problems, including relatively low per capita gross domestic product, high unemployment, poor infrastructure, and stagnant growth. Again, such backwardness is not surprising: poorer countries are more likely to suffer insurgencies than wealthy ones, and insurgencies themselves are a barrier to economic development. The impact of a poor economy is relatively straightforward. Corruption, of course, is more tempting when overall wages are low (and, indeed, corruption may be particularly common, as it is often a source of poor economic growth). A poor economy can also limit the military budget of a country, making it difcult to pay recruits well, buy better equipment for the force, and expand the size of the military.91 Training may also suffer, because a sophisticated training program is expensive and requires more troops, as some must remain actively engaged while others are being trained. In Uzbekistan, the regime cannot afford to modernize its old equipment, making many reform proposals dead on arrival.92 At times, the army may be huge despite a poor economy, placing particularly heavy strains on the budget. In Egypt, the internal security forces have very low pay and recruiting standardslower than the regular forces.93 One observer noted that young Uzbeks joined former Islamic Movement of Uzbekistans leader Juma Namanganis forces over the governments because at least he pays them.94 Not surprisingly, the quality of personnel may also suffer, as the poor pay and limited resources make other opportunities more attractive.95 In Egypt, for example, much of the enlisted personnel is illiterate, and few have a technical education.96 In Uzbekistan, the armed forces had difculty assimilating U.S. military aid because they lack the technical expertise to maintain and repair the equipment.97 Perhaps 70 percent of the trainees in the Afghan army are il91. For example, much of the Indonesian air force is not operational due to funding shortages. International Crisis Group, Indonesia: Next Steps in Military Reform, p. 12. 92. McDermott, The Armed Forces of the Republic of Uzbekistan, 19922002, p. 38. 93. Kline, Egyptian and Algerian Insurgencies, p. 127. 94. Rashid, A Long, Hot Summer Ahead, p. 29. 95. Of course, if unemployment is high, skilled individuals may seek out the military simply because it offers gainful employment. In Iraq today, many Iraqis appear to be joining the security services for nancial reasons despite the great personal risk it involves. 96. Pollack, The Sphinx and the Eagle. 97. McDermott, The Armed Forces of the Republic of Uzbekistan, 19922002, p. 32.

Contending with Terrorism 294

literate;98 illiteracy is also a problem in India, as the emphasis on caste for recruitment means that there are not enough literate recruits to ll out some regiments.99 Personnel who are less educated and less motivated are less able to gather and process intelligence effectively. The challenge of integration is often particularly difcult. If the overall quality of personnel is low, both the ofcer corps and the NCOs will suffer accordingly as well. Insurgents are better able to penetrate the military, both because they can bribe their way into key positions and because overall disaffection in the ranks makes penetration easier. Training may even be inhibited, as ofcers are reluctant to have soldiers leave their control because they are skimming off soldiers pay and supply requisitions. Even Saudi Arabiaby reputation, a wealthy statehas more than its share of economic problems. Despite its oil riches, overall economic development has been limited, and skyrocketing population growth puts serious strains on the country. Moreover, income distribution is heavily skewed, favoring the royal family and those connected to it.100 social exclusion and divisions In many countries, power in all its forms is held in the hands of a relatively small group of people, who in turn exclude or actively inhibit the rise of other groups. In India, remnants of the caste system have preserved a division between the work of the hands and work of the minds, while several ethnicities are particularly prevalent in the military by tradition. Caste and subcaste often dene regiments and battalions.101 Even many Hindu-Muslim divisions are really about caste differences.102 As a result, individuals are cut off from one another. Clan and region are also important in Uzbekistan, while in Pakistan there is a bias against the Shiite minority in education and state services.103 Saudi Arabia, of course, is dominated by the al-Saud family, and

98. Mitch Frank, Brian Bennett, Anthony Davis, and Michael Ware, Army on a Shoestring, Time Europe, August 26, 2002, pp. 3036. 99. Rosen, Societies and Military Power, p. 209. 100. Alan Richards, Economic Reform in the Middle East, in Bensahel and Byman, The Future Security Environment in the Middle East, pp. 107117. 101. Rosen, Societies and Military Power, pp. 203209. For a broader look at the persistence of caste attitudes in India, see Myron Weiner, The Child and State in India (Princeton, N.J.: Princeton University Press, 1990). 102. Cohen, India, p. 114. 103. International Crisis Group, The State of Sectarianism in Pakistan, Asia Report, No. 95 (April 2005), p. i.

Friends Like These 295

many of its security positions are in the hands of particular tribes, particularly individuals from the Najd region. Such domination has several pernicious effects on the military culture. The ofcer corps may actively disdain much of the rank-and-le if they are of a different, less-regarded, group. Promotions and rewards may also be skewed with individuals from certain groups receiving a preference, while others have a formal or informal ceiling on their rise. In addition, the quality of personnel may suffer, as certain groups may deem the military to be unwelcoming. Even without hostility, ethnic differences create more mundane problems. In Afghanistan, the problem of coordinating across multiple languages has hurt training. In India, promotion in some units depends on preserving caste ratios: if a particular regiment that is home to one caste has no vacancies for more senior positions, soldiers from that unit will not be promoted.104 As a result, basic military tasks are more likely to be performed poorly. Initiative will suffer, and a garrison mentality is more likely as ofcers will not motivate the soldiers. In such a system, one of the hardest military tasksintegration across units and among servicesis less likely to be rewarded. More generally, loyalty to the regime or membership in the right group will count for more than creativity and military excellence. The direct political consequences of exclusion and social divisions relevant to counterinsurgency are also considerable. Politically, the military may be seen as an agent of the ruling clique, not of the nation as a whole. In addition, the military may oppose political and social reforms that disadvantage their members privileged position. In some countries, the military may be representative of the majority population but seen as alien by segments linked to the insurgents. The Philippines military, for example, includes much of the population, but is not seen as representative by the Muslim minority in areas where the Abu Sayyaf Group was active.105 Table 4 loosely summarizes the above points. Needless to say, the various structural weaknesses do not necessarily produce the problem in question, but the weakness does make it far more likely the problem will be present.

104. Rosen, Societies and Military Power, p. 212. 105. The United States has taken steps to offset the dominance of Tajiks in the Afghan military by recruiting more Pashtuns for a range of positions. Joshua Kucera, Afghanistan Looks to Army Expansion, Janes Defence Weekly, October 13, 2004; and Kenneth Katzman, Afghanistan: Post-war Governance, Security, and U.S. Policy (Washington, D.C.: Congressional Research Service, May 19, 2005), p. 30, Order Code RL30588.

Contending with Terrorism 296

Table 4. A Summary of How Structural Problems Create Specific Counterinsurgency Weaknesses Illegitimate and Repressive Regimes Poor intelligence flows Poor integration across units and services Leadership problems Unmotivated and poorquality soldiers Lack of training, learning, and creativity Civil-military integration Corruption Insurgent penetration Lack of support from the population Militaries opposition to reform X Civil-Military Tension X X X X X X X X X X X X Poor Economies Social Discrimination

X X X X

X X X X

Why the Tail Wags the Dog

The United States has few means of controlling its allies because it has few means to do so.106 Indeed, Washingtons very efforts to demonstrate its commitment to some of its allies and to domestic audiences worsen this problem. As Timothy Crawford contends, a rm alliance can embolden the ally to act contrary to Washingtons interest. Two particular problems stand out. The rst is adverse selection: the United States attracts allies that are reckless in their policies and create insurgencies, backing them later as an unintentional reward for their policy mistakes. The second is the moral hazard problem: U.S. support enables allies to avoid useful political reforms or otherwise stay a foolish course because the regime has more resources.107 When the United States pro106. As Mark Pollack argues, information that an agent is not following the wishes of the principal is useful only if the principal can threaten the agent with sanctions. Pollack, Delegation, Agency, and Agenda Setting in the European Community, p. 116. 107. Timothy W. Crawford, Pivotal Deterrence: Third-Party Statecraft and the Pursuit of Peace (Ithaca, N.Y.: Cornell University Press, 1993), pp. 2223. See also Douglas MacDonald, Adventures in Chaos: American Intervention for Reform in the Third World (Cambridge, Mass.: Harvard University Press, 1992).

Friends Like These 297

claims an ally to be vital in the war on terrorism, and when the ally is indeed ghting insurgents who might later abet an attack on the United States, it is difcult to walk away. As D. Michael Shafer noted with regard to U.S. support for governments during the Cold War, The more critical the situation, the less leverage the United States can muster.108 One way out of the moral hazard is through what Alan Kuperman labels restricting insurancethe example he uses is the International Monetary Fund requirement of structural adjustments before bailing out a country.109 While possible in theory, playing such a role would in practice be politically difcult for the United States. The repeated U.S. declarations that ghting alQaida is a vital interest (and the linkage of this to the country in question, which is often necessary to get a program under way for political and bureaucratic reasons at home) undermine the ability of the United States to threaten to cut support.110 Moreover, such declarations provide locals an incentive to exaggerate their insurgents links to al-Qaida. Immediately after the September 11 attacks, governments around the world began declaring that the local ghters they faced had al-Qaida links as a way of legitimating their struggle in Washingtons eyes.111 Before the recent schism with the United States over the regimes abuses, the government of Uzbekistan has even stressed the dangers to U.S. military personnel as a way of pushing the United States to give more support to the government.112 Scale is another problem. As discussed above, solving many of the various military problems requires changing the broader society, economy, and political system: a daunting challenge that requires massive resources to tackle. There will be little U.S. appetite for placing pressure on allies if truly changing their security forces requires transforming them from top to bottom. Compounding this challenge is that the United States instrument of change is often the very regime and military forces that are themselves part of the structural problem. The United States cannot by itself foster economic devel108. D. Michael Shafer, The Unlearned Lessons of Counterinsurgency, Political Science Quarterly, Vol. 103, No. 1 (Spring 1998), p. 64. 109. See Alan J. Kuperman, Suicidal Rebellions and the Moral Hazard of Humanitarian Intervention, in Timothy W. Crawford and Kuperman, eds., Gambling on Humanitarian Intervention: Moral Hazard, Rebellion, and Civil War (New York: Routledge, 2006). The economic analogies, including the International Monetary Fund, appear in ibid., pp. 1216. 110. Metz and Millen, Insurgency and Counterinsurgency in the 21st Century, p. 19. 111. Members of the General Directorate for External Security, interviewed by the author, Paris, France, May 2005. 112. Richard Weitz, Storm Clouds over Central Asia: Revival of the Islamic Movement of Uzbekistan (IMU)? Studies in Conict & Terrorism, Vol. 27, No. 6 (November/December 2004), pp. 465490.

Contending with Terrorism 298

opment in Algeria or political reform in Uzbekistan. Such measures require local regimes to take action. For many local interlocutors, reform is more threatening than the insurgency: Political reform would throw them out of power; military reform might increase the chances of a coup; economic reform would lessen opportunities for corruption; and social reform would hinder their groups hold on power. Not surprisingly, foreign leaders often turn the United States down when it presses for reform. At times, they may halfheartedly embrace reform, going through the motions (and taking U.S. money and resources) but perverting the outcome to ensure the stability of the status quo. The United States also suffers from several moral hazards as it seeks to exert inuence. U.S. support of a government often reduces the necessity for the regime to undertake the reforms required to gain popular support. U.S. backing comes with a degree of legitimacy as well as with nancial and other resources. Thus empowered, governments can put off land reform, stop reining in corruption, and avoid other changes that would hurt the insurgent cause.113 In Uzbekistan, for example, the regime used the U.S. embrace to enhance its legitimacy even as it cracked down on dissent at home.114 Ironically, the United States may be tarred with the brush of a brutal ally, even if it is urging that ally to reform.115 Similarly, U.S. support for a military reduces the militarys need to change its leaders and revise its doctrine, organization, and procedures to better ght the insurgency. Change often comes at the point of a knife: if the United States is doing the ghting for, or even with, locals, they may believe they can carry on with inefcient practices without losing. Money transfers, one of the biggest U.S. levers and in theory one of the key ways that the United States could inuence its allies, are also problematic. Denying the Pakistani security services resources means they are less able to do what the United States seeks.116 Indeed, in the most extreme circumstances, the local military and regime may not want to completely defeat the insurgents, for nancial reasons. Kyrgyz Prime Minister Kurmanbek Bakiev, for example, declared the U.S. military presence in his country to be a gold mine, a

113. Metz and Millen, Insurgency and Counterinsurgency in the 21st Century, p. 19. 114. Prepared Statement of Fiona Hill, reprinted in Central Asia: Terrorism, Religious Extremism, and Regional Instability. 115. Prepared Statement of Stephen Blank, reprinted in Central Asia: Terrorism, Religious Extremism, and Regional Instability, p. 53. 116. Pollack identies a similar general problem in principal-agent relationships. Pollack, Delegation, Agency, and Agenda Setting in the European Community, p. 117.

Friends Like These 299

comment that suggests just how benecial such a presence can be to poor areas.117 Even without U.S. aid, war is often nancially benecial to military leaders. In Algeria, elements of the military wanted to keep the war going with the insurgents indenitely because of these nancial benets.118

Must the United States Go to War with the Allies It Has?

This article has argued that U.S. allies in the struggle against al-Qaida-linked insurgencies are likely to suffer from a range of problems at both the tactical and strategic levels. These problems are not simply the result of poor training or ignorance. Rather, they are structural, growing out of the societal and political characteristics of the regimes and countries in question. Thus, when the United States relies on its allies to ght U.S. enemies, it must recognize not only their limited capabilities but also their fundamentally different interests. To guard against some of the problems that will inevitably arise, the United States should take several steps. A rst is to change the target of intelligence collection. One of the biggest problems the United States faces is that its ally can manipulate U.S. intelligence. It is the ally that gathers the intelligence, and in so doing it can both directly and indirectly shape the views of the United States toward the struggle. The ally can portray insurgents with a local agenda as part of al-Qaida or exaggerate the effectiveness of units receiving U.S. assistance in a bid to increase the size of the overall package. An obvious recommendation is to gather intelligence on the ally itself, particularly with regard to its information collection and dissemination activities. The United States must know what information is being gathered, how well it is being analyzed and disseminated, and what is being passed on to (and, more important, withheld from) the United States. The United States must also diversify the sources of information instead of relying on the area regimes, which can easily manipulate the information. One way to gather information is to rely on third parties.119 The international media would be one source of information, as would independent investigative groups such as the International Crisis Group. These bodies can monitor the ally in a credible way, supplementing the overall information available. They also can do so with less risk of a backlash to the overall bilateral relationship than if an ally discovered the United States was spying on its security services.
117. Quoted in Weitz, Storm Clouds over Central Asia, p. 519. 118. Martinez, The Algerian Civil War, 19901998, p. 232. 119. See Pollack, Delegation, Agency, and Agenda Setting in the European Community, p. 111.

Contending with Terrorism 300

In addition to rectifying information asymmetries, the United States must reconsider how it exerts leverage over its allies and, if possible, try to minimize the moral hazard issue. To increase its chances of success, the United States must recognize that it is not always on the side of the allied government. Rather, Washington should at times act as a third party, helping ght the insurgency but also demanding reforms when possible. Aid and other assistance should be contingent when possible. Given domestic political difculties, however, such nimbleness will be difcult. Peacetime engagement activities such as training, military education programs, and military-to-military contacts can help change a military culture, but the effects are likely to be limited at best, as they cannot change the fundamental difference in interests. U.S. ofcials argue that the Georgian military did not interfere in the recent process of democratic change in part due to the inuence of U.S. training and education programs.120 The United States has made such civil-military training a core part of its efforts to engage militaries in the postcommunist world, though the long-term effects are not yet clear. The new perspectives and skills that are learned in these programs, however, often atrophy or are overwhelmed by the powerful cultural, political, and economic forces that created the dysfunctional military culture in the rst place. Uzbekistan, for example, has been a member of the Partnership for Peace since 1994, yet its military culture remains brutal and corrupt. As a result of these barriers, realistic expectations are necessary. The United States has no ready solutions for the problem of weak states, which is at the core of many of its allies weaknesses in counterinsurgency. Diplomatic pressure and peacetime military engagement activities can help improve a governments effort against insurgencies, but their track record is likely to be spotty at best. Moreover, these efforts may take years or even generations to produce signicant results. Recognizing the difculties in this process and the likely limits will help in designing programs that are more realistic and have the proper expectations.

Implications for Iraq

The United States is likely to encounter many of the problems it has faced with other COIN partners if it transfers security responsibilities in Iraq to Iraqi forces. Iraq is a highly divided society with a weak economy, both of which are
120. Thom Shanker and C.J. Chivers, Crackdown in Uzbekistan Reopens Longstanding Debate on U.S. Military Aid, New York Times, July 13, 2005.

Friends Like These 301

posing the danger of massive civil strife. The Sunni insurgency, though divided, is strong numerically, and in many Shiite parts of Iraq, militias hold sway that are not under the governments control. The regime cannot deliver social services, and crime is rampant. The Iraqi government of Prime Minister Nuri al-Maliki is far from unied, making it likely that military forces will be similarly fractionalized or view the elected leadership with contempt. No one of these problems is inherently insurmountable, but together they will prove exceptionally troublesome in the years to come. They pose particular problems for the preferred exit strategy of both the Bush administration and its Democratic critics: handing over the security of Iraq to the Iraqis. If Iraq follows the experiences of the other countries examined in this article, its security forces will likely be corrupt, brutal, poorly trained, weakly led, unable to coordinate, penetrated by insurgents and militias, and out of step with national political leaders. Such problems suggest a fundamental aw with the preferred U.S. strategy of transferring security to Iraqi forces (unless the U.S. goal is to cynically use such a handoff as an excuse for cutting and running). Instead, the extreme positions may be the more realistic ones: either staying in the country with large numbers of U.S. forces for many years to defeat the insurgents or abandoning the country to its fate of strife and civil war.

Is Counterinsurgency Worth the Effort?

Americans may legitimately wonder whether the United States should still work with allies that are weak, corrupt, antidemocratic, and brutal. Is it simply better to devote resources to homeland defense and otherwise wage this ght without such unsavory partners? This is particularly true in cases where U.S. partners might still face signicant political opposition even if their counterinsurgency capabilities were stronger: the demand for reform created by the governments brutality and incompetence is simply too strong. Americans should keep in mind several counterarguments. First, although U.S. allies range from disappointing to abysmal, the jihadists they ght are typically far worse. The jihadists ideals and practices are bloody and backward, and the United States can be said to be on the least worst side. Second, the area regimes do not want to kill Americans, and the jihadists do. Sheer selfinterest dictates recognition of this difference, however distasteful to the allies. Third, the United States can push reform on its allies. As noted above, this is difcult and likely to suffer many problems, but it is not impossible. These arguments will not hold in every case, however. Washington must be

Contending with Terrorism 302

aware that some of its allies problems may grow despite U.S. help and that this expansion of the U.S. role, in turn, can pose a direct danger to U.S. security by raising hostility toward the United States and tarnishing its overall image. In such circumstances, U.S. resources may be better used for improving the United States defenses or increasing its domestic capabilities rather than further expanding programs to help allies.

NATOs International Security Role

NATOs International Security Role in the Terrorist Era

Rene de Nevers

he North Atlantic Treaty Organizations ongoing engagement in missions ranging from Bosnia to Darfur suggests that the alliance has overcome the doubts about its future that arose after the Cold War. The war on terror that followed al-Qaidas attacks on the United States on September 11, 2001, would appear further to reinforce NATOs signicance. While unilateral actions by the United States and U.S. cooperation with loose coalitions in Afghanistan and Iraq have garnered the bulk of international attention, experts agree that multilateral cooperation is essential in ghting terrorism. Moreover, several of NATOs current activities, such as its missions in Afghanistan and the Mediterranean, are closely linked to the war on terror, with other NATO missions also contributing to this ght. These activities have led NATOs secretary-general, Jaap de Hoop Scheffer, to declare that more than ever, NATO is in demand, and NATO is delivering.1 This apparent vibrancy, however, may not accurately reect NATOs true condition. Although its missions have expanded dramatically since the end of the Cold War and alliance members agree on the threat posed by terrorism, NATOs actual role in the multifaceted struggle against terrorists is minor. This could have long-term implications for alliance unity. This article investigates how the United States has worked with NATO in prosecuting the war on terror. The U.S. government conceives of this struggle broadly, with counterinsurgency and efforts to constrain the spread of weapons of mass destruction (WMD) as essential elements. NATO is the United States premier alliance, and most of Washingtons closest allies are members. But how does NATO contribute to this war on terror? To be sure, NATO is not simply a tool of U.S. policy. The war on terror is a U.S. creation, however, and NATO has been forced to adjust to this fact. The United States perceives
Rene de Nevers is Assistant Professor of Public Administration at the Maxwell School at Syracuse University. The author wishes to thank William Banks, Chantal de Jonge Oudraat, Brian Taylor, Terry Terriff, the participants in the Transatlantic Policy Consortiums Colloquium on Security and Transatlantic Relations, and two anonymous reviewers for their helpful comments on previous drafts of this article. 1. Jaap de Hoop Scheffer, Speech at the 42nd Munich Conference on Security Policy, Munich, Germany, February 4, 2006, http:/ /www.securityconference.de/.
International Security, Vol. 31, No. 4 (Spring 2007), pp. 3466 2007 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

303

Contending with Terrorism 304

terrorism as the key national security threat it will face in the coming years. Just as the United States is working to transform its strategies in response to this threat, we would expect it to evaluate key alliances and security relationships with this measure. I argue that NATO is playing a largely supportive role in U.S. efforts to combat terrorism. The focus of both the European ght against terrorism and the U.S. war on terror lies elsewhere, leaving NATOs contribution to efforts to quell terrorism somewhat tangential. NATO is conducting a defensive mission in the Mediterranean in response to the terrorist threat, and it has adopted strategies ranging from new technology development to consequence management to prevent or mitigate terrorist attacks. In Afghanistan the alliance has assumed a frontline role in seeking to deny terrorist groups a foothold there, making this NATOs rst de facto combat operation ever. But many of the essential elements of the ght against terrorism, such as intelligence sharing, occur outside NATO. Afghanistan aside, NATO members participate in offensive efforts to respond to terrorism outside NATO through bilateral activities or loose coalitions of the willing. There are three main reasons for NATOs limited role: shifts in alignments and threat perceptions caused by systemic changes, NATOs limited military capabilities, and the nature of the ght against terror. The United States needs allies in its ght against terrorism, but does it need the alliance?2 To be sure, the United States values NATO, and indeed has been the driving force behind efforts to expand the alliance by incorporating new members. In addition, NATO has become more than simply a military alliance. Glenn Snyder denes alliances as formal associations of states for the use (or nonuse) of military force, in specied circumstances, against states outside their own membership.3 NATO is far more than this. It is commonly de2. At the end of the Cold War, realist scholars predicted the alliances demise, and during the 1990s, realist and institutionalist scholars sought to explain NATOs longevity. For realist approaches, see John J. Mearsheimer, Back to the Future: Instability in Europe after the Cold War, International Security, Vol. 15, No. 1 (Summer 1990), pp. 556; and Kenneth N. Waltz, The Emerging Structure of International Politics, International Security, Vol. 18, No. 2 (Fall 1993), pp. 4479. For realist discussions of NATOs continued utility, see Charles L. Glaser, Why NATO Is Still Best: Future Security Arrangements for Europe, International Security, Vol. 18, No. 1 (Summer 1993), pp. 550; and Robert J. Art, Why Western Europe Needs the United States and NATO, Political Science Quarterly, Vol. 111, No. 1 (Spring 1996), pp. 139. For institutionalist approaches, see John S. Dufeld, NATOs Functions after the Cold War, Political Science Quarterly, Vol. 109, No. 5 (Winter 199495), pp. 763787; Robert B. McCalla, NATOs Persistence after the Cold War, International Organization, Vol. 50, No. 3 (Summer 1996), pp. 445475; and Celeste A. Wallander, Institutional Assets and Adaptability: NATO after the Cold War, International Organization, Vol. 54, No. 4 (Autumn 2000), pp. 705735. 3. Glenn H. Snyder, Alliance Politics (Ithaca, N.Y.: Cornell University Press, 1997), p. 4.

NATOs International Security Role 305

scribed as a political-military alliance that combines the key political function of guiding members foreign and security policy and providing a forum for alliance consultation with the operational function of ensuring that members can train and develop the capabilities to cooperate militarily.4 This dual role helps to explain why NATO has endured.5 The key issues are whether its members continue to agree on its value and what its core tasks should be, as well as the threat that it confronts. Moreover, if NATOs members do not seek to address their core security threats within the alliance, the alliances military value to its members is likely to be questioned. In the next section, I compare U.S. and NATO strategies for confronting terrorism. I then assess NATOs contribution to the U.S. ght against terrorism. The following section examines factors that help to explain why NATOs contribution to the U.S. war on terror has been relatively limited. I look at three elements: systemic changes and their consequences for NATO, alliance capabilities, and the nature of the ght against terrorism. Finally, I discuss the implications of NATOs elusive role in combating terrorism for U.S. policy and for the alliance.

Comparing Strategies for Confronting Terrorism

The U.S. government widely regards NATO as the most important institution that the United States works with, its premier alliance. Not only do government ofcials point out how much the alliance has done to support U.S. activi4. Celeste A. Wallander and Robert O. Keohane propose that NATO has become a security management institution rather than an alliance. See Wallander and Keohane, Risk, Threat, and Security Institutions, in Helga Haftendorn, Keohane, and Wallander, eds., Imperfect Unions: Security Institutions over Time and Space (Oxford: Oxford University Press, 1999), pp. 2147. 5. Relatively little theoretical research has examined NATOs current situation. Some recent exceptions include evaluations of the alliances effort to adjust to shifting power relations, and of soft balancing as an alternative to traditional balance of power behavior. See Galia Press-Barnathan, Managing the Hegemon: NATO under Unipolarity, Security Studies, Vol. 15, No. 2 (AprilJune 2006), pp. 271309; Seyom Brown, Multilateral Constraints on the Use of Force: A Reassessment (Carlisle, Pa.: Strategic Studies Institute, U.S. Army War College, 2006); Robert A. Pape, Soft Balancing against the United States, International Security, Vol. 30, No. 1 (Summer 2005), pp. 745; T.V. Paul, Soft Balancing in the Age of U.S. Primacy, International Security, Vol. 30. No. 1 (Summer 2005), pp. 4671; and Stephen G. Brooks and William C. Wohlforth, Hard Times for Soft Balancing, International Security, Vol. 30, No. 1 (Summer 2005), pp. 72108. For discussions of alliance persistence and maintenance, see Stephen M. Walt, Why Alliances Endure or Collapse, Survival, Vol. 39, No. 1 (Spring 1997), pp. 156179; Patricia A. Weitsman, Intimate Enemies: The Politics of Peacetime Alliances, Security Studies, Vol. 7, No. 1 (Autumn 1997), pp. 156192; and Barry Buzan and Ole Waever, Regions and Powers: The Structure of International Security (Cambridge: Cambridge University Press, 2003).

Contending with Terrorism 306

ties against terrorists, but they also note that virtually all of NATOs activities today are shaped by the struggle against terrorism.6 Terrorism had emerged as a shared alliance concern by the late 1990s. Terrorist attacks in Saudi Arabia in 1996 and in Kenya and Tanzania in 1998 led the United States to urge NATO to address this threat more seriously. As a result, terrorism was incorporated as a factor contributing to NATOs security challenges in the alliances 1999 Strategic Concept.7 It was not, however, a core focus of NATO policy at that time.8 Terrorism crystallized as a central threat to alliance members only after the September 11 attacks on the United States. NATOs initial response to these attacks was twofold: within twenty-four hours, it invoked article 5 of its charter, declaring that the attacks on the United States represented an attack on all alliance members. It followed this in subsequent weeks with agreement on steps to assist the U.S.-led coalition that attacked al-Qaida and the Taliban in Afghanistan in October 2001. These steps included increasing intelligence cooperation, helping to defend states that were participating in the Afghanistan campaign, allowing overight rights, and deploying naval forces in the eastern Mediterranean and NATO airborne warning and control system (AWACS) planes to the United States.9 These actions set the precedent that the alliances article 5 commitment could stretch beyond the territorial defense of member states to include defense against terrorist attacks.10 NATO also moved quickly to develop military guidelines for responding to terrorism. At the November 2002 Prague summit, members endorsed the new Military Concept for Defense against Terrorism as ofcial NATO policy. The political guidance for alliance actions emphasizes that NATOs goal should be to help deter, defend, disrupt, and protect against terrorist attacks, including by acting against state sponsors of terrorists. In addition, the document iden6. U.S. ofcial, interview by author, Brussels, Belgium, June 7, 2006. 7. To be sure, NATOs European members have long experience with terrorism. For discussions of European responses, see Jeremy Shapiro and Bndicte Suzan, The French Experience of Counter-terrorism, Survival, Vol. 45, No. 1 (Spring 2003), pp. 6798; and Esther Brimmer, ed., Transforming Homeland Security: U.S. and European Approaches (Washington, D.C.: Center for Transatlantic Relations, 2006). 8. NATO, The Alliances Strategic Concept, NATO summit, Washington, D.C., April 24, 1999, http:/ /www.nato.int/docu/pr/1999/p99-065e.htm. 9. Several alliance members also participated in the coalition. NATO, Brieng: Response to Terrorism, Online Library, March 2005, http:/ /www.nato.int/docu/brieng/rtt/html_en/ rtt01.html. 10. Transatlantic Homeland Defense, CTNSP/INSS Special Report (Washington, D.C.: Center for Technology and National Security Policy, Institute for National Strategic Studies, National Defense University, May 2006), http:/ /www.NDU.edu/inss/press/CTNSP-INSS/spl-rpt.pdf.

NATOs International Security Role 307

ties four military roles for alliance operations against terrorism: antiterrorism, or defensive measures; consequence management in the event of an attack against a member state; offensive counterterrorism; and military cooperation with nonmilitary forces.11 In some ways, U.S. strategies to combat terrorism differ from NATO strategies in the goals they emphasize. The U.S. National Strategy for Combating Terrorism, published in September 2006, species four short-term policies to address terrorism: preventing terrorist attacks before they occur; denying WMD to rogue states and terrorist groups; denying terrorist groups sanctuary or support from rogue states; and preventing terrorist groups from controlling any nation that they could use as a base of operations.12 The strategy seeks to ensure that the United States confronts terrorism abroad, not at home. As the U.S. National Security Strategy (NSS) of March 2006 notes, The ght must be taken to the enemy.13 Moreover, in this rubric, prevention refers to offensive counterterrorism activities. The goal of preventing attacks before they occur is not only to stop a planned attack, but to hunt down and capture or kill terrorists determined to attack the United States. The denial elements of the strategy are largely focused on states that might support terrorists. NATOs military guidelines are more defensive and reactive than those of the United States. NATO places greater emphasis on reducing vulnerabilities and enhancing capabilities to respond quickly to potential attacks. In contrast, the United States seeks to keep terrorists from striking the homeland or U.S. interests abroad. And, whereas NATOs military guidelines suggest that its forces could play either lead or supportive roles in offensive operations against terrorists, more planning is recommended before NATO-led offensive operations are undertaken, while the recommendations for support missions are more practical. This indicates the alliances greater comfort and experience with its support role. The goals of NATO and the United States as outlined in these strategies do overlap in important ways. For example, each recognizes the usefulness of multilateral actions and seeks to prevent attacks before they occur. In addition, NATOs counterterrorism strategy shares with U.S. policy the recognition that

11. NATO, International Military Staff, NATOs Military Concept for Defense against Terrorism, updated April 14, 2005, http:/ /www.nato.int/ims/docu/terrorism.htm. 12. The long-term policy goal is to advance democracy as a means to quell support for terrorism. National Security Council, The National Strategy for Combating Terrorism (Washington, D.C.: White House, September 2006), pp. 1117, http:/ /www.whitehouse.gov/nsc/nsct/2006/. 13. George W. Bush, The National Security Strategy of the United States of America (Washington, D.C.: White House, March 2006), p. 8, http:/ /www.whitehouse.gov/nsc/nss/2006/nss2006.pdf.

Contending with Terrorism 308

preventing attacks may require offensive action against terrorists or states that support them. As NATOs Military Concept for Defense against Terrorism states, Allied nations agree that terrorists should not be allowed to base, train, plan, stage, and execute terrorist actions, and the threat may be severe enough to justify acting against these terrorists and those who harbor them.14 The overlap notwithstanding, U.S. strategy documents suggest that NATOs deeply institutionalized, consensus-based model is not the United States preferred approach for multilateral cooperation in the war on terror. Moreover, NATO appears to be less central to U.S. policy and planning. Both the 2002 and 2006 NSS documents promote the formation of coalitions, both within and outside NATO, to address a range of threats.15 More critically, the 2006 NSS makes explicit the U.S. preference for a looser form of cooperation, citing as a model the Proliferation Security Initiative (PSI), an activity designed by the George W. Bush administration to constrain the spread of WMD-related technology. The 2006 NSS states as a goal [the] establish[ment of] results-oriented partnerships on the model of the PSI to meet new challenges and opportunities. These partnerships emphasize international cooperation, not international bureaucracy. They rely on voluntary adherence rather than binding treaties. They are oriented towards action and results rather than legislation or rulemaking.16 The 2006 NSS also states that existing international institutions have a role to play, but in many cases coalitions of the willing may be able to respond more quickly and creatively, at least in the short term.17 Similarly, the 2006 Quadrennial Defense Review (QDR) highlights the distinction between static alliances versus dynamic partnerships and the Pentagons preference for the latter.18 Some Pentagon ofcials insist that the apparent disdain for existing alliances is aimed not at NATO, but at bodies such as the Organization for American States, which, for example, has resisted U.S. efforts to revise its charter in an attempt to isolate Venezuelas president, Hugo Chavez, to punish his anti-U.S. stance. Although NATOs European members are less concerned now that the United States would use NATO as a toolbox than they were immediately after the U.S. invasion of Iraq in 2003,19 they may not be reassured that the United States strongly supports the alliance.
14. NATO, NATOs Military Concept for Defense against Terrorism. 15. George W. Bush, The National Security Strategy of the United States of America (Washington, D.C.: White House, September 2002), http:/ /www.whitehouse.gov/nsc/nss.pdf. 16. Bush, National Security Strategy (2006), p. 46. 17. Ibid., p. 48. 18. Donald H. Rumsfeld, Quadrennial Defense Review Report (Washington, D.C.: Department of Defense, February 6, 2006), p. vii, http:/ /www.defenselink.mil/qdr/report/Report20060203.pdf. 19. I thank an anonymous reviewer for reinforcing this point.

NATOs International Security Role 309

A central question for the United States and NATO is whether their aims correspond to the nature of the terrorist threat. Both U.S. strategy and NATOs military concept initially focused on a particular type of terrorism: the threat posed by al-Qaida at the time of the September 11 attacks. The threat has evolved since then, partly in response to U.S. and allied efforts to disrupt terrorist groups and their activities. The 2006 NSS notes the changed nature of the threat, although the framework of U.S. policy has not changed. For example, whether terrorists need states to support their activities is no longer clear.20 Not only have individuals proven willing to fund terrorist actions, but increasingly, terrorists operate virtually, without formal ties, communicating and even training via websites and other electronic media.21 One result has been the localization of threats, as seen in the terrorist attacks in Madrid in 2004 and London in 2005. These attacks were carried out by homegrown terrorists with no clear ties to al-Qaida. NATO and U.S. strategies thus exhibit both similarities and differences in priorities. The crucial issue is how these strategies affect cooperation between the United States and NATO regarding terrorism.

NATOs Role in the U.S. War on Terror

In this section I assess NATOs contribution to the U.S. war on terror in the following categories: (1) prevention and defense, (2) denial, (3) counterterrorism, and (4) consequence managementall of which are essential to confronting terrorism. These categories incorporate both elements of the U.S. strategy and NATOs political and military efforts to ght terrorism. For each category, I evaluate how NATOs efforts correspond to U.S. goals, as well as to the nature of the terrorist threat. prevention and defense against terrorist attacks Efforts to prevent and defend against terrorist actions fall into two main areas: intelligence sharing and surveillance to detect preparations for an attack. NATO has engaged in both activities, primarily through Operation Active Endeavor (OAE). It is also exploring new technologies to detect and defend against terrorist attacks, a third preventive activity.
20. See, for example, Fareed Zakaria, Terrorists Dont Need States, Newsweek, April 5, 2004, p. 37. 21. See, among others, Ron Suskind, The One Percent Doctrine: Deep Inside Americas Pursuit of Its Enemies since 9/11 (New York: Simon and Schuster, 2006); and Benjamin Wallace-Wells, Private Jihad: How Rita Katz Got into the Spying Business, New Yorker, May 29, 2006, pp. 2841.

Contending with Terrorism 310

operation active endeavor. OAE is NATOs only article 5 operation, and it was the rst substantive military action the alliance took after the September 11 attacks to address the terrorist threat.22 This activity corresponds both to Washingtons goal of preventing terrorist attacks and to NATOs antiterrorism strategy. After deploying in the eastern Mediterranean in October 2001 as a deterrent and surveillance measure in support of the U.S. intervention in Afghanistan, OAE evolved into a broader counterterrorism initiative. It expanded to cover the entire Mediterranean in 2003; and during the U.S. invasion of Iraq, it escorted ships through the Strait of Gibraltar (at the United States request) to alleviate concerns that terrorists might target such ships. OAE has focused on monitoring shipping and the safety of ports and narrow sea-lanes. A second goal, particularly since 2003, has been to expand participation by non-NATO states, both by countries that are formal NATO partners and by countries participating in NATOs Mediterranean Dialogue, a consultative forum intended to improve cooperation with countries in the Mediterranean area.23 OAE has devoted much attention to expanding its intelligence-sharing activities, including efforts to develop a network for tracking merchant shipping throughout the Mediterranean, and improving means to share this intelligence with relevant governments. This should help to address not only terrorist concerns but also alliance efforts to prevent drug smuggling and the spread of weapons of mass destruction.24 OAE has clear military objectives, and NATO has developed valuable experience in maritime surveillance and interdiction through this mission. At the same time, the mission has had both strategic and political aims. NATO has sought to include Russia in OAE, for example, to gain Moscows agreement to extend the operations activities into the Black Sea. Expansion into the Black Sea has not happened, due to objections from both Russia and Turkey to allowing NATO operations there, but Russia participated in OAE patrols in the Mediterranean in 2006.25 Efforts to include more Mediterranean countries are
22. General James L. Jones, NATO: From Common Defense to Common Security, Senate Committee on Foreign Relations, 109th Cong., 2d sess., February 7, 2006, http:/ /foreign.senate.gov/ hearings/2006/hrg060207a.html. 23. NATO, NATO Elevates Mediterranean Dialogue to a Genuine Partnership, Launches Istanbul Cooperation Initiative, NATO Update, June 29, 2004, http:/ /www.nato.int/docu/update/2004/ 06-june/e0629d.htm. 24. Robert Cesaretti, Combating Terrorism in the Mediterranean, NATO Review, No. 3 (Autumn 2005), http:/ /www.nato.int/docu/review/2005/issue3/english/art4.html. 25. Russia was unwilling to agree to this move unless it gained a greater decisionmaking role over the activity, which NATO was not willing to accept. Vladimir Socor, Russians Not Joining NATO Operation Active Endeavor, Eurasia Daily Monitor, Vol. 1, No. 136 (November 30, 2004); Jones,

NATOs International Security Role 311

designed to improve cooperation and, if possible, to share the burden for sustaining the operation with a greater number of countries. This is in keeping with NATOs ongoing efforts to explore expanded partnerships with countries around the globe.26 The United States values OAE because it facilitates intelligence sharing and because it is an alliance-wide activity. Still, NATOs efforts do not always go as far as the United States would like. OAEs guidelines allow it to board only ships whose masters and ag states are willing to comply with the boarding, in keeping with international law. In contrast, the United States is willing to act alone if there is confusion about whether ships will comply.27 This has led to instances in which the United States has boarded ships unilaterally, although based on suspicions raised by OAEs monitoring activities. OAE is NATOs most prominent defense activity, but the alliance has also undertaken numerous surveillance and patrolling missions to defend against possible terrorist attacks. NATO AWACS aircraft conducted surveillance at more than thirty events ranging from NATOs Istanbul summit in 2004 to the 2006 World Cup in Germany. NATO also deployed its new Multinational Chemical, Biological, Radiological, Nuclear (CBRN) task force to the 2004 Athens Olympics. intelligence. Intelligence is widely viewed as the most important tool in preventing terrorist attacks, and the United States shares intelligence regarding terrorist activities with a broad range of countries.28 NATOs intelligence contribution to U.S. efforts against terrorism is limited, however, for four reasons. First, most of the military intelligence NATO relies on, which is shared through the NATO Special Committee, is provided by the United States to the alliance. The Multinational Battleeld Information and Exploitation System, for example, is a near-real-time all-source system through which the United States feeds information to NATO commands.29 Second, the United States and its European allies have diverging views about the role of military intelligence. From the U.S. perspective, military intel-

NATO: From Common Defense to Common Security; and Igor Torbakov, Turkey Sides with Moscow against Washington on Black Sea Force, Eurasia Daily Monitor, Vol. 3, No. 43 (March 3, 2006), http:/ /www.jamestown.org/. 26. NATO, NATO Looks to Global Partnerships, NATO Update, April 27, 2006, http:/ / www.nato.int/docu/update/2006/04-april/e0427c.htm. 27. NATO, Brieng: Response to Terrorism, p. 5. 28. Derek S. Reveron, Old Allies, New Friends: Intelligence-Sharing in the War on Terror, Orbis, Vol. 50, No. 3 (Summer 2006), pp. 453468. 29. Richard J. Aldrich, Transatlantic Intelligence and Security Cooperation, International Affairs, Vol. 80, No. 4 (July 2004), pp. 731753.

Contending with Terrorism 312

ligence is an increasingly important component on the battleeld. The Department of Defense emphasizes that military intelligence is no longer just a staff function, but rather a war-ghting function that soldiers on the battleeld will be actively engaged in at all times. In addition, as part of its broader interest in network-centric warfare, the Defense Department is pushing to establish a fully networked battlespace, with the goal of information dominance in any conict.30 NATOs European members do not place the same degree of emphasis on real-time military intelligence. Third, the capabilities gap that has presented a chronic problem for NATO is increasing in the intelligence area, which suggests growing problems for interoperability. Already in the 1990s, the U.S. military had to maintain legacy communications systems to enable it to operate with other NATO members, and allied forces depended heavily on U.S. communications and intelligence during the 1999 Kosovo bombing campaign.31 One reason the United States rejected some European offers of military assistance in its intervention in Afghanistan in 2001 was the difculties presented by different levels of technological sophistication. The United States spends far more on research and development than its allies; the Defense Departments budget request for research and development for FY 2007 is $57.9 billion. In contrast, the entire defense budget for the United Kingdom, NATOs next largest spender, was $50.2 billion in FY 2006.32 The United States also has a more robust domestic high-technology industry than does any of its European allies. To be sure, alliance members agree on the need for improvements in intelligence capabilities and interoperability. NATO adopted an initiative on developing new capabilities, particularly in areas such as intelligence and surveillance, in November 2002. In addition, some alliance members are working to improve their information warfare capabilities.33 That better intelligence
30. William G. Boykin, Intelligence Support to Allied and Coalition Operations: Strategic Environment for Coalition Warfare, Sixteenth Annual NDIA SO/LIC Symposium, Washington, D.C., February 24, 2005, http:/ /www.dtic.mil/ndia/2005solic/2005solic.html. On network-centric warfare, see Nancy J. Wesensten, Gregory Belenky, and Thomas J. Balkin, Cognitive Readiness in Network-Centric Operations, Parameters, Vol. 35, No. 1 (Spring 2005), pp. 94105; and Paul Murdock, Principles of War on the Network-Centric Battleeld: Mass and Economy of Force, Parameters, Vol. 32, No. 1 (Spring 2002), pp. 8695. 31. David S. Yost, The NATO Capabilities Gap and the European Union, Survival, Vol. 42, No. 4 (December 2000), pp. 97128. 32. NATO Europe spends about $12 billion a year on research and development. See Stephen J. Flanagan, Sustaining U.S.-European Global Security Cooperation, Strategic Forum, No. 217 (September 2005), pp. 16; Steven M. Kosiak, Analysis of the FY 2007 Defense Budget Request (Washington, D.C.: Center for Strategic and Budgetary Assessments, April 25, 2006); and International Institute for Strategic Studies, The Military Balance, 20052006 (London: IISS, 2005), p. 107. 33. Aldrich, Transatlantic Intelligence and Security Cooperation, pp. 745748.

NATOs International Security Role 313

capabilities continue to be problematic is evident in repeated references to the need for improved intelligence sharing both among national agencies and internationally.34 The problem, however, is deeper than merely the need for better intelligence capabilities; NATOs members have developed diverging operational concepts because their military capabilities differ. Differences in their views on the role of information in war ghting are one example of this divergence. The United States approaches the use of force differently than do most European militaries, which means that cooperation on the battleeld could be increasingly difcult.35 Although joint exercises may highlight these differences, they do not necessarily resolve them. Fourth, the most vital terrorist-related intelligence information generated in Europe is outside NATOs scope, because it comes from police and domestic intelligence agencies. The bulk of intelligence sharing within Europe, and between the United States and European states, occurs bilaterally or among select groups of states, not in NATO. U.S. intelligence cooperation is closest with the United Kingdom, and its cooperation with members of the UK-USA network is far more intensive than is its intelligence sharing with other NATO countries.36 The major European mechanism for sharing domestic intelligence is the Berne Group, a club of European intelligence organizations to which the United States does not belong. Three factors hinder greater cooperation in nonmilitary intelligence sharing: the problem of ensuring protection of sources when information is dispersed, differences between the United States and many European allies over appropriate domestic privacy standards, and disagreements over legal constraints on intelligence collection.37 The turmoil caused by reports of secret CIA detention centers in Europe, and Italys indictment of several CIA ofcers for operating illegally on Italian soil when they kidnapped a terrorist suspect, illustrate the differences in views regarding the acquisition and use of intelligence.38
34. NATO, Lessons Learned from Recent Terrorist Attacks: Building National Capabilities and Institutions, Chairmans Report, October 6, 2005, http:/ /www.nato.int/docu/conf/2005/ 050727/index.html. 35. I thank Terry Terriff for pointing this out. 36. UK-USA was established in 1947. Denmark, Norway, Turkey, and West Germany later joined the network. Reveron, Old Allies, New Friends, p. 460. 37. Aldrich, Transatlantic Intelligence and Security Cooperation, pp. 738740. 38. Dana Priest, CIA Holds Terror Suspects in Secret Prisons: Debate Is Growing within Agency about Legality and Morality of Overseas System Set Up after 9/11, Washington Post, November 2, 2005; and Craig Whitlock, Prosecutors: Italian Agency Helped CIA Seize Cleric, Washington Post, July 6, 2006.

Contending with Terrorism 314

technology development. NATOs effort to develop counterterrorism technologies represents the alliances third defensive activity. The Defense against Terrorism (DAT) program was established by the alliances Conference of National Armaments Directors after the 2004 Istanbul summit. Its goal is to develop technologies to help prevent terrorist attacks ranging from the use of improvised explosive devices to rocket attacks against aircraft. Different NATO countries have taken the lead on each of the ten project initiatives.39 This effort has strong U.S. support. As part of the alliances defense investment program, the DAT program reects Washingtons interest in persuading U.S. allies to devote more resources to their military capabilities, and it corresponds to the increasingly high-technology approach to warfare adopted by the U.S. military. Some new defensive technologies, such as mechanisms to detect improvised explosive devices, are currently being developed, but it is too soon to determine whether the program will improve the alliances ability to prevent terrorist attacks. Notably, the United States is using the defense investment program to promote allied transformation goals. NATOs new transformation command, Allied Command Transformation, established after the alliance revised its military command structure in 2002, is based in Norfolk, Virginia, near the U.S. Atlantic Fleets headquarters; and the defense investment program is led by an American, Marshall Billingslea, who worked in the Pentagon on counterterrorism and special operations before moving to NATO. U.S. efforts to use this co-location to promote NATOs transformation have led to some resentment within the alliance that the United States is feeding its views through the new headquarters and the defense investment program.40 NATOs contribution to ongoing efforts to prevent terrorist attacks is thus important, but it many not be central to U.S. policy. While OAEs surveillance activities contribute to tracking potential terrorist movements, the intelligence developed through OAE may be more directly relevant to law enforcement than to military missions. To be sure, U.S. ofcials agree that many NATO members have strong intelligence capabilities, and they can provide valuable intelligence that the United States does not possess. But the most important elements of intelligence gathering in Europe take place outside NATO and are conducted by domestic intelligence organizations. This intelligence is more likely to be shared bilaterally, rather than through NATO. Bilateral intelligence

39. NATO, Defense against Terrorism Program: Countering Terrorism with Technology, Topics, November 14, 2005, http:/ /www.nato.int/issues/dat/index.html. 40. NATO ofcial, interviewed by author, Brussels, Belgium, June 2006.

NATOs International Security Role 315

sharing among key allies continued despite severe strains in political relations in the months prior to the U.S. invasion of Iraq, as revelations about German intelligence cooperation with the United States to designate military targets and civilian locations just prior to the invasion made clear.41 Five European states with substantial terrorism concerns developed their own forum for intelligence cooperation in March 2005. In 2006 the alliance established a new intelligence fusion center to ensure that needed intelligence can be distributed to troops in the eld. It also created the Terrorist Threat Intelligence Unit to provide a forum for joint analysis of nonclassied information. But these are better means to share and interpret information.42 NATO as such does not generate raw intelligence useful to preventing terrorism. Finally, NATOs effort to develop new defense technologies to protect against terrorist attacks reinforces the U.S. goal of promoting allied defense transformation. Whether the program will contribute to NATO or U.S. defense remains to be seen. denial: wmd, sanctuary, and state control Denying terrorists certain weapons and the benets of state support, ranging from use of a states territory to outright control over its government, is a central feature of the U.S. strategy against terrorism. Denial is less evident in NATOs strategy; the goal of disrupting terrorist activities comes closest to the U.S. concept. NATOs contribution to this effort varies considerably; it is marginally involved in efforts to deny terrorists WMD, but it plays a signicant role in efforts to deny terrorists state support. denying access to wmd. Preventing the spread of WMD is a core NATO goal, though preventing terrorist acquisition of these weapons is not. Moreover, NATOs contribution to international efforts to confront the problem of WMD proliferation is complicated by two central questions. First, should denial efforts be primarily multilateral or bilateral? If multilateral, what is NATOs value added in seeking to address proliferation problems? Some NATO members have been involved in cooperative threat reduction efforts in Russia and other former Soviet republics. These nonproliferation efforts have been conducted bilaterally or by small groups of states, however, rather than as multilateral NATO initiatives. The United Kingdom, for example, is work41. Bob Drogin, German Spies Aided U.S. Attempt to Kill Hussein in Aerial Attack, Los Angeles Times, January 12, 2006. 42. NATO, Allied Command Operations, Supreme Headquarters Allied Powers Europe, Launch of the Intelligence Fusion Center in Support of NATO, January 17, 2006, http:/ /www.nato.int/ shape/news/2006/01/060117a.htm.

Contending with Terrorism 316

ing with Russia on projects ranging from dismantling nuclear submarines to developing sustainable employment for scientists and engineers formerly employed in Soviet WMD programs. Canada, France, and Germany have engaged in efforts to develop plutonium disposition methods.43 And several core NATO countriesFrance, Germany, and the United Kingdomhave sought to induce Iran to end its nuclear enrichment program since 2004. The group seeking to deal with Iran is generally referred to as the EU-3, and the broader proposals by these countries and the United States, China, and Russia were presented to Iran in June 2006 by Javier Solana, the European Unions foreign policy chiefnot a NATO representative.44 The only alliance-wide efforts to address WMD proliferation have sought to encourage political dialogue through forums such as the NATO-Russia Council, established in 2002, and consultations with other NATO partner states such as Ukraine. Second, what means are appropriate for preventing terrorists from acquiring WMD? Although alliance members generally agreed that Iraq did possess WMD capabilities,45 this question was at the heart of the bitter 200203 debate over invading Iraq. The differences were over the response, and core NATO allies disagreed with the United States about whether preemption was an appropriate counter to Iraqi efforts to develop WMD. This is not a new debate, but it was sharpened by the ght over Iraq policy.46 U.S. policy favors preemption against the potential spread of WMD, as stated in the 2006 NSS. This reects Washingtons concern that rogue states and terrorists will acquire WMD for use in terrorist attacks. Given alliance differences over means, it is not surprising that NATO does not gure centrally in U.S. counterproliferation activities. Moreover, to the degree that it relies on

43. Tony Blair, The Global Partnership Annual Report, 2005 (London: Department of Trade and Industry, 2005), http:/ /www.dti.gov.uk/les/le14426.pdf; Robin Niblett, ed., Test of Will, Tests of Efcacy: Initiative for a Renewed Transatlantic Partnership, 2005 Report (Washington, D.C.: Center for Strategic and International Studies, 2005), pp. 2531; and Matthew Bunn and Anthony Wier, Securing the Bomb, 2006 (Cambridge, Mass., and Washington, D.C.: Project on Managing the Atom, Harvard University, and Nuclear Threat Initiative, July 2006). 44. Breffni ORourke, Iran: Solana Delivers EU Offer on Nuclear Program, Radio Free Europe/ Radio Liberty Newsline, June 6, 2006, http:/ /www.rferf.org/featuresarticle/2006/06/9ee3bf6e955d-4ccd-a1b7-a6f7db90afae.html. 45. The French government supplied the CIA with contrary information from one of Saddam Husseins cabinet ministers, who was a French spy. According to a former CIA ofcial, however, He said there were no weapons of mass destruction . . . so we didnt believe him. Drogin, German Spies Aided U.S. Attempt to Kill Hussein in Aerial Attack. 46. On the ongoing debate over preemption, see Gareth Evans, When Is It Right to Fight? Survival, Vol. 46, No. 3 (September 2004), pp. 5982; and Peter Dombrowski and Rodger A. Payne, The Emerging Consensus for Preventive War, Survival, Vol. 48, No. 2 (June 2006), pp. 115136.

NATOs International Security Role 317

multilateral efforts to prevent WMD proliferation, the Bush administration has stated its preference for more exible partnerships, such as the Proliferation Security Initiative.47 From the U.S. perspective, the PSI has two main advantages: it is a coalition of the willing, involving only those states that share the PSIs goals, and it is results oriented, emphasizing action rather than legislation or rule making.48 Although many NATO members participate in the PSI, this is not an alliance activity. The United States demonstrated its preference for informal coalitions to deny WMD to terrorists by agreeing with Russia on July 15, 2006, to establish the Global Initiative to Combat Nuclear Terrorism. Like the PSI, this informal agreement is open to states that share the United States concern about nuclear terrorism. There is no plan to establish a treaty or institution to formalize this program; its legal authority is based on the International Convention on the Suppression of Acts of Nuclear Terrorism, signed in 2005, as well as on United Nations Security Council resolutions 1373 and 1540, which proscribe terrorist nancing and the spread of WMD-related materials.49 The initiative seeks to set new standards for securing nuclear materials, engaging in law enforcement, and prosecuting terrorist suspects and their supporters. In this, it resembles the requirements of earlier UN resolutions regarding terrorist nancing and law enforcement. Several NATO members are likely to participate in this initiative, which had its rst meeting in October 2006, but it is not a NATO activity. Although many of NATOs European members share the United States concern about the proliferation of WMD and their acquisition by terrorists, this has not translated into cooperation through NATO to actively confront this security problem. Instead, individual states have worked bilaterally or through alliance consultations to address WMD proliferation. The alliance has done better, however, with regard to denying terrorists sanctuary. denying state support or sanctuary. NATO has contributed substantially to the U.S. goals of denying support or sanctuary from rogue states to

47. Chaim Brown and Christopher F. Chyba, Proliferation Rings: New Challenges to the Nuclear Nonproliferation Regime, International Security, Vol. 29, No. 2 (Fall 2004), pp. 549; and Stephen G. Rademaker, Proliferation Security Initiative: An Early Assessment, hearing before House Committee on International Relations, 109th Cong., 1st sess. (Washington, D.C.: U.S. Government Printing Ofce, June 9, 2005), http:/ /www.internationalrelations.house.gov/archives/109/21699 .pdf. 48. Bush, National Security Strategy (2006), p. 46. 49. Global Initiative to Combat Nuclear Terrorism: Joint Fact Sheet, July 31, 2006, http:/ / en.g8russia.ru/docs/7.html. For the International Convention on the Suppression of Acts of Nuclear Terrorism, see http:/ /www.un.org/sc/ctc/law.shtml.

Contending with Terrorism 318

terrorist groups and ensuring that such groups do not gain control over states. NATOs mission in Afghanistan, the International Security Assistance Force (ISAF), has assumed control over international military forces throughout the country. NATO forces are also involved in ongoing efforts to train Iraqi security forces. The U.S. intervention in Afghanistan that began in October 2001 caused some tension within the alliance. While several NATO states offered to contribute troops to this mission, and NATO declared that the September 11 attacks constituted an article 5 attack against all alliance members, the United States did not seek NATOs participation in the invasion. This reected President Bushs desire to avoid having allies dictate how the war would be fought, as well as the preference among some in the Pentagon to avoid the headaches of having to gain allied consensus on strategy similar to those that had developed during NATOs bombing campaign against Serbia in 1999.50 Equally important was the question of whether NATO allies could contribute the specialized capabilities needed for the campaign the United States was planning.51 Nonetheless, a few NATO countries took part in the initial attack against Afghanistanincluding Denmark, France, Germany, Turkey, and the United Kingdomand NATO AWACS aircraft patrolled U.S. airspace in the fall of 2001, backlling to ensure that U.S. territory was defended while freeing U.S. forces for the invasion.52 ISAF was initially established with UN Security Council authorization under British command in October 2001, after the United States overthrew Afghanistans Taliban government. NATO assumed control of ISAF in August 2003.53 Initially ISAFs mission was limited to patrolling Kabul, but since 2004, ISAF has undertaken a four-stage expansion of its mission into the northern and western provinces of Afghanistan, and later to the south and east. It has also deployed several provincial reconstruction teams, which are based on a

50. Bob Woodward, Bush at War (New York: Simon and Schuster, 2002), p. 81; Terry Terriff, Fear and Loathing in NATO: The Atlantic Alliance after the Crisis over Iraq, Perspectives on European Politics and Society, Vol. 5, No. 3 (2004), p. 424; and Tomas Valasek, NATOs New Roles: A Vision from Inside the Alliance (Washington, D.C.: CDI Terrorism Project, Center for Defense Information, October 19, 2001), http:/ /www.cdi.org/terrorism/nato-pr.cfm. 51. Indeed, initial operations in Afghanistan were conducted by the CIA, which was able to act more rapidly than the Pentagon. See Suskind, The One Percent Doctrine, pp. 1821; Ivo H. Daalder and James M. Lindsay, America Unbound: The Bush Revolution in Foreign Policy (Washington, D.C.: Brookings, 2003), pp. 101102; and Peter van Ham, Growing Pains, NATO Review, No. 3 (Autumn 2005), http:/ /www.nato.int/docu/review/2005/issue3/english/analysis_pr.html. 52. Backlling is one of NATOs support goals in its strategy for confronting terrorism. 53. On NATOs role, see NATO in Afghanistan, last updated December 18, 2006, http:/ / www.nato.int/issues/afghanistan/index.html.

NATOs International Security Role 319

model developed by the U.S. military that combines security and reconstruction functions in an effort to help stabilize the countryside.54 ISAF assumed responsibility for security throughout Afghanistan in October 2006. At that point, it was NATOs largest operation, involving about 31,000 troops, including roughly 12,000 U.S. troops under ISAF command. ISAF represents a valuable contribution to the U.S. goal of denying terrorists sanctuary or allies, given al-Qaidas close ties with the previous Taliban regime and ongoing efforts to pursue al-Qaida members in the border region between Afghanistan and Pakistan. All twenty-six NATO members participate in ISAF, as do ten non-NATO partner countries. At the same time, ISAF has suffered from three signicant problems. First, since 2003 the alliance has been unable to secure sufcient troop commitments to meet the target force size. When NATO took control of the southern and eastern regions of Afghanistan in August 2006, its 31,000-strong force represented about 85 percent of the troops and equipment that NATO commanders had requested for the mission. Since July 2006, NATO troops have confronted far more intense ghting than expected.55 The alliance appealed for more troops in September 2006, but only one member country, Poland, offered to send additional troops.56 At the November 2006 summit meeting in Riga, Latvia, new pledges from member states raised the troop and equipment totals to 90 percent of requirements.57 ISAFs commander at that time, Lt. Gen. David Richards, said that it can manage with the current troop strength, but additional troops would allow it to conduct major operations more rapidly and with less risk to NATO soldiers.58
54. While initially designated provisional, these teams are now called Provincial Reconstruction Teams. Robert Borders, Provincial Reconstruction Teams in Afghanistan: A Model for Postconict Reconstruction and Development, Journal of Development and Social Transformation, Vol. 1 (November 2004), pp. 512; and CARE, NGO Concerns Regarding Deployment of U.S. Military Provisional Reconstruction Teams, ACBAR (Agency Coordinating Body for Afghan Relief) Policy Brief, January 15, 2003, http:/ /www.care.org/newsroom/specialreports/afghanistan/ a_policypositions.asp. 55. Franois Heisbourg points out that NATO is trying to stabilize Afghanistan as the Soviet Union sought to do earlier, but the Soviet Union failed, with three times as many troops. International Perspectives on the Use of Force and Legitimacy, transcript (Washington, D.C.: Brookings, October 11, 2006), p. 32, http:/ /www.brook.edu/comm/events/20061011.pdf. 56. Molly Moore and John Ward Anderson, NATO Faces Growing Hurdle as Call for Troops Falls Short, Washington Post, September 18, 2006. 57. NATO, NATO Boosts Efforts in Afghanistan, last updated December 8, 2006, http:/ / www.nato.int/docu/comm/2006/0611-riga/index.htm. 58. U.S. Department of Defense, DoD News Brieng with General Richards from Afghanistan, news transcript, October 17, 2006, http:/ /www.defenselink.mil/transcripts/transcript.aspx ?Transcript1D?3757. As part of its normal rotation, Richards was replaced as ISAF commander by

Contending with Terrorism 320

Second, many troops in Afghanistan operate under national caveats, whereby governments place limits on what military activities their troops are allowed to do or where they are allowed to go in carrying out their missions. These caveats are problematic for two reasons: they hurt operational effectiveness; and alliance members do not share risks equally, which can cause friction.59 Germanys troops can be deployed only near Kabul, for example, and in 2006 Poland resisted sending additional troops to southern Afghanistan, where they are needed the most. Only six NATO members operate without caveats. The problem is not unique to ISAF; national caveats caused headaches during NATOs peacekeeping mission in Bosnia as well, and they have long been a problem in UN peacekeeping missions.60 Recognition of the operational problems such caveats pose has led to a marked decline in their use, but they have made both multinational cooperation and operations in general more difcult in Afghanistan.61 Caveats tend to creep back in, moreover, as is evident in repeated efforts to eliminate them. NATO leaders agreed to reduce caveats at the 2006 Riga summit, for example, with the result that 26,000 troops of the increased force of 32,000 had broader freedom to act.62 Third, the Afghan leadership fears that the United States will abandon it, and it is unsure what NATOs authority over both the security and counterterrorism mission will mean in the long run. Concern has also been raised about whether NATO has the political will and capabilities to ght a sustained counterinsurgency campaign.63 Since NATO forces assumed responsibility for security in southern Afghanistan, the frequency and intensity of Taliban attacks have increased.64 This renewed ghting forced the United States to reverse plans to reU.S. Gen. Dan McNeill on February 4, 2007. Carlotta Gall, America Takes Over Command of NATO Force as Its Mission Grows, New York Times, February 5, 2007. 59. Some of the national caveats, for example, are imposed for technical reasons, such as aircraft that are not properly equipped to conduct nighttime operations. Philip H. Gordon, Back Up NATOs Afghanistan Force, International Herald Tribune, January 8, 2006. 60. I thank Chantal de Jonge Oudraat for pointing this out. 61. James L. Jones, Prague to Istanbul: Ambition versus Reality, presentation at Global Security: A Broader Concept for the 21st Century, Twenty-rst International Workshop on Global Security, Berlin, Germany, May 710, 2004, pp. 2730; and John D. Banusiewicz, National Caveats among Key Topics at NATO Meeting, American Forces Press Service, Department of Defense, February 9, 2005, http:/ /www.defenselink.mil/news/Feb2005/n02092005_2005020911.html. 62. NATO, NATO Boosts Efforts in Afghanistan. 63. Kristin Archick and Paul Gallis, NATO and the European Union, CRS Report for Congress (Washington, D.C.: Library of Congress, April 6, 2004), Order Code RL 32342; and Seth Jones, Averting Failure in Afghanistan, Survival, Vol. 48, No. 1 (Spring 2006), pp. 111128. 64. This was apparently intended to test the will of NATOs member states to sustain their commitment. U.S. Department of Defense, DoD News Brieng with General Richards from Afghanistan.

NATOs International Security Role 321

duce its military commitment in Afghanistan and led the British to expand their troop contribution to ISAF.65 The United States decided in January 2007 to extend the tours of 3,200 troops in Afghanistan, and further troop increases were under consideration.66 Notably, U.S. forces, ISAFs largest contingent, will continue to conduct the bulk of counterterrorism activities aimed at alQaida. The U.S. military also retains 11,000 troops outside ISAFs command to sustain a separate counterinsurgency function in addition to peacekeeping.67 NATO has played a far smaller role in Iraq. Whereas the Bush administration sought to frame the March 2003 invasion as part of the war on terror, the alliance remained deeply split and did not formally participate in the invasion. The NATO Council never discussed Iraq, an indication of the depth of discord within the alliance. NATO did offer some support for the U.S.-led operation, however. It contributed to Turkeys defense against possible Iraqi retaliation during the invasion, and it agreed to the Polish governments request for allied support when Poland took over leadership of one sector of the stabilization force in Iraq in May 2003.68 Fifteen NATO states have contributed forces to the coalition since 2003.69 And although Germany did not participate in the coalition and strongly opposed the U.S. invasion, it deployed hundreds of chemical and biological weaponsdetection troops in Kuwait and Turkey to aid coalition forces in the event of a WMD attack.70 In 2004, at U.S. urging, NATO agreed to play a central role in training Iraqi security forces. NATOs training effort has several elements: mentoring of Iraqi military ofcers by NATO personnel; creation of an ofcer training facility in Iraq; and training of Iraqi ofcers in NATO facilities. NATOs target is to train 1,000 ofcers inside Iraq annually, and 500 outside the state; by September

65. Eric Schmitt, Springtime for Killing in Afghanistan, New York Times, May 28, 2006; and Britain to Send More Troops to Afghanistan, Agence France-Presse, July 10, 2006. 66. Gall, America Takes Over Command of NATO Force. 67. Amin Tarzi, Afghanistan: NATO Expansion Demands Common Approach, Radio Free Europe/Radio Liberty, October 6, 2006, http:/ /www.rferl.org/featuresarticle/2006/10/18ba3c2dcd22-4963-af8d-52ac0963832b.html. 68. Agreement to aid Turkey was reached in spite of French resistance by taking the decision to NATOs military committee, of which France is not a part. On Polands request, see NATO, NATO Council Makes Decision on Polish Request, NATO Update, May 22, 2003, http:/ / www.nato.int/docu/update/2003/05-may/e0521b.htm. On NATO debates over aid to Turkey, see NATO, Consultations on Measures to Protect Turkey, NATO Update, March 6, 2003, http:/ / www.nato.int/docu/update/2003/02-february/e0210a.htm. 69. Some of these troop contributions are quite small. Participating states include Bulgaria, the Czech Republic, Denmark, Estonia, Hungary, Italy, Latvia, Lithuania, the Netherlands, Norway, Poland, Romania, Slovakia, Spain, and the United Kingdom. 70. Daalder and Lindsay, America Unbound, p. 190.

Contending with Terrorism 322

2006, NATO had trained 650 Iraqi ofcers in European facilities and roughly 2,000 ofcers overall.71 NATO has also donated military equipment to Iraqs security forces. This equipment comes primarily from former Warsaw Pact countries that have become NATO members, and it is compatible with Iraqs Soviet-supplied military hardware. NATOs training mission has faced signicant difculties, however. First, the need to gain consensus on all decisions hamstrung efforts to get the mission up and running and greatly slowed the process; residual bitterness over the U.S. decision to invade Iraq contributed to this problem. Some members objected to the precedent set by taking on the training mission, which also slowed decisionmaking.72 Second, as in Afghanistan, some troop contributions have operated under national caveats, which has hindered commanders efforts to coordinate NATOs activities. Third, funding for the mission has been a serious problem. Countries contributing troops are expected to cover their own costs. NATO set up a trust fund to pay for the establishment of a defense university in Iraq, but contributions to the fund have thus far been insufcient. As a result, although the Iraqi government has stressed its preference for in-country training to help gain popular trust and support for the new security forces, more ofcers have been trained outside Iraq.73 offensive measures: counterterrorism NATOs offensive role in ghting terrorism is limited. Although its Military Concept for Defense against Terrorism stresses that NATO must be prepared to take on offensive missions if required, the alliance has operated primarily in support of member state efforts to conduct offensive counterterrorist operations. NATO ofcials point out that this is appropriate given that the alliances task is to defend its members and support them in the event of attacks, rather than to take the lead in offensive operations.74
71. Rick Lynch and Phillip D. Janzen, NATO Training Mission Iraq: Looking to the Future, Joint Force Quarterly, No. 40 (1st Quarter 2006), p. 33; IISS, The Military Balance, 20052006, p. 45; and NATO, Iraq Gets National Defense College, September 3, 2006, NATO Update, http:/ /www .nato.int/docu/update/2006/09-september/e0903a.htm. 72. NATO ofcials, interviewed by author, Brussels, Belgium, June 8, 2006. 73. Lynch and Janzen, NATO Training Mission Iraq, pp. 3234. The funding problem is not unique to this mission; the NATO Response Force has been referred to as a reverse lottery because troop-contributing nations must bear all the costs. De Hoop Scheffer, Speech at the 42nd Munich Conference on Security Policy; and Jones, NATO: From Common Defense to Common Security. 74. Indeed, some members, notably France, continue to insist that NATOs military role should be limited to article 5 operations. NATO ofcials, interviewed by author.

NATOs International Security Role 323

The only example of NATO assuming a combat role is found in Afghanistan, where ISAF took control over security operations throughout the country in October 2006. Although the security problems on the ground in Afghanistan range from crime and drug trafcking to counterterrorism, ISAFs rules of engagement do not explicitly cover missions other than peacekeeping; ISAF does not have a formal counterterrorist mission.75 ISAFs limited mandate is due in large measure to member states nervousness about the prospect of taking on counterterrorism or counterinsurgency responsibilities, as well as concern that acknowledging the potential combat elements of the mission would make it even harder to obtain sufcient troop commitments to ISAF. NATOs mission in Afghanistan shifted from a primary focus on reconstruction and stability to counterinsurgency when it took responsibility for security in the southern region of the country at the end of July 2006, and it adopted more robust rules of engagement.76 Since then, British, Canadian, and Dutch forces on the ground have been engaged in counterinsurgency operations. Commanders dropped their earlier vagueness about their activities in Afghanistan and now acknowledge the war-ghting nature of the mission. Lieutenant General Richards of ISAF stated in August 2006 that NATOs goal was to strike ruthlessly at Taliban ghters seeking to undermine the Afghan government.77 Similarly, in parliamentary debates over extending Canadas deployment to Afghanistan in May 2006, members openly acknowledged the combat nature of the mission.78 NATOs mission in Afghanistan is of crucial importance to the alliance. NATO troops are engaged in their most militarily challenging mission since the alliance was formed in 1949. NATO leaders regarded the mission as a suc-

75. Ali A. Jalali, The Future of Afghanistan, Parameters, Vol. 36, No. 1 (Spring 2006), pp. 419. 76. Oral Statement of General James L. Jones, United States Marine Corps, Supreme Allied Commander, Europe, before the Senate Foreign Relations Committee, 109th Cong., 2d sess., September 21, 2006. 77. Richard Norton-Taylor, British General Takes Command and Promises Ruthless Strikes on Taliban, Guardian (London), August 1, 2006. British forces had killed at least 700 combatants in ongoing battles with Taliban forces in southern Afghanistan by August 2006. It is notoriously difcult, however, to determine the number of Taliban casualties. According to one estimate, 4,000 people died in ghting in 2006, many of whom were believed to be rebels. See France to Withdraw Special Forces from Afghanistan, Agence France-Presse, December 17, 2006, http:/ / www.france24.com/france24Public/en/news/france/20061217-Afghanistan.html; Thomas Harding, Paras Claim 700 Taliban Lives, Telegraph, July 29, 2006; and Taleban Death Toll Inaccurate, BBC News, December 10, 2006, http:/ /news.bbc.co.uk/2/hi/south_asia/6166577.stm. 78. Quoted in Stephen Harper, Canadas Commitment in Afghanistan, 39th Parl., 1st sess., May 17, 2006, http:/ /www.parl.gc.ca/39/1/parlbus/chambus/house/debates/025_2006-05-17/ HAN025-E.htm#OOB-1543354.

Contending with Terrorism 324

cess at the end of 2006; the alliance has shown the ability to ght and to maintain support for the mission. But continued success is not guaranteed. Both NATO and U.S. commanders expressed growing concern in late 2006 that without signicant advances in reconstruction and development in Afghanistan, the military effort would ultimately fail. As Lieutenant General Richards noted, Fighting for its own sake in a counterinsurgency will get us nowhere over time.79 The military commanders argued that the international community had failed to convince the Afghan people that international involvement would make their lives better in terms of either security or living standards. If it did not do so soon, the resurgent Taliban would win their support, or at least acquiescence.80 There are also worrisome hints, however, that some NATO members may be unwilling to sustain their commitments to ISAF. Not only has the alliance had difculty convincing members to send more troops to Afghanistan, but some governments were starting to pull out troops at the end of 2006. The French government decided to withdraw 200 special forces troops from the southern region of Afghanistan in December 2006, for example, although some 1,100 French troops would remain near Kabul.81 The Italian government has faced pressure to set a deadline for withdrawal of its 1,800 troops in ISAF.82 The ambivalence toward ISAF is partly explained by the fact that even governments that acknowledged the dangers confronting their troops were surprised by the strength of the insurgency in Afghanistan; in other participating states, the failure to prepare their publics for the true nature of the mission damaged support for ISAF. Finally, the Afghan mission suffers from the European publics residual resentment about Iraq, with which it is often associated.83
79. U.S. Department of Defense, DoD News Brieng with General Richards from Afghanistan. 80. Ibid.; and Oral Statement of General James L. Jones. On the fragility of the situation in Afghanistan, see also Barnett R. Rubin, Still Ours to Lose: Afghanistan on the Brink, prepared testimony for the House and Senate Committees on Foreign Relations (Washington, D.C.: Council on Foreign Relations, September 21, 2006). 81. France to Pull Troops Fighting against Taliban in Afghanistan, New York Times, December 18, 2006. In both the Netherlands and Canada, parliaments raised serious questions about sending troops to Afghanistan; both Dutch and Canadian troops are actively engaged in ghting insurgent forces there. Canada Committed to Afghan Mission, Harper Tells Troops, CBC News, March 13, 2006, http:/ /www.cbc.ca/world/story/2006/03/13/harper_afghanistan060313.html; and More Dutch Troops for Afghanistan, BBC News, February 3, 2006, http:/ /news.bbc.co.uk/2/hi/ europe/4673026.stm. 82. Helene Cooper, NATO Allies Wary of Sending More Troops to Afghanistan, New York Times, January 27, 2007. 83. Moore and Anderson, NATO Faces Growing Hurdles as Call for Troops Falls Short.

NATOs International Security Role 325

consequence management The U.S. strategy for confronting terrorism gives priority to preventing attacks. In contrast, NATOs Military Concept emphasizes consequence management and has two main goals: to respond to an attack once it occurs, and to minimize the effects of an attack that has taken place. At the 2002 Prague summit, the alliance adopted a Civil Emergency Planning Action Plan to improve its ability to respond to attacks involving WMD, and it has developed a range of capabilities to respond to such attacks, including the multinational CBRN Defense Battalion, a deployable analytical laboratory, an event response team, a chemical and biological defense stockpile, and a disease surveillance system.84 The alliance has also established a center of excellence to further explore defenses against WMD. NATO has also expanded its disaster relief plans to enhance its terrorism response capabilities. The alliance has conducted several WMD-related exercises, and its Euro-Atlantic Disaster Response Center has assisted in responding to oods, forest res, and snowfalls in states both within and outside the alliance, as well as to Hurricane Katrina in the United States. NATO also contributed to disaster relief operations after a major earthquake in Pakistan in October 2005. This was considered a signicant step because it represented the rst time the alliance had offered disaster assistance outside its own geographic area, which could set a precedent for such operations. Some analysts have argued that NATO should expand its consequence management abilities because this is a logical role for the alliance, and it would encourage NATO members to do more to enhance their militaries capabilities to respond and coordinate in the event of terrorist attacks or other disasters. There are two obstacles to expanding NATOs role in this area. First, NATO can act only if its members request assistance. For example, NATO was not asked to aid Spains response to the terrorist bombings in Madrid in 2004, nor did it respond to the 2005 London bombings.85 Whether other states will turn to NATO in the future is likely to depend on the gravity of the attack and on

84. Dagmar de Mora-Figueroa, NATOs Response to Terrorism, NATO Review, No. 3 (Autumn 2005), http:/ /www.nato.int/docu/review/2005/issue3/english/art1.html; Eric R. Terzuolo, Combating WMD Proliferation, NATO Review, No. 3 (Autumn 2005), http:/ /www.nato.int/ docu/review/2005/issue3/english/art3.html; and Boosting NATOs CBRN Capabilities, NATO Review, No. 3 (Autumn 2005), http:/ /www.nato.int/docu/review/2005/issue3/english/features1 .html. 85. Jean-Yves Haine, Military Matters: ESDP Transformed? NATO Review, No. 2 (Summer 2005), http:/ /www.nato.int/docu/review/2005/issue2/english/military_pr.html; and NATO, Press Conference by NATO Secretary-General on the Extraordinary Meeting of the North Atlantic Council, Online Library, July 8, 2005, http:/ /www.nato.int/docu/speech/2005/s050708a.htm.

Contending with Terrorism 326

NATOs ability to provide specialized capabilities such as CBRN. Second, European states disagree about whether this function should be carried out by NATO or by the European Union. Moreover, although the need for greater cooperation between NATO and the EU in areas such as emergency response is widely recognized, little progress has been made in this area. summary This examination of NATOs participation in the U.S. war against terror yields three main insights. First, the chronic capabilities gap between the United States and its allies is growing, and increasingly may be limiting the alliances operational cooperation. Second, the gap between what the United States and European states are willing to do militarily is also growing. This reects their disparate capabilities and the differences in operational planning that follow from them. Third, the United States commitment to working through the alliance is unclear. While NATOs chief military ofcer stresses the value of NATOs military mechanisms and political consultation, the U.S. government is often unwilling to rely on NATO in campaigns that relate directly to defending the United States. Not only do Pentagon leaders want to avoid the effort of working to build consensus within the alliance, but they see it as compromising the mission and the safety of U.S. forces. Moreover, some U.S. government ofcials note that the Bush administration sees NATO as unreliable, because of the difculty in gaining troop commitments from member states in recent years. This has fostered the view expressed by one State Department ofcial, who noted, We ad hoc our way through coalitions of the willing. Thats the future.86

NATOs Limited Role in the U.S. War on Terror

Three factors help to explain why NATOs contribution to the U.S. war on terror has been relatively limited: shifting alignments and threat perceptions due to systemic changes, NATOs chronic and growing capabilities gap, and the war against terror itself. alignments and threats after the cold war Two critical changes in the nature of the international system have inuenced NATOs evolution since the collapse of the Soviet Union in 1991. First, the dis86. Guy Dinmore, U.S. Sees Coalitions of the Willing as Best Ally, Financial Times, January 4, 2006.

NATOs International Security Role 327

tribution of power in the international system is no longer bipolar. For all its tensions, the international system was relatively stable during the Cold War, with the two superpowers and their allies aligned in opposition. This made core alliance cohesion relatively easy to maintain. NATOs European members were unlikely to defect from the U.S.-led alliance, and the United States was unlikely to abandon Europe, though fears to the contrary emerged periodically. Although scholars continue to debate whether the current international system is unipolar, multipolar, or something else, today U.S. military power dwarfs that of other powers in the international system. Second, a security community developed among the European powers, Japan, and the United States concurrently with the erosion and eventual collapse of Soviet power. As a result, the option of war between these powers has become virtually unthinkable. This is an equally momentous change in the international system, given that great power war has been a constant in history.87 NATO has deed realist predictions that it would not survive in the absence of the threat it was created to defend against: the Soviet Union. But the aforementioned systemic changes have led alliance members to perceive security threats differently. They have also renewed uncertainties about alliance stability and the U.S. commitment to NATO. The cross-cutting pressures created by changing and sometimes competing interests have made it more difcult to reach consensus on how to deal with security threats confronting NATO.88 The disappearance of the Soviet threat made intra-alliance differences on issues including trade, the environment, and human rights more salient. NATOs European members also responded to concern about U.S. willingness to act through NATO, and Europes inability to act collectively without the United States, by expanding efforts to build a viable European security identity that would make it possible for European states to operate independent of the United States.89 NATO has always been a political-military alliance, rather than a purely military union. It has had a long-standing goal of cementing its members shared democratic ideology and values,90 as evidenced by NATOs expansion in the

87. Robert Jervis, American Foreign Policy in a New Era (New York: Routledge, 2005), pp. 1136. See also Karl W. Deutsch, Sidney A. Burrell, and Robert A. Kann, Political Community and the North Atlantic Area: International Organization in the Light of Historical Experience (New York: Greenwood, 1969), pp. 58; and Buzan and Waever, Regions and Powers, pp. 343376. 88. Brown, Multilateral Constraints on the Use of Force, pp. 1921. 89. Press-Barnathan, Managing the Hegemon, pp. 290301. 90. Thomas Risse-Kappen, Collective Identity in a Democratic Community: The Case of NATO,

Contending with Terrorism 328

1990s.91 In expanding eastward, NATO accomplished two objectives: it extended the European security community, and it removed the potential threat of unstable countries along the borders of member states. This expansion, however, has led to further divergences in threat perception among alliance members. While some of NATOs new members continue to have territorial security concerns due to their proximity to Russia, some long-standing members feel less threatened by traditional security concerns. Moreover, European perceptions of the gravity of the terrorist threat vary widely. Some states perceive the threat to be limited, whereas others view it as signicant. Notably, only ve statesFrance, Germany, Italy, Spain, and the United Kingdomhad legislation dealing with terrorism before the September 11 attacks. NATOs members also differ on the means to respond to threats confronting the alliance. This was most apparent in the bitter dispute over the 2003 U.S. invasion of Iraq. The dispute illustrated three points of disagreement. First, it reected different understandings of the nature of the terrorist threat and how to combat it. Second, it exposed deep differences about the appropriate use of force, and in particular about the U.S. policy of preventive war. Whereas the United States insisted that the urgency of the threat posed by Saddam Husseins supposed possession of WMD mandated immediate action, several European allies argued that Hussein was contained and could be deterred. Third, the dispute illustrated increasing European concern about U.S. unilateralism and the fear that NATOs European members might be entrapped by their alliance commitments to support a reckless military operation.92 As a result, both France and Germany balked at supporting the United States.93 Although the Bush administration sought to repair relations with key European allies and institutions after the 2004 presidential elections, the acrimony caused by this dispute has left a residue of ill will. The shifting alignments and attitudes toward threats confronting NATO have reduced the United States willingness to accept alliance constraints.94
in Peter J. Katzenstein, ed., The Culture of National Security: Norms and Identity in World Politics (New York: Columbia University Press, 1996), pp. 357399. 91. On NATO expansion, see James M. Goldgeier, Not Whether but When: The U.S. Decision to Enlarge NATO (Washington, D.C.: Brookings, 1999); David S. Yost, NATO Transformed: The Alliances New Roles in International Security (Washington, D.C.: United States Institute of Peace Press, 1998); and Jonathan Eyal, NATOs Enlargement: Anatomy of a Decision, International Affairs, Vol. 73, No. 4 (October 1997), pp. 695719. For a contrary view, see Dan Reiter, Why NATO Enlargement Does Not Spread Democracy, International Security, Vol. 25, No. 4 (Spring 2001), pp. 4167. 92. Press-Barnathan, Managing the Hegemon, pp. 301303. 93. Brown, Multilateral Constraints on the Use of Force, p. 19. 94. To be sure, frictions over U.S. unilateralism are not new to the alliance. Similar tensions were evident during President Bill Clintons administration, which rejected several attempts to establish

NATOs International Security Role 329

Moreover, the United States strategic focus has changed, with greater attention being given to the Middle East, Central Asia, and East Asia. This is evident both in the changing base deployments in Europe and the State Departments decision to shift at least 100 diplomatic positions from Europe to other regions, including Africa, South Asia, East Asia, and the Middle East.95 This move is a logical step and if anything overdue, given the end of the Cold War, but it is telling of shifts in U.S. policy priorities. Differences in member states views of the role of military force have probably contributed to U.S. frustration with alliance constraints. National caveats attached to military operations sometimes reect limited capabilities, but they can also reveal different political goals. U.S. inability to count on allies for military cooperation in Iraq, Afghanistan, and elsewhere has produced great frustration within the Bush administration. This is evident in Secretary of Defense Donald Rumsfelds comment, Its kind of like having a basketball team, and they practice and practice and practice for six months. When it comes to game time, one or two say, Were not going to play. Well, thats fair enough. Everyone has a free choice. But you dont have a free choice if youve practiced for all those months.96 alliance capabilities From the United States perspective, the perennial frustration regarding NATO is the difculty of convincing its European allies to increase defense spending and thus to improve capabilities. During the Cold War, NATO members agreed that defense spending should be roughly 3 percent of a states gross domestic product; in 2006 the minimum spending level set by the alliance was 2 percent. Yet only six states other than the United States met this threshold.97 Furthermore, how this money is spent causes concern in Washington. Many states continue to expend the bulk of their resources on manpower, rather than on transforming their forces in ways the United States hopes they willand to which NATO has agreed. Some, such as Greece and Turkey, still have military forces focused at least in part on each other.
international institutions that it believed might undermine U.S. national security, such as the 1997 Ottawa Convention, which banned antipersonnel land mines, and the International Criminal Court. 95. U.S. Department of State, Ofce of the Spokesman, Transformational Diplomacy, January 18, 2006, http:/ /www.state.gov/r/pa/prs/ps/2006/59339.htm. 96. Quoted in Banusiewicz, National Caveats among Key Topics at NATO Meeting. 97. Donald H. Rumsfeld, Speech at the 42nd Munich Conference on Security Policy, February 4, 2006, http:/ /www.securityconference.de.

Contending with Terrorism 330

As NATO has taken on new tasks, differences in capabilities have hindered the alliances operational cooperation. Moreover, willingness to address this gap appears limited; although many European states acknowledge the problem, defense spending and military transformation remain low domestic priorities in several states because they do not see major military threats to their security. Additionally, the large number of operations in which NATO is currently engaged means that many states do not have the manpower and resources to devote to transformation given current budgets. Notably, the United States has funded operations in both Iraq and Afghanistan through supplemental budget requests, rather than the annual defense budget. It has also nanced some allied contributions, such as the Polish division in Iraq.98 U.S. ofcials would like to see NATO undertake a transformation similar to what the United States is adopting: that is, to develop expeditionary forces and what Defense Department ofcials call twenty-rst century military capabilities. The alliances adoption of transformation as one of its goals in 2003 represents a welcome step in this direction; indeed, NATOs reorganized command structure designates one of its two new commands the Allied Command Transformation.99 The United States and some other alliance members disagree over the kinds of capabilities NATO members should seek to acquire. The QDR afrms U.S. support for efforts to create a NATO stabilization and reconstruction capability and a European constabulary force that could build on the EUs existing constabulary forces. It suggests that allied states should aim to [tailor] national military contributions to best employ the unique capabilities and characteristics of each ally, [to achieve] a unied effort greater than the sum of its parts.100 This implies the development of niche capabilities. If the alliance moves in this direction, it could reinforce the perception that the United States, along with only a few alliance members such as the United Kingdom and
98. Ofce of the Press Secretary, Fact Sheet: Request for Additional FY 2004 Funding for the War on Terror, September 8, 2003, http:/ /www.whitehouse.gov/news/releases/2003/09/200309081.html. 99. Brieng on NATO Military Structure, August 2005, http:/ /www.nato.int/docu/brieng/ nms/nms-e.pdf; and NATO, Allied Command Operations, Supreme Headquarters Allied Powers Europe, New NATO Command Structure, May 4, 2004, www.nato.int/shape/issues/ncs/ ncs_index.htm. 100. Rumsfeld, Quadrennial Defense Review, p. 88. On constabulary forces for NATO, see David T. Armitage Jr. and Anne M. Moisan, Constabulary Forces and Postconict Transition: The EuroAtlantic Dimension, Strategic Forum, No. 218 (November 2005), pp. 17; and Hans Binnendijk and Richard L. Kugler, NeededA NATO Stabilization and Reconstruction Force, Defense Horizons, No. 45 (September 2004), pp. 18.

NATOs International Security Role 331

France, has combat-ready troops able to conduct frontline operations, while NATO as a whole is relegated to postconict operations and cleanup. The need for such postconict capabilities is increasingly apparent, notably in both Afghanistan and Iraq. Less clear, however, is whether NATO states are willing to codify this division of labor, and how this would affect alliance unity and decisionmaking. Yet there is an element of hypocrisy in Washingtons annoyance with its NATO allies, because the U.S.-backed expansion of the alliance contributed to the erosion of NATOs military capabilities. The United States was the strongest voice encouraging two rounds of NATO expansion to incorporate Central European states after the Cold War, in 1999 and 2004, and it supported the decision at the Riga summit to consider further enlargement of the alliance as early as 2008. When the second expansion decision was made in 2004, however, it was already apparent that the states that joined NATO in 1999 had failed to meet the goals set for transforming their militaries to accord with NATO standards; and they were not likely to achieve them soon. Moreover, none of the second-round applicants met NATOs military capability standards, and they were far behind the rst-round states in their ability to transform their militaries.101 Lessened alliance capabilities may have been unproblematic to the United States because it did not intend to incorporate alliance forces in its major operations. the nature of the war on terror The nal factor explaining NATOs limited role in the war on terror is the very nature of that conict. Although the Bush administration determined that this is a war, it is not one in which most battles are fought by military forces. Different means are required as well. Indeed, this is partly why Europeans tend to refer to the ght, rather than war, against terrorism. Combating international terrorism requires cooperation in a wide range of areas among different countries and international organizations. Key elements of the struggle include diplomatic efforts to maintain and strengthen international treaties and norms proscribing terrorism, as well as economic cooperation to nd and eliminate sources of terrorist nancing. In addition to military activities, security cooperation is needed to ensure that states can share intelligence regarding ter-

101. On this point, see in particular Wade Jacoby, Is the New Europe a Good Substitute for the Old One? International Studies Review, Vol. 8, No. 1 (March 2006), pp. 178197; and Zoltan Barany, NATOs PostCold War Metamorphosis: From Sixteen to Twenty-six and Counting, International Studies Review, Vol. 8, No. 1 (March 2006), pp. 165178.

Contending with Terrorism 332

rorists and that law enforcement agencies can work together across borders. U.S. efforts to combat terrorism in these areas involve extensive cooperation with European states, but this does not occur through NATO. Rather, it occurs either bilaterally or with the EU. Additionally, the threats facing the United States and its European allies are different. The United States has chosen to combat terrorism as far from its shores as possible, through military actions in Afghanistan and Iraq; it also emphasizes the role of covert special-operations forces to conduct counterterrorist missions overseas.102 In contrast, several European states face a domestic threat. Their large Muslim minorities create the potential for homegrown terrorists, particularly to the degree that these minorities have not been integrated into the broader society. Indeed, the July 2005 bombings in London were carried out primarily by British citizens. This gives European states a different set of priorities in ghting terrorism, because the threat they confront is local, not distant.103

Conclusion
NATO plays a largely supportive role in the war on terror. To the degree that NATO countries are engaged in key elements of U.S. efforts to combat terrorism, they do so on the basis of bilateral ties or loose coalitionsnot through NATO. Operation Active Endeavor provides important support for U.S. military operations in the Middle East. It contributes to the prevention of and defense against terrorism; it is not, however, a combat operation. The contribution that NATO members make by providing intelligence in the struggle against terrorists occurs largely bilaterally, and it is generated primarily by law enforcement agencies, rather than by allied military intelligence capabilities. NATOs defense investment programs may help to create better defenses against terrorism, but it is too soon to tell how successful these will be. Similarly, NATO does not have a direct role in denying terrorists access to WMD. NATO maintains political dialogues with countries at risk for the theft or sale of weapons or WMD-related products, and individual members participate in threat reduction activities. These are not designed to address the problem of terrorist acquisition, however. The alliance is split on the use of

102. Rumsfeld, Quadrennial Defense Review, pp. 2324; and Jonathan Stevenson, Demilitarizing the War on Terror, Survival, Vol. 48, No. 2 (Summer 2006), pp. 3754. 103. Jeremy Shapiro and Daniel L. Byman, Bridging the Transatlantic Counterterrorism Gap, Washington Quarterly, Vol. 29, No. 4 (Autumn 2006), pp. 3350.

NATOs International Security Role 333

preemption as a means to prevent the spread of WMD to states that might let terrorists obtain weapons of mass destruction, and the United States prefers the PSI and the new Global Initiative to Combat Nuclear Terrorism to NATO as means to prevent the spread of WMD to terrorists. NATOs ISAF mission in Afghanistan directly contributes to the U.S. goal of denying terrorists sanctuary there. ISAF troops are in essence conducting counterterrorism as well as counterinsurgency operations, and this is NATOs rst combat mission since its creation. The lead role in counterterrorism in Afghanistan continues to be played by U.S. special forces, however, not NATO; and U.S. troops are the largest contingent in ISAF. NATOs role in Iraq is even more limited. Many member states have individually contributed troops to the U.S.-led coalition in Iraq, but the alliances sole contribution to stabilizing the country has been the training of Iraqi military ofcers. NATO has begun to develop consequence management capabilities to respond to terrorist attacks, particularly those with WMD. This could give the alliance a valuable support role. The nature of the terrorist threat confronting individual member-states will likely determine whether they take advantage of NATOs support capability. Three factors explain Washingtons circumvention of the alliance in prosecuting the war on terror. First, two critical changes in the international system, U.S. hegemony and the emergence of a security community, particularly among European states, have led NATOs members both to differ among themselves on a broad range of global issues and to perceive security threats differently. They also differ on the appropriate means for responding to perceived threats, as was most evident in the dispute over the U.S. invasion of Iraq. These shifting alignments and attitudes have reduced U.S. willingness to accept alliance constraints. Second, U.S. military capabilities are greater and more sophisticated than those of its allies, which makes it difcult for even close U.S. allies to coordinate with U.S. forces in frontline military activities. Some U.S. ofcers point out that one goal of NATO training exercises is to illuminate these differences, as a way to spur allies to improve their capabilities.104 But NATOs expansion has eroded its military capabilities further. Combined with the increasing use of national caveats, which constrain what individual military forces can do in NATO operations, the alliances ability to work with the United States in confronting immediate military threats appears limited, at best.
104. U.S. ofcial, interview by author.

Contending with Terrorism 334

Third, the nature of the war on terror itself constrains NATOs contribution to U.S. strategy. Iraq and Afghanistan notwithstanding, terrorism is fought primarily by nonmilitary means, such as law enforcement and intelligence gathering. Moreover, NATOs members face different threats. The United States is unlikely to abandon NATO, however. In spite of its rejection of alliance constraints on its own actions, NATO provides a crucial forum in which the United States can discuss foreign and security policy with its key allies to reach common understandings of shared problems. This is particularly vital to the United States as the EUs Common Foreign and Security Policy begins to coalesce and inuence the policies of European states. Only in NATO does the United States have a voice in European security affairs. This helps explain U.S. support for expanding the alliance, and it has sought to make NATO the forum for discussion of a broad range of security problems affecting Europe and North America. Moreover, Washington recognizes that combating international terrorism requires extensive cooperation, both bilateral and multilateral. This is best built on a shared understanding of the problem states confront, and NATO can play an important role in generating common views regarding terrorism. So long as the United States views NATO as a valuable forum in which it can convince its European allies that they share the same goals and that they confront the same threat in the war on terror, it will continue to value the alliance. If threat perceptions within the alliance diverge further, however, this could make it harder to reach agreement on common policies. Notably, European states appear to have differing views of the threats they face; this is not simply a transatlantic divide. U.S. policy increasingly acknowledges the importance of nonmilitary measures such as public diplomacy and the war of ideas in combating terrorism.105 Increasing recognition that the terrorist threat is evolving means that U.S. approaches are likely to move toward greater concurrence with European policies on terrorism, which stress intelligence, law enforcement, and quiet engagement with the Muslim world. This would ease some of the frictions in Washingtons relations with its European allies. It would not, however, lead to a greater role for NATO in confronting terrorism. Rather, it could accelerate the tendency to utilize mechanisms outside the alliance framework to address this urgent threat.

105. Bush, National Security Strategy (2006), p. 9.

NATOs International Security Role 335

NATOs military value as a partner to the United States in the war against terrorism also remains in question. Should the United States confront terrorists militarily in the future, it will likely do so with special operations forces working either alone or with host-government troops. A few alliance members may participate in such operations, but the alliance itself will not. Further, the bulk of the struggle against terrorism requires substantial nonmilitary means. NATO may have a useful diplomatic role to play, both among its members and with regard to key states such as Russia. But many of the critical tasks in this ght are outside the military domain, leaving NATO with little role. In 2003 NATO Supreme Cmdr. James Jones noted that if the attempt at defense transformation fails, the alliance may lose its military value.106 Others point to NATOs ISAF mission as the essential test for its survival. Its success or failure in Afghanistan will be a critical indicator of the alliances ability to address the type of security threats that will emerge in contested regions around the globe. Success would conrm NATOs unity and capability to act out of area, but a defeat would undermine NATOs claim to a broader global mission. The alliance would continue to provide for the defense of Europe, and the alliance members shared values may be sufcient to sustain NATO as an organization, assuming its political consultation and dialogue functions continue to thrive. But such a defeat would raise serious questions about NATOs contribution to its members core security concerns, if these are seen as out of area. If NATOs major member-states do not seek to address their most urgent threats within the alliance framework, its military value could atrophy.

106. Jim Garamone, Jones Discusses Changing Troop Footprint in Europe, American Forces Press Service, October 10, 2003.

Assessing the Dangers of Illicit Networks

Assessing the Dangers of Illicit Networks

Why al-Qaida May Be Less Threatening Than Many Think

Mette EilstrupSangiovanni and Calvert Jones

he globalization of transportation, communication, and nance has beneted not only licit businesses but also professional criminals and terrorists. Arms dealers, drug trafckers, money launderers, human trafckers, terrorists, and other sundry criminals, enabled by new, affordable technologies, are increasingly organizing into sprawling global networks. As a result, understanding international organized crime and terrorism in terms of networks has become a widely accepted paradigm in the eld of international relations. In this article we seek to clarify that paradigm, probe deeper into the consequences of the network structure, and challenge conventional wisdom about network-based threats to states. A common theme in recent international relations scholarship dealing with organized crime and terrorism is the great difculty states face in combating network-based threats. According to a growing literature, the primary confrontation in world politics is no longer between states but between states and terrorist networks such as al-Qaida, drug smuggling networks such as those in Colombia and Mexico, nuclear smuggling networks in places such as North Korea and Pakistan, and insurgent networks such as those in Iraq.1 And states are widely reputed to be losing the battle. The main reason, according to the existing literature, is the organizational advantages enjoyed by networked acMette Eilstrup-Sangiovanni is Lecturer in International Studies at the Centre of International Studies at the University of Cambridge. She is also Fellow of Sidney Sussex College. Calvert Jones is a doctoral student in the Department of International Relations at Yale University. 1. See, for example, Fiona B. Adamson, Globalisation, Transnational Political Mobilisation, and Networks of Violence, Cambridge Review of International Affairs, Vol. 18, No. 1 (April 2005), pp. 31 49; Jrg Raab and H. Brinton Milward, Dark Networks as Problems, Journal of Public Administration Research and Theory, Vol. 13, No. 4 (October 2003), pp. 413439; Manuel Castells, The Rise of the Network Society (Cambridge, Mass.: Blackwell, 1996); Mark Dufeld, War as a Network Enterprise: The New Security Terrain and Its Implications, Cultural Values, Vol. 6, Nos. 12 (2002), p. 161; John Arquilla and David Ronfeldt, eds., Networks and Netwars: The Future of Terror, Crime, and Militancy (Santa Monica, Calif.: RAND, 2001), especially Phil Williams, Transnational Criminal Networks, pp. 6198; Michael Kenney, From Pablo to Osama: Trafcking and Terrorist Networks, Government Bureaucracies, and Competitive Adaptation (University Park: Penn State University Press, 2007); Audrey Kurth Cronin, Behind the Curve: Globalization and International Terrorism, International Security, Vol. 27, No. 3 (Winter 2002/03), pp. 3058; Sheena Chestnut, Illicit Activity and Proliferation: North Korean Smuggling Networks, International Security, Vol. 32, No. 1 (Summer 2007), pp. 80111; and Marc Sageman, Understanding Terror Networks (Philadelphia: University of Pennsylvania Press, 2004).
International Security, Vol. 33, No. 2 (Fall 2008), pp. 744 2008 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

336

Assessing the Dangers of Illicit Networks 337

tors. A uid structure is said to provide networks with a host of advantages including adaptability, resilience, a capacity for rapid innovation and learning, and wide-scale recruitment. Networked actors are also said to be better at exploiting new modes of collaboration and communication than hierarchically organized state actors. The suggestion that it takes a network to ght a network is therefore gaining currency, both among academics2 and in the wider security community.3 We agree that network-based threats pose serious challenges to state security. But, we argue, the prevailing pessimism about the ability of states to combat illicit networks is premature. The advantages claimed for networks vis--vis hierarchical organizations in the existing literature are often not well characterized or substantiated. Although they tend to enjoy exibility and adaptability, networks have importantand often overlookedstructural disadvantages that limit their effectiveness. Given the high premium on battling networked threats, it is surprising that these disadvantages have not received more attention from international relations scholars. To ll this lacuna, we combine theoretical and empirical evidence to illuminate some important network weaknesses. A caveat is in order. We are neither terrorism experts nor criminologists. Our engagement with the phenomenon of dark networks is motivated primarily by dissatisfaction with the growing international relations literature, where the term network is often used metaphorically and is not clearly dened or expounded.4 Little systematic use has been made by this literature of theoretical approaches to networks developed in other social sciences, and too little effort has been made to build on ndings from extant studies on terrorism, insurgency, and organized crime. As a result, we argue, strategic thinking about how networks can be combated lacks both imagination and historical grounding. The article begins with a brief discussion of the term network in the social sciences. We rst offer a review of the current literature on networks. Next we highlight some limitations that raise doubts about whether networks, in gen-

2. See, for example, Arquilla and Ronfeldt, Networks and Netwars; Richard M. Rothenberg, From Whole Cloth: Making Up the Terrorist Network, Connections, Vol. 24, No. 3 (2002), pp. 3642; and Anne-Marie Slaughter, A New World Order (Princeton, N.J.: Princeton University Press, 2004). 3. For an analysis of the inuence of this idea on U.S. intelligence reform, see Calvert Jones, Intelligence Reform: The Logic of Information Sharing, Intelligence and National Security, Vol. 22, No. 3 (Summer 2007), pp. 384401. 4. The term dark networks was coined by Arquilla and Ronfelt in Networks and Netwars.

Contending with Terrorism 338

eral, and clandestine networks, in particular, are as effective as postulated in their ability to challenge states.5 In general, we argue, international relations scholars have been too quick to draw parallels to the world of the rm where networked organization has proven well adapted to the fast-moving global marketplace. They have consequently overlooked issues of community and trust, as well as problems of distance, coordination, and security, which may pose serious organizational difculties for illicit networks. The second section presents historical evidence on the life cycle and operational effectiveness of networks. Despite a near consensus in the existing literature about the superiority of networks to other forms of social organization, only limited historical and comparative research justies the claim.6 To remedy this weakness, we draw from a wider body of research on the dynamics of participation in underground movements, the life cycle of terrorism and insurgency, and vulnerabilities in organized crime to unearth potential sources of network debilitation in greater theoretical and historical depth. In the third section, we use these ndings as a springboard for analyzing a contemporary, highly potent networked organization: al-Qaida. Although there is much we do not know about this network, the evidence in the public domain suggests al-Qaida is subject to many of the same weaknesses that have beset clandestine networks in the past. We conclude by exploring the theoretical and practical implications of our ndings.

Understanding Networks
The term network has been among the most widely used by social scientists in the last four decades. Economists, organizational theorists, sociologists, and anthropologists have long applied the concept to analyze social and economic systems in which actors are linked through enduring formal and informal rela-

5. We are interested only in the organizational advantages that may or may not ow from the network form. We do not address potential advantages stemming from the psychological proles, belief systems, or other attributes of illicit actors. 6. In international relations, sophisticated single case studies of networks abound, but broad comparative work is rare and often exploratory. See, for example, Arquilla and Ronfelt, Networks and Netwars. Exceptions are Michael Kenney, From Pablo to Osama, who explores the comparative adaptive abilities of drug trafcking and terrorist networks against centralized law enforcement, and Audrey Kurth Cronin, who compares al-Qaida to earlier terrorist networks. See Cronin, How alQaida Ends: The Decline and Demise of Terrorist Groups, International Security, Vol. 31, No. 1 (Summer 2006), pp. 748.

Assessing the Dangers of Illicit Networks 339

tions. More recently, political scientists have adopted the network concept to analyze the organization of nonstate actors at both the domestic and transnational level,7 to study new forms of public administration linking governments and nongovernmental actors,8 or to map the international structure.9 In security studies, scholars increasingly emphasize the role of networks in insurgency, terrorism, and organized crime.10 Much research on networks originates in sociology and organizational theory, and its genesis in these elds is central to understanding why networks, in international relations, are thought to be so effective. Economic sociologists have typically invoked the network concept to analyze the shift away from the classical model of a vertically integrated rm, which relies on top-down management, set bureaucratic routines, and centralized investment to minimize transaction costs.11 This model has been challenged by new forms of horizontal coordination. Many rms now collaborate with competitors, subcontractors, and research institutions through formal and informal networks. Broadly speaking, the literature nds that a networked structure enables exible, on-demand production models that are far better adapted to the shortened product life cycles and accelerating technological changes that typify todays globalized economy than hierarchically organized production.12 As we show below, international relations scholarship has largely imported this logic, without signicant revision, in its conception of transnational networks as effective actors. Yet, this logic may be misleading, at least with regard to illi7. Margaret E. Keck and Kathryn Sikkink, Activists beyond Borders: Advocacy Networks in International Politics (Ithaca, N.Y.: Cornell University Press, 1998). 8. Bernd Marin and Renate Mayntz, eds., Policy Networks: Empirical Evidence and Theoretical Considerations (Boulder, Colo.: Westview, 1991); James N. Rosenau and Ernst-Otto Czempiel, eds., Governance without Government: Order and Change in World Politics (Cambridge: Cambridge University Press, 1992); and Fritz W. Scharpf, Coordination in Hierarchies and Networks, in Scharpf, ed., Games in Hierarchies and Networks: Analytical and Empirical Approaches to the Study of Governance Institutions (Boulder, Colo.: Westview, 1993), pp. 125165. 9. Emilie M. Hafner-Burton and Alexander H. Montgomery, Power Positions: International Organizations, Social Networks, and Conict, Journal of Conict Resolution, Vol. 50, No. 1 (February 2006), pp. 327. 10. See, for example, Raab and Milward, Dark Networks as Problems; Arquilla and Ronfeldt, Networks and Netwars; Kenney, From Pablo to Osama; Cronin, Behind the Curve; Chestnut, Illicit Activity and Proliferation; and Sageman, Understanding Terror Networks. 11. Ronald H. Coase, The Nature of the Firm, Economica, Vol. 4, No. 16 (November 1937), pp. 386405; and Oliver E. Williamson, Markets and Hierarchies: Analysis and Antitrust Implications: A Study in the Economics of Internal Organization (New York: Free Press, 1975). 12. See, for example, Michael J. Piore and Charles F. Sabel, The Second Industrial Divide: Possibilities for Prosperity (New York: Basic Books, 1984); and Laurel Smith-Doerr and Walter W. Powell, Networks and Economic Life, in Neil J. Smelser and Richard Swedberg, eds., Handbook of Economic Sociology, 2d ed. (Princeton, N.J.: Princeton University Press, 2005), p. 384.

Contending with Terrorism 340

cit networks. Clandestine organizationswhether terrorist groups, guerrilla movements, or drug-smuggling enterprisesface a unique set of constraints that distinguish them from their legal commercial counterparts, and their effectiveness cannot be reduced to models of economic efciency. As a result, many advantages claimed for illicit networks in their confrontation with states must be tempered. dening networks Despite growing scholarly attention to the network mode of organization, signicant ambiguity remains about what constitutes a network. A formal denition describes a network as a specic set of relations making up an interconnected chain or system for a dened set of entities that forms a structure.13 This is a loose denition, designating nothing more than a set of linked elements (or nodes). It could refer to a system of computers as well as individuals and could embrace both market and hierarchical structures. In international relations, however, most follow Walter Powell in conceiving of networks as a distinct form of organization, separate from both hierarchies and markets, which link actors working toward common goals.14 From this perspective, a network can be dened as any collection of actors (N 2) that pursue repeated, enduring exchange relations with one another and at the same time lack a legitimate organizational authority to arbitrate and resolve disputes that may arise during the exchange.15 In contrast to markets, exchange relations in networks are enduring; in contrast to hierarchies, networks lack top-down command and authoritative dispute settlement.16 Networks come in many shapes and forms, but all are united by a family of structural properties that, taken together, support assumptions about their
13. Grahame F. Thompson, Between Hierarchies and Markets: The Logic and Limits of Network Forms of Organization (New York: Oxford University Press, 2003), p. 54. 14. Walter W. Powell, Neither Market nor Hierarchy: Network Forms of Organization, Research in Organizational Behavior, Vol. 12 (1990), pp. 295336. 15. Joel M. Podolny and Karen L. Page, Network Forms of Organization, Annual Review of Sociology, Vol. 24, No. 1 (August 1998), pp. 5859. 16. Two broad approaches to network analysis can be distinguished. Social network analysis seeks to reveal how relational ties among individuals affect social outcomes. See Mark Granovetter, The Strength of Weak Ties, American Journal of Sociology, Vol. 78, No. 6 (May 1973), pp. 13601380. Organizational network analysis (ONA) focuses on the organizational level of analysis, examining how networked groups make decisions, pool resources, and engage in collective action. See SmithDoerr and Powell, Networks and Economic Life, p. 369. Viewing networks as a form of governance, as we do, favors an ONA approach because it assumes the existence of common goals, values, or other considerations sustaining collective action. Networks, on this approach, are seen not merely as sets of linked individuals but as self-conscious collective actors in world politics.

Assessing the Dangers of Illicit Networks 341

efcacy. First, whereas traditional hierarchies are based on top-down management, networks are at and decentralized with decisionmaking and action dispersed among multiple actors exhibiting a high degree of local autonomy.17 Although hierarchy in a traditional sense is absent from the network, the boundaries between networks and hierarchies are not always clear-cut. The existence of relatively few nodes with a large number of connections to other nodes (hubs) may introduce an element of hierarchy into the otherwise at network structure. What distinguishes networks from hierarchies is the capacity of lower-level units to have relationships with multiple higher-level centers as well as lateral links with units at the same organizational level.18 Networks are never managed by a single (central) authority. Second, unlike hierarchies, which can rely on authoritative rules and legal arbitration to govern relations, networks are self-enforcing governance structures disciplined primarily by reputation and expectations of reciprocity. As a result, networks tend to require higher levels of trust than other organizational forms.19 Third, unlike the impersonal, rule-guided relations that characterize interactions in hierarchies, networks tend to be based on direct personal contacts. As a result, they are often composed of members with similar professional backgrounds, interests, goals, and values. Relations and connections within networks tend to be informal and loosely structured. Finally, the lack of central authority and rule-guided interaction implies that decisionmaking and coordination in networks tend to be based on consensus and mutual adjustment rather than administrative at. (For a summary of differences between hierarchies and networks, see table 1.) Beyond these core characteristics, networks differ in structure, size, and goals. Some networks are open insofar as they place no restrictions on membership; others are conned to small numbers of like-minded individuals. Some networks are dense, with a large number of connections between individual cells; others are more sparsely linked. Structurally, networks can be di17. Renate Mayntz, Organizational Forms of Terrorism: Hierarchy, Network, or a Type Sui Generis? MPIfG Discussion Paper, No. 04/4 (Cologne, Germany: Max-Planck-Institut fr Gesellschaftsforschung, 2004); and Thompson, Between Hierarchies and Markets, pp. 2224. 18. On hierarchy in networks, see Chris Ansell, The Networked Polity: Regional Development in Western Europe, Governance, Vol. 13, No. 3 (July 2000), pp. 303333, at p. 306; Albert Lszi Barabsi, Linked: The New Science of Networks: How Everything Is Connected to Everything Else and What It Means for Science, Business, and Everyday Life (Cambridge, Mass.: Perseus, 2002); and Ranjay Gulati, Diana A. Dialdin, and Lihua Wang, Organizational Networks, in Joel A.C. Baum, ed., The Blackwell Companion to Organizations (Oxford: Blackwell, 2002), pp. 281303, at p. 289. 19. Powell, Neither Market nor Hierarchy, pp. 301304; Thompson, Between Hierarchies and Markets, p. 43; and Podolny and Page, Network Forms of Organization, pp. 6065.

Contending with Terrorism 342

Table 1. Characteristics of Networks and Hierarchies Networks Structure Membership Unit relations Decision mode Decentralized/horizontal Homogeneous Trust-based, informal (Qualified) majority voting or top-down command Hierarchies Centralized/vertical Diverse, professional Rule-based, formal Consensus

vided into three types: the chain network where people, goods, or information move along a line of separated contacts and where end-to-end communication must travel through intermediate nodes; the hub-and-spoke (or wheel network) where actors are tied to a central (but not hierarchical) node, and must go through that node to communicate with each other; and the all-channel network where everybody is connected to everybody else.20 Terrorist and criminal networks tend to take the form of either chain or wheel networks, whereas all-channel networks are commonly associated with the internet world or some social movements. advantages of networks Much effort has gone into illuminating the benets of networked cooperation. But the potential drawbacks have received far less attention. This section reviews some of the most common organizational advantages claimed for networks vis--vis hierarchical state authorities in the existing literature. efcient communication and information processing. A key advantage claimed for the network is efciency of communication and information processing. In traditional hierarchies, such as state bureaucracies, information typically passes through a centralized processing unit, increasing the risk of congestion and delay. It may be difcult to transmit information on the local characteristics of problems and potential solutions to central decisionmakers. By contrast, the decentralized yet tightly interconnected nature of networks means communication can ow unhindered from one part of the network to another, enabling actors to acquire, process, and act on local information faster than in centralized organizations.21
20. Arquilla and Ronfeld, Networks and Netwars, pp. 78. 21. Scharpf, Coordination in Hierarchies and Networks, p. 135; Powell, Neither Market nor Hierarchy, p. 325; and Wayne E. Baker and Robert R. Faulkner, The Social Organization of Conspir-

Assessing the Dangers of Illicit Networks 343

Information not only ows more freely in networks; it is also thought to be of higher quality. According to Powell, The most useful information is rarely that which ows down the formal chain of command. . . , rather, it is that which is obtained from someone whom you have dealt with in the past and found to be reliable.22 Because cooperation is based on trust and reciprocity rather than on impersonal transactions, networks encourage people to share and collectively interpret information rather than merely pass it on, thereby creating new interpretations and connections.23 As a result, networks are often thought to be more innovative than hierarchies. scalability. A second advantage of networks is scalability, that is, the ability to grow by adding sideways links to new individuals or groups. In principle, a loose organizational structure allows networks to expand freely, integrating new nodes as necessary. If new requirements or problems arise, networks can adapt by adding new links to groups with relevant expertise.24 A networked structure also facilitates recruitment. Due to their dispersed, transnational structure, networks can tap into wider sets of resources such as diaspora populations, and local autonomy allows networks to tailor their message and activities to different communities, thereby increasing their support base.25 Scalability is also enhanced by advances in information and communication technologies. As information ows and transfers of funds become quicker, cheaper, and more secure, the construction of complicated networked organizations over long distances becomes more feasible. A large literature documents how transnational, networked actors use the internet to raise funds, coordinate activities, and recruit new members.26 The upshot, according
acy: Illegal Networks in the Heavy Electrical Equipment Industry, American Sociological Review, Vol. 58, No. 6 (December 1993), pp. 837860, at p. 844. 22. Powell, Neither Market nor Hierarchy, p. 304; and Keck and Sikkink, Activists beyond Borders, pp. 1822. 23. Podolny and Page, Network Forms of Organization, p. 62. 24. Smith-Doerr and Powell, Networks and Economic Life, p. 384; Mayntz, Organizational Forms of Terrorism, p. 12; and Podolny and Page, Network Forms of Organization, p. 66. 25. Bert Klandermans, Hanspeter Kriese, and Sidney Tarrow, eds., From Structure to Action: Comparing Social Movement Research across Cultures (Greenwich, Conn.: JAI Press, 1988); and David A. Snow, E. Burke Rochford Jr., Steven K. Worden, and Robert D. Benford, Frame Alignment Processes, Micromobilization, and Movement Participation, American Sociological Review, Vol. 51, No. 4 (August 1986), pp. 464481. 26. See, for example, Michele Zanini and Sean J.A. Edwards, The Networking of Terror in the Information Age, in Arquilla and Ronfeldt, Networks and Netwars, pp. 2960; Dorothy E. Denning, Activism, Hacktivism, and Cyberterrorism: The Internet as a Tool for Inuencing Foreign Policy, in Arquilla and Ronfeldt, Networks and Netwars, pp. 239288; and Gabriel Weimann, www.terror.net: How Modern Terrorism Uses the Internet, Special Report, No. 116 (Washington, D.C.: United States Institute of Peace Press, March 2004).

Contending with Terrorism 344

to the existing literature, is that networks can expand their scope and boost their ranks with great speed and at low cost. adaptability. A third advantage claimed for networks is adaptability to environmental changes. Compared to hierarchies, network boundaries are more easily redenable and can adjust more rapidly to situational exigencies. As discussed, networks can scale to meet new requirements or needs. Similarly, loose coupling of nodes prevents locking in of ineffective relationships. If a particular organizational link is not providing the expected payoffs, it can be terminated at relatively low cost and replaced with alternative links.27 A relative lack of physical infrastructure also enables networks to relocate operations from one geographic area to another in response to changing constraints. This provides a clear benet for illicit groups that can migrate quickly from areas where the risks from law enforcement are high. For example, Michael Kenney describes how Colombian drug networks respond to police crackdowns by routinely moving their drug plantings and processing labs and creating fresh transportation routes, thereby escaping capture.28 resilience. Research on networks emphasizes their robustness and resistance to inltration and fragmentation. The personal nature of network relationshipsoften based on ties of kinship, loyalty, and trustmeans that networks are more resistant than more impersonal organizational forms to temptations of voice and exit.29 Structural characteristics such as loose coupling and redundant design also reduce systemwide vulnerability. Loose coupling (i.e., minimal interaction and dependency among nodes) means that state authorities cannot use a compromised unit to roll up an entire network, as they might with a vertically integrated adversary.30 Redundancy (i.e., the existence of a large number of structurally equivalent nodes) means that, unlike a hierarchical organization that can be disconnected or debilitated if a top node fails, a large number of nodes can be removed without causing a network to fragment. If one node fails, bypass links can be established around it, allowing business to continue as usual.31 Such resilience is particularly useful to criminal actors that must evade detection and capture. The literature on drug trafcking documents how, to protect themselves from police, trafcking en-

27. Thompson, Between Hierarchies and Markets, p. 144. 28. Kenney, From Pablo to Osama. 29. Thompson, Between Hierarchies and Markets, p. 43. 30. Mayntz, Organizational Forms of Terrorism, p. 14. 31. Duncan J. Watts, Six Degrees: The Science of a Connected Age (London: William Heinemann, 2003), pp. 285286.

Assessing the Dangers of Illicit Networks 345

terprises often compartmentalize their participants into separate groups and limit communication among them.32 Also, many criminal networks allegedly build redundancy into their active groups and leadership to prevent law enforcers from immobilizing the entire network by dismantling a single node.33 learning capacity. The existing literature highlights the advantages of networks over hierarchies when it comes to facilitating learning and innovation.34 By promoting rapid transfers of information, it is said, networks allow participants to learn quickly about new events, opportunities, and threats. Networks also encourage learning through experimentation. In hierarchies, top-down command and heavy initial investments in dedicated machinery and routine tend to lock people into particular ways of working and discourage experimentation.35 By contrast, a at decisionmaking structure allows ideas and methods to be tested more readily, without having to wait for approval from above, thereby allowing wider sets of lessons to be learned.

Critical Questions and Historical Evidence

In current international relations literature, the advantages claimed for networks vis--vis hierarchies are typically assumed to apply to all networked actors, whether they are transnational advocacy groups, human rights coalitions, or criminal syndicates.36 Terrorists and criminalsdue to their constantly changing environments and their dependence on covertnessare believed to prot exceptionally from the network form. The 9/11 commission describes terrorist networks as agile, fast moving, and elusive, difcult for hierarchical states to combat.37 Others depict such networks as nimble, exible, and adap32. Kenney, From Pablo to Osama. 33. Gerben Bruinsma and Wim Bernasco, Criminal Groups and Transnational Illegal Markets: A More Detailed Examination of the Basis of Social Network Theory, Crime, Law, and Social Change, Vol. 41 (2004), pp. 7984; Kenney, From Pablo to Osama; and Russell D. Howard and Reid L. Sawyer, eds., Terrorism and Counterterrorism: Understanding the New Security Environment (Guilford, Conn.: McGraw-Hill, 2004). 34. See, for example, Keck and Sikkink, Activists beyond Borders; Kenney, From Pablo to Osama; Powell, Neither Market nor Hierarchy; and Podolny and Page, Network Forms of Organization, p. 63. 35. See, for example, Bonnie H. Erikson, Secret Societies and Social Structure, Social Forces, Vol. 60, No. 1 (September 1981), pp. 188210; and Kenney, From Pablo to Osama, p. 7. 36. See, for example, Arquilla and Ronfeldt, Networks and Netwars; and Keck and Sikkink, Activists beyond Borders. 37. The 9/11 Commission Report: Final Report of the National Commission on Terrorist Attacks upon the United States (New York: W.W. Norton, 2004), pp. 87, 399.

Contending with Terrorism 346

tive,38 as liable to change . . . structure according to circumstances,39 and as large, uid, mobile, and incredibly resilient.40 These somber assessments appear to be partly vindicated by the recent experiences of leading states in their confrontation with network-based adversaries. Iraqs fractured, unrelenting insurgency, the distressing evolution of media-savvy al-Qaida, and the apparently futile war against Colombian drug lords all suggest a world in which illicit networks are both formidable and adaptive. Most clandestine networks, however, are not as agile and resilient as they are made out to be. In this section we seek to show why many illicit networks may be prone to inefciencies and short life cycles. Our analysis expands the existing literature in two main ways. First, although a great deal of scholarly effort has sought to uncover and explain the advantages of networks, potential weaknesses and constraints have received far less attention. As Miles Kahler notes, successful networks are relatively easy to spot; failed ones much less so. Failures are seldom revealed or evaluated, either with regard to the formation of networks or their ability to achieve their stated goals. As a result, researchers may be overestimating the capacity of networks for collective action, and indeed the overall strength of the network form.41 Also problematic is that researchers rarely evaluate networks explicitly against their organizational alternatives.42 How do networks in international relations compare to other forms of governance? Are they, for example, superior or poorer at coping with change, stress, or failure? A second limitation of the current literature is the insistence on treating contemporary networked threats as newwitness the emphasis on so-called new terrorism. The notion that contemporary criminal actors are fundamen-

38. Bruce Hoffman, Al Qaeda, Trends in Terrorism, and Future Potentialities: An Assessment (Santa Monica, Calif.: RAND, 2003), p. 12. 39. Rohan Gunaratna, Inside Al Qaeda: Global Network of Terror (New York: Columbia University Press, 2002), p. 79. 40. Peter Clarke, cited in Economist, Waiting for al-Qaedas Next Bomb, May 3, 2007. 41. Miles Kahler, Collective Action and Clandestine Networks: The Case of Al-Qaeda, in Kahler, ed., Networked Politics: Agency, Power, and Governance (Ithaca, N.Y.: Cornell University Press, forthcoming). 42. See Smith-Doerr and Powell, Networks and Economic Life. Sociologists in the social movement tradition have explored the comparative utility of a centralized bureaucratic model versus a decentralized, informal approach, but international relations scholars do not typically draw from this work. See, for example, John D. McCarthy and Mayer N. Zald, Resource Mobilization and Social Movements: A Partial Theory, American Journal of Sociology, Vol. 82, No. 6 (May 1977), pp. 12121241.

Assessing the Dangers of Illicit Networks 347

tally different from those in the past makes comparative data difcult to nd because it suggests history is obsolete.43 Yet extant studies of terrorism, insurgency, and organized crime are far from obsolete when it comes to understanding todays networked threats. Opposition to the state has often taken a networked approach, as illustrated by the decentralized Greek resistance to the Ottomans in the early nineteenth century, the Muslim Brotherhoods loosely organized, dispersed resistance to the Egyptian state, and the sprawling international anarchist movement in the late nineteenth century. In the Middle East, informal, loosely structured networks of religion and political activism, sustained as much by personal ties as abstract ideas and common purpose, have a long history.44 Existing research on these and similar entities provides a wealth of data about the strengths and weaknesses of networks. Below we draw from these data to illuminate potential weaknesses of the network form. The analysis is guided by (and limited to) the question of what the distinct structural characteristics of networksthat is, limited central control, local autonomy, and informal, exible interaction based on direct, personal relationsimply for the effectiveness and life cycle of an illicit actor. We do not address potential advantages or disadvantages stemming from other factors, such as the psychological prole or belief systems of criminal actors. Each subsection highlights an area of potential weakness of a networked structure and gives both theoretical and empirical reasons to support the claim, using highly prominent examples of networked cooperation in the history of terrorism, insurgency, and organized crime.45 As such, they should be the most likely to conrm the advantages of network theory. Although we focus spe-

43. According to Ian O. Lesser, The new tendency to organize in networks renders much previous analysis of terrorism based on established groups obsolete. See Lesser, Introduction, in Lesser, Bruce Hoffman, John Arquilla, David Ronfeldt, and Michele Zanini, Countering the New Terrorism (Santa Monica, Calif.: RAND, 1999), pp. 16, at p. 2. 44. See Guilain P. Denoeux, Urban Unrest in the Middle East: A Comparative Study of Informal Networks in Egypt, Iran, and Lebanon (Albany: State University of New York Press, 1993). 45. The cases from which we draw are not chosen randomly to be broadly representative of the universe of illicit networks. Rather we have focused on prominent networks, which, due to the relative success of their activities, have captured signicant media and scholarly attention. These cases should be most likely to conrm the theoretical advantages of a networked structure and thus represent hard cases from our viewpoint. We realize there are key differences in the operational procedures and constraints facing drug smugglers, terrorists, and revolutionaries. Still, we nd, along with others (e.g., Matthew Brzezinski, Re-engineering the Drug Business, New York Times Magazine, June 23, 2002, p. 48; and Kenney, From Pablo to Osama), that there are important structural commonalities in the way these types of groups organize, which justify a comparative analysis.

Contending with Terrorism 348

cically on illicit actors, we suggest these weaknesses may apply to networks more generally. information limitations and communication failure The existing literature portrays networks as highly efcient information providers. Yet, networks may not always be superior organizational structures for gathering, sharing, and processing information. First, decentralization implies that searching for information in networks may be difcult and cumbersome. In principle, a central directory renders the problem of nding information trivial, even in a large, dispersed network. But central directories are expensive to establish and maintain, and may be impracticable in illicit networks due to security concerns.46 Decentralized or distributed systems, on the other hand, tend to be less efcient information providers.47 Because such systems lack a central directory to catalogue information, searches effectively involve each node querying neighboring nodes, which query other neighboring nodes until the information is found (or the search is abandoned). As a result, each unit of information tends to be associated with higher transaction costs than in a centralized system. Compartmentalization of nodes, necessary for security reasons, may also present a barrier to effective information sharing in illicit networks. As discussed, criminal and terrorist networks often seek to minimize potentially destabilizing contacts between cells. To shield them from complicity, members are kept minimally informed about the activities of others, thereby making communication difcult. Dark networks may also nd it hard to source reliable information from outsiders. According to the literature, a key advantage of the network is its structural access to wider, more diverse sources of information. Yet, sociological research on underground participation shows that illicit actors are often isolated from wider social communities.48 Loyalty tends to
46. Secrecy makes information sharing difcult for any type of organization, including hierarchical ones such as, for example, the Central Intelligence Agency, but a decentralized networked structure compounds these difculties due to the absence of reliable means for authenticating information and for controlling who gains access to sensitive data. 47. Duncan Watts gives the example of a purely distributed peer-to-peer internet network called Gnutella. Because it lacks, by design, a central directory, searches in a purely decentralized system such as Gnutella are often far less efcient than a Napster-like network, where queries go to a central high-capacity server. See Watts, Six Degrees, pp. 157158. 48. Martha Crenshaw, Decisions to Use Terrorism: Psychological Constraints on Instrumental Reasoning, in Donatella Della Porta, ed., Social Movements and Violence: Participation in Underground Organizations (Greenwich, Conn.: JAI Press, 1992), pp. 2942; Donatella Della Porta, On In-

Assessing the Dangers of Illicit Networks 349

run deeper when members are cut off from countervailing inuences. Hence, terrorist and insurgent leaders often limit contact with the outside world, thereby restricting the in-ow of information to the network. History provides numerous examples of networks that fail in their missions due to inefcient communication and information sharing or because isolation and secrecy undermine their ability to identify and react to critical changes in their environment. Problems of information sharing are implicated in the demise of the Quebec Liberation Front (known by the French acronym FLQ).49 Although it portrayed itself, and was portrayed in the media, as a centralized, monolithic organization, the FLQ was a loosely organized network of militants, clustered around various charismatic personalities. No central leadership controlled FLQ cellsthese were groups of friends and family sharing the ideal of an independent Quebec. Ronald Crelinstens account of the 1970 October crisis emphasizes obstacles to effective communication and information sharing as a result of this decentralization. The crisis erupted when two friendly but physically separated cells failed to communicate their strategies to each other. The Liberation cell decided to kidnap a British diplomat. The Chenier cell, apparently not well informed about the strategy, carried out a separate kidnapping of Pierre Laporte, the Quebec labor minister. When the Liberation celltaken by surprise by the Laporte kidnappingannounced publicly the FLQ would release both hostages if two of its demands were met, the Chenier cell insisted separately that all demands must be met or Laporte would be killed. Failures of communication and coordination deepened as the network was driven farther underground, until in late 1970 Laporte was killed and popular support for the network lost. What is striking about this example is that the FLQ cells were not initially feuding. Rather, as Crelinsten suggests, the loose network structure and lack of central authority made reliable communication and information sharing difcult, with the result that cooperation broke down. One might object that similar communication failures are less likely in the electronic age, where the internet, cell phones, and videoconferencing enable real-time communication across distances. But as we explain

dividual Motivations in Underground Political Organizations, in Della Porta, ed., Social Movements and Violence, pp. 328; and Donatella Della Porta, Left-wing Terrorism in Italy, in Martha Crenshaw, ed., Terrorism in Context (University Park: Penn State University Press, 1995), pp. 105159. 49. Ronald D. Crelinsten, The Internal Dynamics of the FLQ during the October Crisis of 1970, in David C. Rapoport, ed., Inside Terrorist Organizations (New York: Columbia University Press, 1988), pp. 5989.

Contending with Terrorism 350

below, modern information technologies do not solve all the problems of communication between geographically dispersed actors lacking central command, and police monitoring of phones and internet sites often makes reliance on such technologies perilous. poor decisionmaking and excessive risk-taking A second problem in networks involves strategic decisionmaking. In the theoretical literature, networks assume an almost organic ability to respond exibly to the environment, weighing the options and adjusting their composition and operating procedures as needs change. Yet, real-world networks are not likely to be so smart. First, decisionmaking is unlikely to be as fast or as coherent as the literature suggests. In a network, as in a hierarchy, complex decisions have to be made regarding resource allocation, tactics, whether and when to use violence, what social and political levers to manipulate, and so on. Because these decisions will not ow from centralized leadership, decisionmaking is likely to be a complicated, protracted process as all members try to have a sayor go their own way.50 Decisions also may not be respected as readily due to the lack of an authoritative stamp. As a result, resources may be used poorly, contradictory tactics selected, and activities carried out that serve parochial short-term interests rather than the larger mission. Second, strategy may be virtually nonexistent, or at least rudimentary, without experienced central leadership. Local autonomy means operations can go forward without evaluation, coordination, and a sober assessment of the overall benets and risks. Likewise, the absence of central direction implies that important tasks may be left unassigned or efforts duplicated, causing operational costs to spiral. Strategic processes may be further undermined by selfcensorship. Although the literature highlights deliberation and free exchange of ideas as key advantages of networks, in reality, the norms of collective decisionmaking within cells and strong group loyaltycombined with fear of negative sanctions, even purging, if one does not go along with the majority often lead members to keep off the agenda issues that are threatening to consensus building.51 Free exchange of ideas is therefore not likely to be the norm in illicit networks, and deliberation and rational decisionmaking may suffer as a result.

50. Powell, Neither Market nor Hierarchy, p. 318. 51. Gordon H. McCormick, Terrorist Decision Making, Annual Review of Political Science, Vol. 6 (June 2003), p. 498.

Assessing the Dangers of Illicit Networks 351

Illicit networks are also prone to excessive risk taking. Sociologists nd that decision processes within highly insulated underground organizations are prone to groupthink, and various forms of delusion, including a sense of invulnerability, which encourage extreme risk taking.52 Risk taking may also be induced by the pressure to reciprocate. Networks, like other organizations, thrive on results: terrorists must engage in violent attacks to maintain visibility; drug smugglers need to shift their goods lest they lose supplies and clients. Pressure to reciprocate, combined with the high reputation costs associated with failure, can induce recklessness, forcing members to choose a course of action before they are ready. Research on a Montreal criminal network reveals how a networked structure may induce risk taking.53 This network was subjected to intense police surveillance and disruption over a two-year period. Shipments were routinely seized, but the network itself was allowed to survive so law enforcement could gather enough intelligence to bring it down fully. The effects of stress on the network are revealing. When seizures took place, trafckers tended to blame each other and worry about their own reputations instead of trying to learn how to avoid future seizures.54 According to conversation logs, those involved with failed shipments grew increasingly reckless in their attempts to compensate for losses and get back into the networks good graces. Rather than seek to interpret their failures and revise their strategy, participants seem to have been so frazzled by their failures and dread of what might happen to them as a result that they were moved to irrational behavior. Although the threat of punishment might have the same effect in a hierarchical structure, it is likely that more formal patterns of cooperation, which give members a sense that their place in an organization is relatively stable and secure, rather than based precariously on informal reciprocity, might prevent similar recklessness. restrictions on scope and structural adaptability As we have shown, networks are often depicted as highly elastic entities that combine, recombine, and expand to adapt to transformations in their environment. Yet, both common sense and empirical evidence suggest there are limits
52. A wide literature in sociology nds that decisionmaking by consensus induces risk-taking behavior, particularly in underground organizations. See, for example, Michael A. Wallach and Nathan Kogan, The Roles of Information, Discussion, and Consensus in Group Risk Taking, Journal of Experimental Social Psychology, Vol. 1, No. 1 (1965), pp. 119, at p. 1. For an overview of this literature, see McCormick, Terrorist Decision Making. 53. Carlo Morselli and Katia Petit, Law-Enforcement Disruption of a Drug Importation Network, Global Crime, Vol. 8, No. 2 (May 2007), pp. 109130. 54. Ibid.

Contending with Terrorism 352

to the scalability and structural adaptability of most networks. There are several reasons why networks, and illicit networks in particular, may nd it difcult to scale. As discussed, because illicit networks cannot depend on hierarchy or the legal system to resolve disputes, they are crucially dependent on interpersonal trust. The high premium on trust both limits the feasible size of networks and restricts recruitment. It is well known, for example, that it is easier to generate trust and generalize expectations of reciprocity in small collectivities when the social distance between actors is short, and chains of action are not extended.55 This favors small networks. It is also easier to generate trust when actors are homogeneous in outlook, life style, and culture. As a result, recruitment to illicit networks mostly proceeds through preexisting networks of personal relationships, typically ones that rest on kinship or previous bonding experiences.56 For example, contemporary drug trafcking largely occurs within ethnically homogeneous groups, where kinship generates trust and reciprocity among criminals reluctant to transact with people they have not known for long periods.57 Gerben Bruinsma and Wim Bernascos analysis of Turkish heroin trafcking suggests the entire trade chains from production in Turkey to sale in Europe are based on close family relationships.58 In Donatella Della Portas sample of Italian Red Brigades, 70 percent of recruits had at least one friend involved already,59 and in Marc Sagemans sample of mujahideen, at least 75 percent had preexisting bonds of family or friendship.60 These recruitment practices restrict the scope of illicit networks, casting doubt on rapid expansion.

55. Thompson, Between Hierarchies and Markets, p. 45. 56. Raab and Milward, Dark Networks as Problems; Erikson, Secret Societies and Social Structure, p. 195; and Klandermans, Kriese, and Tarrow, From Structure to Action. The constraint set by preexisting social networks on recruitment varies across criminal activities. For lower-risk activities, such as trade in stolen cars or some drug-trafcking activities, signicant migration movements have stimulated the growth of extensive cross-border networks held together by shared ethnicity. See, for example, John McFarlane, Transnational Crime as a Security Issue, in Carolina G. Hernadez and Gina R. Pattugalan, eds., Transnational Crime and Regional Security in the Asia Pacic (Manila: Institute for Strategic and Development Studies, 1999), p. 53. But, in general, the more dangerous and risky the activity, the more networks rely on strong personal ties. Thus, Martha Crenshaw nds that marijuana users are willing to share information or drugs with a wider range of people than the more endangered and hence more cautious users of heroin. See Crenshaw, An Organizational Approach to the Analysis of Political Terrorism, Orbis, Vol. 39, No. 2 (Fall 1985), pp. 465489. On this logic, terrorists are likely to be among the most cautious in whom they trust. 57. Brzezinski, Re-engineering the Drug Business; Kenney, From Pablo to Osama, p. 28; and Raab and Milward, Dark Networks as Problems, p. 8. 58. Bruinsma and Bernasco, Criminal Groups and Transnational Illegal Markets, p. 87. 59. Della Porta, Left-wing Terrorism in Italy, p. 139. 60. Sageman, Understanding Terror Networks, pp. 111113. Justin Magouirk, Scott Atran, and Marc

Assessing the Dangers of Illicit Networks 353

Networks grow not only by recruiting new members but also by linking sideways to other networked groups. Evidence suggests, however, that networks that seek to scale in this way often end up splintering as a result of differences of ideologies, goals, and strategies. The splintering of the network of Egyptian militants who assassinated President Anwar al-Sadat highlights the tendency toward fragmentation when networks attempt to grow by linking to other, like-minded groups.61 The network of Islamic militants that assassinated Sadat in October 1981 was a loose coalition of autonomous groups that had decided the previous year to coordinate their activities. Although they shared the goal of establishing an Islamic state, these various groups held different views on how to achieve that goal, some favoring a violent coup and others focusing on a broad popular uprising. When presented with the opportunity to strike Sadat, local leaders disagreed widely on the desirability of such a strategy. This lack of unity, in turn, led to poor planning and preparation. To avoid dissent and possible leakage, many local leaders were informed of the plot very latein some cases only hours before it unfolded. As a result, they were caught by surprise and ill prepared for the wave of arrests that followed. The insurrection that followed the assassination was crushed in a matter of days, and more than 300 members of the network were arrested and put on trial. Once the network was hit, it quickly fragmented, as it broke into rival groups. The Egyptian militants are far from a unique example. There are numerous instances of networks that fracture when they attempt to scale. The original Palestinian Liberation Front (PLF), which was founded by Ahmad Jibril in 1959, merged with other nationalist militant groups in 1967 to form the Popular Front for the Liberation of Palestine (PFLP). In 1977 the PLF splintered from the PFLP due to internal conict, and in 1983 and 1985 the organization split again into pro-Palestine Liberation Organization (PLO), pro-Syrian, and proLibyan factions, each of which claimed to represent the mother-organization.62 The much-publicized split within the ranks of Jemaah Islamiyah (JI) in 1999 2000 occurred after new links between senior JI operatives and al-Qaida took the organization in a more militant direction, causing rifts among its leaderSageman document the prevalence of kin relationships within Jemaah Islamiyah in Connecting Terrorist Networks, Studies in Conict and Terrorism, Vol. 31, No. 1 (January 2008), pp. 116. 61. See Sageman, Understanding Terror Networks, pp. 3033. For a more detailed account, see David Sagiv, Fundamentalism and Intellectuals in Egypt, 19731993 (New York: Routledge, 1995), pp. 5461; and Steven Brooke, Jihadist Strategic Debates before 9/11, Studies in Conict and Terrorism, Vol. 31, No. 3 (June 2008), pp. 201226. 62. In the Spotlight: The Palestinian Liberation Front (PLF), Center for Defense Information, updated November 14, 2002, http:/ /www.cdi.org/terrorism/plf.cfm.

Contending with Terrorism 354

ship.63 Aum Shinrikyo, the cult known for the deadly sarin gas attack in Tokyo in 1995, experienced rapid growth in the late 1990s counting as many as 40,000 members around the world. However, unity suffered as a result. Around 2000, a leadership contest broke out between the founder, Shoko Asahara, and challenger Fumihiro Joyu, who sought to distance himself from the violent teachings of Asahara, and in 2007 the organization split in two. Today the cults membership stands at about 1,500.64 That some networks disintegrate when seeking to expand does not suggest that all networks nd it difcult to scale. Research suggests, however, that networks that grow too large often nd it difcult to sustain unity of purpose, and that their effectiveness declines as a result. The main reason is the difculty in coordinating behavior and nourishing agreement among large, diverse groups lacking central leadership. The Basque nationalist group ETA and the PLO both provide examples of networks that have struggled to sustain unity as they grew. The demise of the international anarchist movement also highlights the difculty of sustaining community and commitment in dispersed networked structures. By the early twentieth century, scattered in Europe and the United States, the anarchists were exposed to a variety of countervailing inuences, drawing them away from their commitment to transnational anarchism.65 Historically, many nonstate organizations that have expanded abroad have found their base of popular support weakened as a result.66 A networked structure that thrives on local empowerment often struggles to project a unied image of itself to the world. With various centers of power claiming to speak for the network, its legitimacy and very identity can easily be called into question. The decentralized Algerian insurgency in the early 1990s is a striking example.67 The Islamic Salvation Front (known by the French acronym FIS), the Islamist party banned by the Algerian regime after it won parliamentary elections in 1991, had exiled leaders in both the United
63. See Magouirk, Atran, and Sageman, Connecting Terrorist Networks; and Elena Pavlova, From a Counter-Society to a Counter-State Movement: Jemaah Islamiyah According to PUPJI, Studies in Conict and Terrorism, Vol. 30, No. 9 (September 2007), pp. 777800. 64. Holly Fletcher, Aum Shinrikyo (Japan, Cultists, Aum Supreme Truth), Council on Foreign Relations, Backgrounder, updated May 18, 2008, http:/ /www.cfr.org/publication/9238/. 65. Martin A. Miller, The Intellectual Origins of Modern Terrorism in Europe, in Crenshaw, Terrorism in Context, pp. 2762. 66. See David C. Rapoport, The International World as Some Terrorists Have Seen It: A Look at a Century of Memoirs, in Rapoport, Inside Terrorist Organizations, pp. 3258. 67. Martin Stone, The Agony of Algeria (New York: Columbia University Press, 1997); and Michael Willis, The Islamist Challenge in Algeria: A Political History (Washington Square: New York University Press, 1997).

Assessing the Dangers of Illicit Networks 355

States and Germany. Concerned with building international support, these leaders were eager not to espouse violence for fear it would endanger the status of the FIS as a legitimate political party. But the FIS also had leaders in Algeria who refused to reject violence. Conicting communiqus often emerged from various centers of power, and, as a result, it was unclear who represented the FIS and broader Islamist insurgency. No one had rm control over all the armed groups, and the violence of some ended up tainting the broader political movement, undermining popular support.68 A similar schism appears to affect Gamaa al-Islamiyya (Islamic Group, or IG) today. IG, an international terrorist network seeking to create an Islamic state in Egypt and implicated in the 1993 World Trade Center bombing, split into violent and nonviolent factions after announcing a cease-re in 1997. Since then, the network has continued to grow more divided as exiled leaders abroad have advocated the use of mass violence while members of the groups leadership in Egypt reject it.69 Not only scalability but also structural adaptability is likely to be limited in many networks. Indeed, networks ties may be stickier than the image of loose coupling suggests; contrary to conventional network wisdom, networks could be more resistant to change than bureaucratic ties. To be sure, a lack of physical infrastructure and an absence of bureaucracy ensure some exibility. But consider that much of the exibility claimed for networks stems from loosely coupled or weak links, which can be easily redrawn and remodeled. In economics weak ties are seen as a boon because they give rise to more diverse sources of information, fresh perspectives on problems, and new collaboration opportunities. Most clandestine networks, however, are built on strong ties based on kinship and previous bonding experience. Adding or severing links may be difcult, and physical movement and relocation not so easy, when personal relationships are involved. Because networked cooperation builds on expectations of reciprocity, strong reputation constraints on breaking network ties may also limit exibility.70 Consequently, not only the ability of networks to grow but also their ability to adapt through restructuring themselves may be far more modest than the literature recognizes.

68. Willis, The Islamist Challenge in Algeria, p. 315. 69. U.S. Department of State, Ofce of the Coordinator for Counterterrorism, Chapter 6: Terrorist Organizations, in Country Reports on Terrorism, April 30, 2008, http:/ /www.state.gov/s/ct/rls/ crt/2007/103714.htm. 70. Podolny and Page, Networks Forms of Organization, pp. 6162.

Contending with Terrorism 356

collective-action problems due to coordination As suggested, networks face a variety of collective-action problems. A frequent source of collective-action failure is internal conict. In-ghting commonly aficts clandestine groups, whether modeled hierarchically or not,71 but networks, with their lack of centralized control, are especially susceptible to internal strife. Local autonomy, though advantageous in some respects, easily nourishes the growth of competing centers of power with independent bases of legitimacy, loyalty, and material support. Challengers are more willing to assert themselves if central leadership is weak. In networks, moreover, rules and regulations about the use of tactics, allocation of resources, and so on are not formally established. As a result, competing centers of power may ght over such issues more readily, and without the aid of formal arbitration. The PLO vividly illustrates how tensions can go unresolved in informal networks, regardless of close social linkages. This loose federation of nationalist groups historically suffered from acute inghting, as autonomous segments vied for control. Sponsor states such as Egypt, Syria, and the conservative Gulf monarchies deepened internal strife by funding rival groups, but they were only partly to blame for Palestinian fractiousness.72 As David Schiller suggests, structural features typical of networksthe absence of central authority, the unchecked autonomy of rival groups, and the inability to arbitrate quarrels through formal mechanismsmade the PLO excessively vulnerable to outside manipulation and internal strife. ETA, which began as a diffuse, heterogeneous movement with limited central control, provides another example of network fractiousness.73 In the 1960s, aiming for a broad-based insurgency against President Francisco Franco, it was open to people of widely different political and social backgrounds, some supporting Basque independence, others unication with the French Basque territories, still others self-rule within

71. See McCormick, Terrorist Decision Making. 72. David Schiller, A Battlegroup Divided: The Palestinian Fedayeen, in Rapoport, Inside Terrorist Organizations, pp. 90108. 73. For this example, see Cyrus Ernesto Zirakzadeh, From Revolutionary Dreams to Organizational Fragmentation: Disputes over Violence within ETA and Sendero Luminoso, Terrorism and Political Violence, Vol. 14, No. 1 (Winter 2002), pp. 6692. Not only terrorist groups but also other networked actors are prone to disunity. Research on U.S. street gangs suggests that loosely congured gangs, lacking clear roles, a corporate structure, and central control, cannot effectively control the behavior of their members. A study of warring gangs in St. Louis, Missouri, revealed that gang homicides were more likely to take place within gangs than between them, mainly because of their lack of central control and discipline. See Scott H. Decker and David Curry, Gangs, Gang Homicides, and Gang Loyalty: Organized Crimes or Disorganized Criminals, Journal of Criminal Justice, Vol. 30, No. 4 (JulyAugust 2002), pp. 343352.

Assessing the Dangers of Illicit Networks 357

Spain. Differences over the acceptability of violent tactics triggered serious internal friction, and splinter groups multiplied over the generations, with little in common except the franchise label ETA. Although fractiousness is certainly not uncommon in terrorist groups, a networked structure appears to nourish it. The Shining Path in Peru, one of the most cohesive, enduring insurgent organizations in the twentieth century, provides a helpful contrast. Research on the groups high degree of social cohesion and effectiveness suggests it was closely tied to its hierarchical structure, centered on the powerful leadership of Abimael Guzmn.74 Jemaah Islamiyah is another case in point. JI under the leadership of Abdullah Sungkar was a highly centralized organization with a top-down chain of command and clearly dened objectives. Sungkar ruled JI with an iron hand and did not allow any rival centers of power to arise within the organization. After Sungkars death in 1999, JI split into fractious groups as a militant minority under the leadership of Hambali broke with the moderate majority and carried out a series of terrorist attacks in Southeast Asia from 2000 to 2005. The nominal leadership of Sungkars second in command, Abu Bakar Baasyir, has been largely unsuccessful in stemming fractiousness within what is now a much more loosely structured network.75 Distance coordination poses another obstacle to collective action. The network literature tends to assume away problems of distance. The internet and other communications technologies are thought to enable seamless cooperation among geographically dispersed actors. Yet, research on computersupported collaborative work nds that such technologies do not solve all the problems of distance cooperation, and may generate new difculties.76 People

74. Gordon H. McCormick, The Shining Path and Peruvian Terrorism, in Rapoport, ed., Inside Terrorist Organizations, pp. 109128. Zirakzadeh, too, comes to this conclusion in his explicit comparison of the ETA with the hierarchical Shining Path; hierarchy helped prevent fragmentation of the Peruvian group. Zirakzadeh, From Revolutionary Dreams to Organizational Fragmentation. 75. See Magouirk, Atran, and Sageman, Connecting Terrorist Networks; and Pavlova, From a Counter-Society to a Counter-State Movement. 76. On the problems of distance collaboration, especially conict, misunderstanding, and trust, see Pamela J. Hinds and Diane E. Bailey, Out of Sight, Out of Sync: Understanding Conict in Distributed Teams, Organization Science, Vol. 14, No. 6 (NovemberDecember 2003), pp. 615632; Gary M. Olson and Judith S. Olson, Distance Matters, Human-Computer Interaction, Vol. 15, Nos. 23 (September 2000), pp. 139178; Catherine Durnell Cramton, The Mutual Knowledge Problem and Its Consequences for Dispersed Collaboration, Organization Science, Vol. 12, No. 3 (MayJune 2001), pp. 346371; Pamela J. Hinds and Mark Mortensen, Understanding Conict in Geographically Distributed Teams: The Moderating Effects of Shared Identity, Shared Context, and Spontaneous Communication, Organization Science, Vol. 16, No. 3 (MayJune 2005), pp. 290307; Sirkka L. Jarvenpaa and Dorothy E. Leidner, Communication and Trust in Global Virtual Teams, Organization Science, Vol. 10, No. 6 (June 1999), pp. 791815; and Patricia Wallace, The Internet in the

Contending with Terrorism 358

easily misinterpret and misunderstand one another when they must rely on voice transmissions, email, and instant messaging. Without face-to-face interaction, people often fail to identify and correct for misjudgments.77 Consequently, conicts are more likely to erupt in geographically dispersed teams, as opposed to colocated ones. This research also casts doubt on the ability of dispersed networks to establish and preserve social cohesion. Face-to-face interaction is crucial in building a social-support structure for communication. People separated by signicant distances often lack the contextual information to make sense of behavior, and as a result tend to be less cohesive and trusting than their face-to-face counterparts. The absence of a headquarters, though considered an advantage for security, may also undercut social cohesion in networks. A central base where recruits live, train, or plan activities together can be essential for building the trust that sustains collective action. Loyalty and commitment may be less easy to instill in networks that are transnational, dispersed, and reliant on temporary, makeshift bases. Germanys experience with terrorism in the 1970s highlights these difculties. Research comparing left- and right-wing groups suggests the dispersed, networked structure of right-wing groups severely reduced their cohesion.78 The far more unied and successful left-wing terrorists tended to organize hierarchically, with professional management and clear divisions of labor. They were concentrated geographically in universities, where they could establish central leadership, trust, and camaraderie through regular, face-to-face meetings. Under interrogation, they rarely betrayed their comrades. By contrast, the right-wing networks were decentralized, scattered about the country, and involvement was often temporary. They were routinely inltrated and their members arrested; those captured frequently betrayed their associates. security breaches Research on illicit networks emphasizes their resistance to inltration and dismantlement. Yet, despite practices of loose coupling and redundant design, which make networks less vulnerable to leadership interdiction and
Workplace: How New Technology Is Transforming Work (Cambridge: Cambridge University Press, 2004). 77. Cooperation that relies on distance technologies is also hampered by the difculty of sharing communication aids. For example, in face-to-face meetings, a map is easily shared and participants can use gestures to communicate movements or locations. This efcient war room style of coordination, easily arranged for colocated teams, is far more difcult for dispersed team members using communications technologies. 78. See Frieldhelm Neidhardt, Left-wing and Right-wing Terrorism Groups: A Comparison for the German Case, in Della Porta, Social Movements and Violence, pp. 215235.

Assessing the Dangers of Illicit Networks 359

random arrests, networks are not necessarily more secure than hierarchical organizations. In fact, once networks are hit, they may unravel spontaneously as participants begin to blame one another, as suggested by the demise of the Montreal drug-trafcking network or the Egyptian assassins.79 More generally, the absence of central authority, while protecting networks by leaving no obvious locus for attack, can seriously jeopardize security. Dispersed authority makes it difcult to monitor activities and screen new recruits. Rules for safe conduct are not dened and enforced centrally but often evolve locally in an impromptu, precarious way, which may undermine security. Security breaches are easily observed in the experiences of underground networks. According to research on the Italian underground in the 1970, those militant left-wing organizations that experimented with a decentralized approach were highly vulnerable.80 They typically engaged local leaders (squadre), who were relatively autonomous in their activities. Although local autonomy boosted grassroots recruitment, it was also risky. Because of their operational freedom, squadre were able to recruit whole groups of supporters without having to wait for background checks and approval from above. Internal security was fragile, and inltration by the state rampant. Many groups, in fact, evolved in the direction of greater centralization precisely because of the risks associated with a networked strategy. The Red Brigades, for instance, instilled more hierarchy in their organization and stricter controls over the process of recruitment. Communication practices among international drug trafckers also highlight the security liabilities of a decentralized approach. Research suggests that, although unofcial guidelines among drug trafckers discourage the use of cell phones and encourage coded language, coordination among actors on the street is often so demanding that agents communicate in an excessively simple, transparent manner that risks security.81 As a former trafcker explained: International dope smugglers have to make thousands of phone calls. There are many who say they never use the phone because its too insecure. They are either lying or not doing any business. Referring to the use of coded language, he continued: Any attempt at sophisticated coding quickly leads to disastrous misunderstandings. I have never heard or made a dope79. Of course, not all networks unravel easily, as the debate about torture demonstrates. Resilience in the face of opposition, however, often has more to do with the extreme loyalty of a few individuals than with a networked structure per se. 80. Della Porta, Left-wing Terrorism in Italy. 81. Morselli and Petit, Law-Enforcement Disruption of a Drug Importation Network.

Contending with Terrorism 360

smuggling call which isnt obviously just that.82 Such practices may explain why communications among drug trafckers, as well as terrorists, are habitually intercepted by security services. learning disabilities Learning is crucial for illicit actors. To stay aoat in a hostile environment, criminals must learn to identify and circumvent rapidly shifting countermeasures, avoid past mistakes, and recover from missteps. In the existing literature, networks are cast as highly efcient learners.83 As with much of the logic on networked effectiveness, however, the theoretical learning advantages of networks t a legal, economic context better than the world of illicit actors. One reason networks are said to be good learners is that they facilitate rapid information ows, which allow actors to nd out about new opportunities and threats. As we have shown, however, information does not necessarily ow freely in networks, especially illicit ones. Loose coupling, combined with pressure to separate the network from its social base for security reasons, tends to reduce social embeddedness and network connectivity, thereby making information sharing difcult. And, as Michael Kenney shows with respect to drugtrafcking networks, compartmentalization means individual cells often absorb only those lessons they have learned directly rather than benet from the experience of others.84 Another alleged learning advantage of the networkits superior ability to produce on-the-job expertiseis also doubtful. Because networks build on personal relationships, they are said to be better than hierarchies at transmitting tacit knowledgethat is, knowledge that cannot be explicitly codied but is associated with learning-by-doing and hands-on techniques.85 Yet, bureaucrats learn on the job too, and personal, mentored relationships also exist in hierarchies. Much like their networked adversaries, agents in counterterrorism and drug law enforcement learn when they operate in the eld, and much like their networked adversaries, they can leverage social networks of friends,
82. Quoted in ibid., p. 17, of copy available online from the Social Science Research Network, http:/ /ssrn.com/abstract?944829. 83. See, for example, Powell, Neither Market nor Hierarchy; and Thompson, Between Hierarchies and Markets. On the learning abilities of illicit networks, see especially Kenney, From Pablo to Osama. 84. See Kenney, From Pablo to Osama, p. 115. Many network scholars recognize that compartmentalization and secrecy may present impediments to learning, but they still insist that networks are better learners than bureaucracies. See, for example, ibid., p. 7. 85. Thompson, Between Hierarchies and Markets, pp. 121123; and Kenney, From Pablo to Osama.

Assessing the Dangers of Illicit Networks 361

colleagues, and informers to learn about changing conditions on the ground. But agents have the added advantage of better access to formal training. Most agents receive training in a variety of elds, including investigation, surveillance, intelligence analysis, and undercover operations and technology, and many pursue advanced education.86 The same amount of routinized knowledge is difcult to transmit in dispersed clandestine networks. The network literature tends to downplay the importance of formal instruction and skill transfer. Informal contacts facilitated by communications technologies are often considered sufcient to transmit skills among networked actors. For example, the literature is rife with examples of how terrorist manuals and bomb recipes can be downloaded freely from the internet. Yet, internet manuals and recipes (on bomb building and other tactics) are often too imprecise to be of practical use.87 To learn how to use advanced weapons and tactics, already knowledgeable students must be taught by experts. It is no coincidence that some of the most successful militant organizations have relied on central training camps to teach their associates basic skills. Fatah, the PLO, Hezbollah, and the Taliban all built training camps where militants took courses on intelligence gathering, bombing, the organization of cells, and so on.88 Such centralized learning camps, however, have proven a security liability in an age where these networks have been ferociously targeted by states. A third barrier to learning in networks is lack of organizational memory. Organizations learn by distilling lessons and storing them in ways that are accessible to others, despite the turnover of personnel and passage of time.89 To be sure, criminal enterprises can record knowledge, relying on manuals, notebooks, and computers.90 Yet, there are limits to what can be written down or saved to a computer without compromising security. Informal organizational memories, dependent on error-prone human recollection, are unlikely to be as reliable as formal ones. Organizational memory may end up fragmented and
86. Michael Kenney, in From Pablo to Osama, p. 81, details the comprehensive training of narcs (narcotics police) but does not believe that this gives them any crucial advantage in the field. 87. See David E. Smith, The Training of Terrorist Organizations, CSC Report, Global Security .org, 1995, http:/ /globalsecurity.org/military/library/report/1995/SDE.htm; and Javier Jordan, Fernando M. Manas, and Nicola Horsburgh, Strengths and Weaknesses of Grassroot Jihadist Networks: The Madrid Bombings, Studies in Conict and Terrorism, Vol. 31, No. 1 (January 2008), pp. 1739. 88. For an overview of illicit training camps, see Gunaratna, Inside Al Qaeda, pp. 93101. 89. Barbara Levitt and James G. March, Organizational Learning, Annual Review of Sociology, Vol. 14 (1988), pp. 319338. 90. On these practices, see Kenney, From Pablo to Osama, p. 56.

Contending with Terrorism 362

fallacious, because no central authority is responsible for consolidating and vetting it. Thus, networked actors may not be able to translate lessons learned into solid improvements in organizational practice.91 History suggests illicit networks are far from nimble learners. As we have shown, in the Montreal drug network, police seizures did not prompt any sustained attempts to learn and adapt. Trafckers instead struggled ever more recklessly to get things back on the road to sustain their reputation, and the network collapsed. The persistent inability in the late nineteenth century of the anarchists to learn from past mistakes is another example. Research on the German anarchists, in particular, highlights security liabilities and learning difculties owing from a networked structure. The German anarchists were an informal group, largely autonomous within a broader transnational structure.92 No one was responsible for establishing and enforcing security procedures. As a result, by the early 1880s, the Berlin police had thoroughly inltrated their ranks and curtailed their activities. In response, individual anarchists attempted to institute changes that would enhance security. A London-based anarchist, Viktor Dave, proposed creating a small commission of known, trusted anarchists that would be responsible for smuggling Freiheit, their banned journal, to its subscribers on the continent. The smuggling process had previously been informal and unregulated, leading to frequent arrests of couriers. Rival leaders on the continent, however, felt that reliance on a central distribution center in London would heighten vulnerability. As no one was in overall charge, no central decision was made, and the police continued to inltrate the group and arrest anarchists. Although they recognized the pitfalls of their practices, the anarchists could not transcend their networked structure to institute changes that would make them more secure.

Perils of Networking: The Case of al-Qaida

So far we have focused on limitations of the network form using theoretical and mainly historical evidence. But are clandestine actors in the twenty-rst century more sophisticated users of the network form? In this section we con91. Michael Kenney documents this problem with respect to drug-trafficking networks. Kenney, From Pablo to Osama, p. 115. 92. On the anarchists and their structure, see Andrew R. Carlson, Anarchism in Germany, Vol. 1: The Early Movement (Metuchen, N.J.: Scarecrow, 1972), especially pp. 334376; and James Joll, The Anarchists (London: Eyre and Spottiswoode, 1964).

Assessing the Dangers of Illicit Networks 363

sider a contemporary networked threat: al-Qaida.93 Al-Qaida is a particularly good case in which to probe the presence of network weaknesses. Received wisdom about al-Qaida emphasizes classic advantages of a networked structure, including adaptability, resilience, and rapid learning.94 In its evolution so far, al-Qaida has proved both adaptable and robust. From an essentially visible organization, running training camps and occupying territory in Afghanistan, al-Qaida has transformed itself into a global jihad movement increasingly consisting of associate groups and ad hoc cells all over the world.95 Despite losing its base in Afghanistan, al-Qaida has not lost its ability to mount terrorist attacks. The attacks against the Ghriba synagogue in Tunisia in April 2002 provided the rst signs of the movements resiliency. These incidents were followed by attacks in Pakistan (May 2002); Kuwait, Yemen, and Indonesia (October 2002); Kenya (November 2002); Turkey (November 2003); Madrid (March 2004); London (July 2005); Jordan (November 2005); and Algeria (December 2007). Add to this a series of spectacular plots that have been foiled only by a concentrated international effort by police and intelligence services. Yet, like its predecessors, the al-Qaida network reveals familiar weaknesses. In this section we highlight three points that cast doubt on the strength of alQaida as a networked actor. First, al-Qaida carried out its most successful missions when it was relatively hierarchically structured. The al-Qaida that perpetrated the September 11, 2001, attacks was not really organized as a network. Indeed, many of al-Qaidas traditional strengths seem to build on a hierarchical structure, which has been increasingly difcult to sustain as the organization has come under stress. Second, as the organization fans out into a more loosely structured network, it appears to be losing unity, cohesion, and collective-action capacity. To regain capacity for large-scale attacks, al-Qaida may have to recentralize at least some of its core activities; yet this will make it more vulnerable to attack. Third, although the al-Qaida network has perpetrated some spectacular terrorist attacks, many more plots have been foiled.96
93. Much has been written and said about al-Qaida, and we do not pretend to be experts. What we have to say is all based on publicly available information. 94. See, for example, Gunaratna, Inside Al Qaeda; David Benjamin and Steven Simon, The Age of Sacred Terror: Radical Islams War against America (New York: Random House, 2003); Jason Burke, AlQaeda: Casting a Shadow of Terror (London: I.B. Tauris, 2003); and Sageman, Understanding Terror Networks. 95. Cronin, How al-Qaida Ends. 96. A statement in November 2006 by Eliza Manningham-Buller, then director-general of the British Security Service, indicated that authorities were aware of nearly thirty plots, many with links to al-Qaida in Pakistan. See Bruce Hoffman, The Myth of Grass-Roots Terrorism: Why Osama bin Laden Still Matters, Foreign Affairs, Vol. 87, No. 3 (May/June 2008), pp. 133138.

Contending with Terrorism 364

That a majority of plots by the worlds allegedly best-led and best-trained terrorist organization are foiled is perhaps of scant comfort to police chiefs and frightened publics, but it does throw doubt on the network as a superior organizational form. hierarchy as key to success What is most revealing, perhaps, is the evidence that al-Qaidas most successful operations took place when the organization possessed a hierarchical structure. In the 1990s al-Qaida had a signicant degree of hierarchy and formal organization in the top tier, though lower levels remained more loosely structured. Organizationally, the core of al-Qaida (central staff) was a tight hierarchy with Osama bin Laden at the top, supported by a shura majlis (consultative council). This leadership oversaw a tidy organization of committees with welldened positions and responsibilities.97 When dened to include its regional afliates, al-Qaida assumed a more networked form with regional hubs acting as subcontractors, who maintained substantial autonomy.98 But although some operations were carried out with local autonomy and limited hierarchical management, successful ones typically received close supervision from above. Indeed, the top tier closely managed the 1998 bombings of U.S. embassies in East Africa and the September 11 attacks. Moreover, al-Qaida until 2001 was not stateless; it used Afghanistan under the Taliban as a base to centrally plan and coordinate terrorist operations around the world. Despite conventional wisdom about the efcacy of dispersed networks, it appears therefore that many of al-Qaidas traditional strengths may have built on hierarchy and centralized training and coordination. networked vulnerabilities The loss of Afghanistan as a base in 2001 scattered al-Qaida, forcing it to adapt by becoming more decentralized and networked. Although observers disagree on the extent to which al-Qaidas core is still operationally intact, most experts agree that al-Qaida today operates less like a top-down structure and more like a loose umbrella group, offering inspiration and legitimacy to radical Islamists from varying backgrounds but not necessarily providing much strategic or tactical support.99 To many the transformation of al-Qaida from a fairly central97. The second tier consisted of a military committee, a nance and business committee, a religious committee, and a media and publicity committee. See Gunaratna, Inside Al Qaeda, p. 77. 98. Kahler, Collective Action and Clandestine Networks. 99. Some observers, such as Marc Sageman, argue that al-Qaida has ceased to exist as either an or-

Assessing the Dangers of Illicit Networks 365

ized hierarchical organization into a more diffuse transnational network has made it a more formidable enemy, better capable of scaling and of avoiding detection. But decentralization and segmentation have also exposed al-Qaida to the gamut of organizational dilemmas associated with a networked structure. A major problem owing from a looser networked structure is poorer security. Before the destruction of al-Qaidas logistical infrastructure, the consultative council considered and approved all major operations.100 Members or associate groups would typically submit proposals to the council, which would select a small number for further development and assist with seed money, training, and tactical support. After al-Qaida was forced to decentralize into smaller operational units, afliated groups started to act on their own initiative without centralized clearance for attacks and with limited links to the network. The Bali cell, which conducted the 2002 bombings, consisted of around twelve activists that apparently came together of their own initiative, chose their own target, and executed the attack independently.101 The Madrid train bombings in 2004 were carried out by a local cell of Moroccan immigrants, which was inspired but not directed by al-Qaida.102 The 2005 London bombings too were carried out autonomously.103 As one might expect, enhanced autonomy has led to security problems. Like
ganizational or an operational entity; others, such as Bruce Hoffman, insist that the centre holds. They point to recent intelligence analyses by U.S. and British security services that suggest alQaida still exercises top-down planning and command and control capabilities from its new position along the Afghani-Pakistani border. See Hoffman, The Myth of Grass-Roots Terrorism; and National Intelligence Council, National Intelligence Estimate, July 2007. We agree that al-Qaida has not been operationally neutralized. Pursuit by U.S. and coalition forces has, however, forced it to grow much more dispersed and decentralized. 100. See Peter L. Bergen, The Osama bin Laden I Know: An Oral History of Al Qaedas Leader (New York: Free Press, 2006). 101. The Bali bombing was implemented by a fringe within JI with no input from the majority of the organization, including high-level leadership. See Burke, Al-Qaeda, pp. 265266; Magouirk, Atran, and Sageman, Connecting Terrorist Networks; and Pavlova, From a Counter-Society to a Counter-State Movement. 102. See Jordan, Manas, and Horsburgh, Strengths and Weaknesses of Grassroot Jihadist Networks; Loren Vidino, The Hofstad Group: The New Face of Terrorist Networks in Europe, Studies in Conict and Terrorism, Vol. 30, No. 7 (July 2007), pp. 579592; and Aidan Kirby, The London Bombers as Self-Starters: A Case Study in Indigenous Radicalization and the Emergence of Autonomous Cliques, Studies in Conict and Terrorism, Vol. 30, No. 5 (May 2007), pp. 415428. 103. The group responsible for the London bombings appears to have received some support from al-Qaidas network in Pakistan, but the group was not formally linked to al-Qaida and had limited links to the network. See Vidino, The Hofstad Group; and Kirby, The London Bombers as Self-Starters.

Contending with Terrorism 366

the London bombers in 2005, newcomers seek guidance, legitimacy, and tactical support from the network, where they can get it, but are not afraid to act independently. Relying on informal connections and rules of conduct and operating in the absence of institutionalized training or recruitment, many are strikingly nave about security.104 Some of the most prominent cells in Europe and North America, including those in Montreal, London, and Milan, were closely monitored by the police before being uncovered. Recent plots in Britain involved militants under easy surveillance for months and even years.105 The group that orchestrated the Madrid bombings also provides a good example of amateurism in autonomous groups. Only one of the groups members had passed through training camps or had experience in terrorist campaigns.106 The groups lack of professionalism meant it committed grave mistakes, which eventually led to its downfall. For example, the bombs the groups members used were of poor quality, and three of the thirteen bombs did not explode. One of these unexploded bombs provided information that led to arrests just days after the bombings.107 The groups efforts to secure logistical support also put it in danger. Unlike the September 11 suicide pilots who maintained few ties with other individuals in the United States, the Madrid group needed to be in close contact with its social environment to gather logistical support. Lacking secure links to the al-Qaida network, the groups efforts to recruit members and acquire arms through local mosques brought them into contact with a Moroccan police informant, who was posing as an imam, and several police informants on drug matters who almost thwarted the plot.108 This is far from an unusual case. In 2006 seven men in Miami reportedly planning attacks against the Chicago Sears Tower sought to work with someone they thought was an al-Qaida member who turned out to be an FBI informant.109 Lacking formal avenues of access, these militants did not have a secure way of connecting with the greater al-Qaida community. As
104. For evidence of security navet, see Olivier Roy, Globalized Islam: The Search for a New Ummah (London: Hurst, 2004). 105. See, for example, John Ward Anderson and Karen DeYoung, Plot to Bomb U.S.-bound Jets Is Foiled, Washington Post, April 11, 2006; and David Stringer, London Subway Attackers Linked to Al-Qaida, Colluded with Other Terror Cells, Canadian Press Newswire, April 30, 2007. 106. See Jordan, Manas, and Horsburgh, Strengths and Weaknesses of Grassroot Jihadist Networks. 107. Ibid. 108. Ibid. 109. Jerry Seper, Miami Terrorism Suspects Planned War against U.S., Washington Times, June 29, 2006.

Assessing the Dangers of Illicit Networks 367

these examples show, informal links established by dispersed militants are often precarious.110 The more the al-Qaida network attempts to expand through weak links, the less secure it may become. The lack of centralized sanctioning of missions is also proving perilous. To network enthusiasts, al-Qaidas increasingly diffuse structure encourages expansion and innovation through local experimentation. Yet, increased local initiative is a mixed blessing. Take the example of Zuhair Hilal Mohammed al-Tubaiti, who was arrested for planning an operation against U.S. naval vessels in the Strait of Gibraltar. Upon his arrest, he told the Moroccan authorities that al-Qaida had originally rejected him for a martyrdom mission. Reduced hierarchical control following the loss of the Afghan base empowered him to experiment on his own instead of waiting for approval from above.111 Although he acted alone, his failure harmed the wider organization via his betrayal of sensitive information. Al-Qaida has also displayed weaknesses in coordination and strategic planning.112 The 1999 millennial plot against the Los Angeles airport fell apart when the plotters were identied and arrested, one by one, until Ahmed Ressam was left to carry it out on his own. Inexperienced and acting without reliable organizational support, he was easily caught as he tried to enter the United States with a car full of explosives. In 2000 the overloaded boat of explosives targeting the USS The Sullivans in the port of Aden actually began sinking before it could do any damage. The locally planned attack against U.S. naval vessels in the Strait of Gibraltar in 2002 appears to have crumbled as a result of communication failures.113 If the London car bomb plot and Glasgow airport attack discovered in June 2007 are related to al-Qaida, these too reveal poor planning. Even the September 11 attacks show signs of strategic planning failure. Although it demonstrated al-Qaidas ability to hit its far enemy at
110. A case in point is Ishtiaque Parker, a student whom Ramzi Yousef, one of the conspirators in the 1993 World Trade Center bombing, attempted to recruit. Yousef wanted to use Parker to transport explosives but their connection was weak. Parker eventually betrayed information leading to Yousefs arrest in 1995. See Bill Keller, Self-portrait of Informer: An Innocent, New York Times, February 21, 1995; and Sageman, Understanding Terror Networks, p. 109. 111. Francina Bester, New Trends in Contemporary International and Transnational Terrorism, University of Pretoria, 2007. 112. For these examples, see Alan Cowell, Police Find Two Car Bombs in Central London, International Herald Tribune, June 29, 2007; Elaine Sciolino, Casablanca Bombers Were Probably Lost, International Herald Tribune, May 20, 2003; and Sageman, Understanding Terror Networks, pp. 4556, 99103. 113. See Bester, New Trends in Contemporary International and Transnational Terrorism.

Contending with Terrorism 368

home, September 11 turned out to be a disaster for al-Qaida because it led to the destruction of its extensive hierarchical infrastructure in Afghanistan.114 This outcome does not appear to have been anticipated. As Kahler notes, what is striking about the planning for September 11 is the apparent absence of efforts to comprehend or undermine the likely U.S. response: apart from escaping Afghanistan, the leadership does not appear to have made substantial efforts to disperse key assets such as their training camps. This and other strategic failures may be the result of a highly secretive organization that thrives on limited social embeddeness and whose members are therefore cut off from wider sources of information.115 With increased decentralization, such problems are likely to worsen. Learning too has proven feeble in an increasingly diffuse al-Qaida network. In particular, the loss of Afghanistan as a central headquarters for professional training has made learning a more improvised and unreliable affair. The Casablanca bombers, for instance, were trained haphazardly on weekend camping trips, and their homemade explosives were erratic, with only one resulting in mass casualties.116 Improvised learning is also likely to reduce accountability.117 In the camps, al-Qaida had the opportunity to evaluate trainees, choosing only the best for formal participation in operations. With less hierarchical oversight, members are now left to evaluate their own capacities and learn from their own mistakes. Rohan Gunaratna refers to al-Qaida as a learning organization, noting that when bin Laden discovered his satellite phone conversations were being monitored, he used this knowledge to mislead and evade coalition forces targeting him at Tora Bora in 2001.118 Yet alQaida at large has still not learned the lesson. Its informal networked militants continue using easily monitored phones. In 2004 London authorities tapped more than 100 phone lines during their operation against the militants planning the infamous fertilizer bomb attack.119 Perhaps bin Laden is learning, but the broader network appears less advanced. The main benet of increased decentralization and segmentation is that it
114. Bergen, The Osama bin Laden I Know. 115. According to Miles Kahler, al-Qaidas relatively closed network simply did not provide it with useful information about the motivations and likely actions of its principal adversary, the United States. See Kahler, Collective Action and Clandestine Networks. 116. Sageman, Understanding Terror Networks. 117. Bester, New Trends in Contemporary International and Transnational Terrorism. 118. Gunaratna, Inside Al Qaeda, p. 107. 119. Stringer, London Subway Attackers Linked to al-Qaida.

Assessing the Dangers of Illicit Networks 369

has enabled al-Qaida to scale by building informal links to regional hubs.120 But although many see al-Qaidas increased reliance on informal connections to other groups as an indicator of increasing strength, it can equally be interpreted as a weakness.121 Indeed, there are signs that growing inclusiveness is leading to a more disjointed and disunited movement. The al-Qaida network has involved ambitious upstarts such as Abu Musab al-Zarqawi, more traditional organizations such as Egyptian Islamic Jihad, regionally focused groups such as Jemaah Islamiyah, ragtag insurgents in Iraq, converts such as John Walker Lindh and Richard Reid, and home-grown militants living in Europe, often with little connection to the Middle East. As Richard Matthew and George Shambaugh note, it would be a grave mistake to assume that all the members of al-Qaida share an understanding of goals and strategy.122 Problems of disunity could be seen already in the 1990s. Fawaz Gerges, for example, argues that the inner group was riven by ethnic, regional, and ideological rivalries.123 Such problems intensied in the late 1990s. After 1998, when bin Laden called for abandoning the struggle against the near enemy in favor of global jihad against the United States, al-Qaida lost many members who hesitated to take on the United States.124 As the network fans out, new rifts are emerging. Challenging the West means different things to different cells that are often engaged in local power struggles. The increasing autonomy of leaders within the network has also fueled acute internal conicts, limiting expansion and collective action. For example, when Ayman al-Zawahiri brought his Egyptian Islamic Jihad organization into the fold of al-Qaida, several of his top lieutenants opposed the merger and left.125

120. A sample of groups that are allegedly connected to al-Qaida includes the Moro Islamic Liberation Front (Philippines), Jemaah Islamiyah (Southeast Asia), Egyptian Islamic Jihad, al-Ansar Mujahidin (Chechnya), al-Gamaa al-Islamiyya (Egypt), the Abu Sayyaf Group (Philippines), Hezbollah (Lebanon), the Islamic Movement of Uzbekistan, the Salast Group for Call and Combat (Algeria), and Harakat ul-Mujahidin (Pakistan). See Daniel Byman, Al-Qaeda as an Adversary: Do We Understand Our Enemy? World Politics, Vol. 56, No. 1 (October 2003), pp. 139163, at p. 258; and Cronin, How al-Qaida Ends, p. 33. 121. See Cronin, How al-Qaida Ends, pp. 3334. 122. Richard Matthew and George Shambaugh, The Limits of Terrorism: A Network Perspective, International Studies Review, Vol. 7, No. 4 (December 2005), pp. 617627. 123. Fawaz A. Gerges, The Far Enemy: Why Jihad Went Global (Cambridge: Cambridge University Press, 2005), p. 19, quoted in Kahler, Collective Action and Clandestine Networks. 124. Sageman, Understanding Terror Networks, pp. 4547; Burke, Al-Qaeda, p. 150; and Brooke, Jihadist Strategic Debates before 9/11. 125. Roy, Globalized Islam, p. 73.

Contending with Terrorism 370

an adaptive network of a centralized organization in decline? The above discussion suggests a network plagued by internal conict and coordination problems. The diverse, dispersed community that makes up alQaida is likely to grow increasingly fragile as the memories of Afghanistan fade. Al-Qaida clearly has relied on the common experiences of the mujahideen in Afghanistan and the Afghan training camps as a main source of identication and integration.126 After the loss of the camps, militants from around the world can no longer meet face-to-face in a central location, where they might forge strong social ties, unity of purpose, and a clear sense of belonging. Contrary to the expectation of organizational network theory, the loss of a stable central base may reduce al-Qaidas ability to scale by attracting new recruits.127 Forcing al-Qaida to adopt a more networked structure may have done a lot to reduce the terrorist threat. Perhaps the best indication of the networks increased fragility is the changing nature of its operations. Al-Qaida continues to inspire violent actions around the world. Yet, following September 11, al-Qaida has demonstrated little ability to plan and execute complex attacks. Since 2001, terrorist actions linked to jihadist groups have nearly all been aimed at soft (nongovernmental) targets, and all appear to have been initiated by local groups with scant involvement by the al-Qaida leadership.128 Indeed, Gerges notes, al-Qaida may have been reduced to desperate local afliates and cells, and what remains of the core is an ideological label, a state of mind, and a mobilizational outreach program to incite attacks worldwide.129 If al-Qaida as a brand name or ideology succeeds in inspiring widespread violence, then this is clearly a dangerous trend, but it is not one that can be attributed to the strength of the network itself.

Conclusion
The prevailing picture painted by mass media, public ofcials, and academics concerned with international terrorist organizations and other criminal net126. Raab and Milward, Dark Networks as Problems, p. 20. 127. See Kahler, Collective Action and Clandestine Networks. The reconguration of al-Qaida along the Pakistani border may provide a new base but is likely to be a poor match for the Afghan safe haven. 128. Sageman, Understanding Terror Networks, pp. 5255. 129. Gerges, The Far Enemy, p. 40, quoted in Kahler, Collective Action and Clandestine Networks.

Assessing the Dangers of Illicit Networks 371

works is that of a mounting danger, relentlessly on the increase around the world with governments severely hampered in their ability to combat it. But is this view justied? Theoretical arguments for why we should expect networks to be so formidable are unpersuasive, and a brief review of networks in their historical confrontations with states raises doubts about their effectiveness as actors. To be sure, networks have many potential advantages, including exibility, scalability, and resilience. Yet, networks cannot enjoy all these advantages at once. The network structure, though making it easier to survive, makes it far harder to engage in concerted action. Efciency of information and communication, and hence ability to learn, often comes at the expense of covertness and security. Expansion through recruitment based on informal weak ties may enhance a networks potential impact, but it can also reduce trust and security, trigger internal strife, and intensify collective-action problems. As we have shown, a more centralized structure may often be better at dealing with complex tasks. In the end, several of the challenges facing transnational criminal organizations, including distance collaboration, collective action, learning, training, and security, are best tackled by more centralized hierarchical structures. Yet, if diffuse, clandestine, network-based groups seek to increase their organizational capacity through centralization, they will tend to generate new structural vulnerabilities that make them easier to target and neutralize.130 Knowing ones adversary, understanding how he is organized and what advantage this provides, is key to developing sound responses to security challenges. Many of todays transnational threats are said to be so dangerous precisely because they are evolving into diffuse networks. There is veritable nostalgia for the days of the highly centralized maa and hierarchical drug cartels. To keep up with todays criminal networks, it is suggested, law enforcement must itself adopt a more networked structure. But this recommendation rests on uncertain logic and evidence. The very fact that problems such as drug trafcking and terrorism persist, despite law enforcements efforts to combat them, does not imply that networks are superior to the bureaucracies pursuing them. Disadvantages claimed for hierarchies are often based on bad management and are not inherent to form. As we have demonstrated, from an organizational viewpoint, law enforcement agencies enjoy several advantages over clandestine networks, such as centralized information processing, moni-

130. Matthew and Shambaugh, The Limits of Terrorism.

Contending with Terrorism 372

toring of activities, formal training, and reliable organizational memory. All this, combined with a signicant force advantage, implies that states can inict costs on illicit networks with a greater efciency, and for longer periods of time, than illicit networks can with regard to states. Illicit networks are themselves subject to several weaknesses that may be relatively easy for law enforcement to exploit. Generally, networks may be destabilized by (1) reducing the ow of communication and information through the network; (2) hampering decisionmaking and consensus formation; and (3) intensifying collective-action problems and security vulnerabilities.131 As we have shown, a cursory look at the historical evidence suggests these sorts of failures occur spontaneously in many illicit networks when they come under stress. Law enforcement can precipitate them by targeting networks repeatedly, forcing actors to change their practices abruptly, or sowing doubt and mistrust through inltration and manipulation of information. Counterterrorism ofcials may be able to take advantage of organizational splits within terrorist groups by appealing to more moderate members. Indeed, such splits can be encouraged. According to reports from prominent dissidents, the internal power struggle that ripped the Abu Nidal Organization (ANO) in 1989 and led to the execution of more than 150 of its members and 22 of its leaders was instigated by PLO agents who persuaded some members that the random violence perpetrated by ANO harmed the Palestinian cause.132 Terrorism and organized crime are complicated, eclectic phenomena that have multiple causes and require multifaceted responses. In this article we have sought only to draw attention to, and to question, structural features that are claimed to advantage networked criminals vis--vis law enforcement. Again, we are not arguing that illicit networks pose no signicant threat to state security. But if they are formidable enemies, it is probably not due to their networked structure so much as other factors, such as the personal attributes of their members or their sheer depth in numbers. The argument that a hierarchical structure does not per se disadvantage state actors should not lead to complacency about the way illicit networks are combated, and there is much room for improvement in the way counterterrorism and drug law enforcement are practiced. As states have widely realized, we need better international re-

131. See Kathleen M. Carley, Ju-Sung Lee, and David Krackhardt, Destabilizing Networks, Connections, Vol. 24, No. 3 (Winter 2001), pp. 7992. 132. Youssef M. Ibrahim, Arabs Say Deadly Power Struggle Has Split Abu Nidal Terror Group, New York Times, November 12, 1989.

Assessing the Dangers of Illicit Networks 373

sponses to transnational threats, along with smoother intelligence sharing and interagency cooperation. Red tape can also be protably reduced. Yet, bureaucracy may still be better than its alternatives. Despite the current mood of skepticism about traditional, inexible bureaucracies, centrally instituted changes do reduce inefciencies. In a network, inefcient practices and red tape may be bypassed more easily by autonomous local players, but the resulting ad hoc solutions may prove counterproductive. Although law enforcement agencies may still have far to go, there is nothing inherent in the network form that makes it impossible for hierarchical state structures to combat networked adversaries.

Part IV: The Future of Terrorism

How al-Qaida Ends

How al-Qaida Ends

The Decline and Demise of Terrorist Groups

Audrey Kurth Cronin

he war on terrorism might be perpetual, but the war on al-Qaida will end. Although the al-Qaida network is in many ways distinct from its terrorist predecessors, especially in its protean ability to transform itself from a physical to a virtual organization, it is not completely without precedent. And the challenges of devising an effective response over the long term to a well-established international group are by no means unique. Al-Qaida shares elements of continuity and discontinuity with other terrorist groups, and lessons to be learned from the successes and failures of past and present counterterrorist responses may be applicable to this case. Current research focuses on al-Qaida and its associates, with few serious attempts to analyze them within a broader historical and political context. Yet this context sheds light on crucial assumptions and unanswered questions in the campaign against al-Qaida. What do scholars know about how terrorist movements end? What has worked in previous campaigns? Which of those lessons are relevant to understanding how, and under what circumstances, al-Qaida will end? Radical Islamists will pose a threat to the United States and its interests for a long time to come. But there is a difference between sporadic and local acts of terrorism by religious extremists and the coordinated growth of al-Qaida, with its signature of meticulous planning, mass casualties, and global reach. A central assumption of early U.S. planning was that the elimination of al-Qaida would bring the war on terrorism (or the global struggle against violent extremism) to an end. Yet al-Qaida itself is a moving target, with experts arguing that it has changed structure and form numerous times. As a result, the strategy to counter this group is composed of tactics such as targeting its leader, Osama bin Laden, and his top lieutenants and denying the organization the ability, nances, and territory to regroup. Similar approaches have been employed against other terrorist organizations, with sharply varied outcomes.
Audrey Kurth Cronin is Director of Studies of the Changing Character of War Program at Oxford University. She is also on the universitys international relations faculty. The author is indebted to Robert Art, David Auerswald, John Collins, Patrick Cronin, Timothy Hoyt, Joel Peters, Harvey Rishikof, Adam Roberts, Karen Wilhelm, and former colleagues and students at the U.S. National War College. The detailed comments of an anonymous reviewer were invaluable, and the Leverhulme Trust provided much-appreciated nancial support.
International Security, Vol. 31, No. 1 (Summer 2006), pp. 748 2006 by the President and Fellows of Harvard College and the Massachusetts Institute of Technology.

377

Contending with Terrorism 378

Careful analysis of comparable situations can shed light on what is required to close out an epoch dominated by al-Qaida terrorism. Terrorism studies are often event driven, spurred by attacks and the need to analyze and respond more effectively to a specic threat.1 As a result, the bulk of traditional research on terrorism has been descriptive analysis focused on one group, detailing its organization, structure, tactics, leadership, and so on. True to this pattern, since the terrorist attacks of September 11, 2001, there has been an outpouring of research (bad and good) on al-Qaida, but little attention to analyzing it across functional lines within a wider body of knowledge and research on terrorist groups. To the extent that broader crosscutting research has been done, the weight of it rests on questions of the causes of this threat, as well as the arguably narrow matters of the weapons and methods being used or likely to be used. This agenda reects the strengths of the established international security and defense community, where there is far more expertise, for example, on nuclear weapons and proliferation than on the Arabicspeaking networks that might use them, on operational methods such as suicide attacks than on the operatives who employ them, and on the causes of wars than on how they end. Yet just as war termination may be more vital in its implications for the international system than how wars begin, the question of how the al-Qaida movement ends may be vital to understanding the strategic implications for the United States, its allies, and the shape of the new era. The question of how terrorist groups decline is insufciently studied, and the available research is virtually untapped. Yet it has a raft of implications for the challenges posed by al-Qaida and its associates, as well as for the counterterrorist policies of the United States and its allies, many of which reect little awareness or scrutiny of the assumptions upon which they rest. For example, national leaders focus on the capture or death of bin Laden as a central objective in the campaign against al-Qaida.2 Past experience with the decapitation of terrorist groups, however, is not seriously examined for insights into this case. Some analysts concentrate on the root causes of terrorism and urge policies that will shift local public support away from al-Qaida, suggesting a longterm approach toward the movements gradual decline. Experience from cases where populations have become unwilling to support other causes is little
1. On this point, see Frederick Schulze, Breaking the Cycle: Empirical Research and Postgraduate Studies on Terrorism, in Andrew Silke, ed., Research on Terrorism: Trends, Achievements, and Failures (London: Frank Cass, 2004), pp. 161185. 2. See, for example, CIA Chief Has Excellent Idea Where bin Laden Is, June 22, 2005, http:/ / www.cnn.com/2005/US/06/20/goss.bin.laden/index.html.

How al-Qaida Ends 379

tapped, and resulting changes in the behavior of terrorist organizations separated from their constituencies are hardly known. In other cases, the use of force or other repressive measures against terrorist groups has been successful. Yet the conditions under which that approach has succeeded or failed have not been examined for parallels with al-Qaida. Most observers assume that negotiations would never lead to the end of al-Qaida because it has nonnegotiable, apocalyptic demands. But experience with other terrorist groups that had open-ended or evolving demands is little scrutinized. In short, the substantial history of how terrorism declines and ceases has not been analyzed for its potential relevance to al-Qaida. The argument here is that past experience with the decline of terrorist organizations is vital in dealing with the current threat, and that the United States and its allies must tap into that experience to avoid prior mistakes and to effect al-Qaidas demise. The article proceeds in four sections. The rst provides a brief review of previous research on how terrorism declines or ends; the second is an examination of the endings of other relevant terrorist organizations, with an eye toward determining what has worked in previous campaigns and why; the third offers an analysis of al-Qaidas unique characteristics to determine where comparisons with other groups are appropriate and where they are not; and the fourth addresses how what came before has implications for U.S. and allied policy toward al-Qaida today.

Previous Research on How Terrorism Ends

The study of terrorism is often narrowly conceived and full of gaps; it is not surprising, therefore, that the question of how the phenomenon ends is understudied. The vast majority of contemporary research on terrorism has been conducted by scholars who are relatively new to the subject and unaware of the body of work that has gone before: in the 1990s, for example, 83 percent of the articles published in the major journals of terrorism research were produced by individuals writing on the subject for the rst time.3 Thus far they have made little effort to build on past conclusions, with only halting and disappointing progress in understanding the phenomenon outside its present political context.4 Not unrelated, a crippling aspect of much of the research on
3. Andrew Silke, The Road Less Travelled: Recent Trends in Terrorism Research, in Silke, Research on Terrorism, p. 191. 4. Andrew Silke, An Introduction to Terrorism Research, in Silke, Research on Terrorism, pp. 129.

Contending with Terrorism 380

terrorism is its often applied nature; analysts willing to examine more than one group or broader, noncontemporary, conceptual questions are rare.5 This is somewhat understandable, given that different groups undertake terrorist acts for different reasons, and it is safer to specialize; efforts to accelerate the demise of al-Qaida, however, require more lateral thinking. The thinness of terrorism studies may be giving way to more sustained substantive research in the postSeptember 11 world, though it is too early to say whether current attention will persist and mature. Nonetheless, serious research conducted thus far has produced several overlapping themes and approaches in three areas: the relationship between how a terrorist group begins and ends; the search for predictable cycles or phases of terrorist activity; and the comparison of historical counterterrorism cases. links between beginnings and endings Hypotheses about how terrorist groups end are frequently connected to the broader body of hypotheses about what causes terrorism.6 The assumption is that the origins of terrorism persist throughout the life of terrorist organizations and shed light on sources of their eventual demise. But this is often an oversimplication. Given the close ties between terrorism analysis and government support, when the perception of imminent attacks subsides, support for solid research declines. Work on a declining or defunct terrorist group is therefore typically sparser than is the tackling of its origins and evolution. With such a glaring imbalance in the available research, great care must be taken in generalizing about beginnings and endings of specic terrorist groups. Recognition of the interplay of internal and external forces in the evolution of terrorism is also crucial. In any given case, the evolution from political awareness to the formation (usually) of a terrorist group to the carrying out of a terrorist attack is a complex process. Some steps in this process may be acci-

5. Silke, Research on Terrorism, pp. 208209. 6. Martha Crenshaw, How Terrorism Declines, Terrorism and Political Violence, Vol. 3, No. 1 (Spring 1991), p. 73; and Jeffrey Ian Ross and Ted Robert Gurr, Why Terrorism Subsides: A Comparative Study of Canada and the United States, Comparative Politics, Vol. 21, No. 4 (July 1989), pp. 407408. Examples of articles on the drivers of terrorism include Michael Mousseau, Market Civilization and Its Clash with Terror, International Security, Vol. 27, No. 3 (Winter 2002/03), pp. 529; and Audrey Kurth Cronin, Behind the Curve: Globalization and International Terrorism, International Security, Vol. 27, No. 3 (Winter 2002/03), pp. 3058. See also Charles Knight and Melissa Murphy, and Michael Mousseau, Correspondence: The Sources of Terrorism, International Security, Vol. 28, No. 2 (Fall 2003), pp. 192198.

How al-Qaida Ends 381

dental or opportunistic.7 Likewise, the process by which a terrorist group declines may be as much determined by innate factors as by external policies or actors.8 A group may make a bad decision, engage in a counterproductive strategy, or simply implode. It may also have an innate compulsion to actfor example, it may be driven to engage in terrorist attacks to maintain support, to shore up its organizational integrity, or even to foster its continued existence.9 Studies of the causes of terrorism frequently begin with analyses of the role of individual operatives or their leaders. These include examinations of the psychologies of individual terrorists,10 proles of terrorists (and future terrorists) and their organizations,11 assessments of the conditions that encourage or enable individuals to resort to terrorism,12 and studies of the distinctive characteristics of terrorist leaders and their followers.13 The relationship be7. For more on this argument, especially as it relates to the causes of terrorism, see Audrey Kurth Cronin, Sources of Contemporary Terrorism, in Cronin and James M. Ludes, eds., Attacking Terrorism: Elements of a Grand Strategy (Washington, D.C.: Georgetown University Press, 2004), pp. 19 45, especially p. 22. 8. See United States Institute of Peace, How Terrorism Ends, Special Report, No. 48 (Washington, D.C.: United States Institute of Peace, May 25, 1999), especially the overview section written by Martha Crenshaw, pp. 24. 9. For example, Bruce Hoffman compares al-Qaida to the archetypal shark that must continue swimming to survive. See, for example, Ann Scott Tyson, Al-Qaeda Broken, but Dangerous, Christian Science Monitor, June 24, 2002. See also Martha Crenshaw, Decisions to Use Terrorism: Psychological Constraints on Instrumental Reasoning, International Social Movements Research, Vol. 4 (1992), pp. 2942. 10. Among numerous sources on these subjects, see Martha Crenshaw, The Psychology of Terrorism: An Agenda for the Twenty-rst Century, Political Psychology, Vol. 21, No. 2 (June 2000), pp. 405420; Martha Crenshaw, The Logic of Terrorism, in Walter Reich, ed., Origins of Terrorism: Psychologies, Ideologies, Theologies, States of Mind, 2d ed. (Washington, D.C.: Woodrow Wilson Center Press, 1998), pp. 724; and Jerrold Post, Terrorist Psycho-Logic, in Reich, Origins of Terrorism, pp. 2540. 11. This approach is well established in criminology and intelligence analysis. See, for example, Irene Jung Fiala, Anything New? The Racial Proling of Terrorists, Criminal Justice Studies, Vol. 16, No. 1 (March 2003), pp. 5358; Rex A. Hudson, Who Becomes a Terrorist and Why? The 1999 Government Report on Proling Terrorists (Guilford, Conn.: Lyons, 2002); and Paul R. Pillar, Counterterrorism after al-Qaeda, Washington Quarterly, Vol. 27, No. 3 (Summer 2004), p. 105. As Pillar points out, the Transportation Security Administration already uses proling to screen airline passengers. Ibid. 12. See, for example, Alan Krueger and Jitka Maleckova, Education, Poverty, Political Violence, and Terrorism: Is There a Causal Connection? Working Paper, No. 9074 (Cambridge, Mass.: National Bureau of Economic Research, July 2002), p. 2; Alan Richards, Socio-economic Roots of Radicalism? Towards Explaining the Appeal of Islamic Radicals (Carlisle, Pa.: Strategic Studies Institute, Army War College, July 2003); Daniel Pipes, God and Mammon: Does Poverty Cause Militant Islam? National Interest, No. 66 (Winter 2001/2002), pp. 1421; and Gary T. Dempsey, Old Folly in a New Disguise: Nation Building to Combat Terrorism, Policy Analysis Series, No. 429 (Washington, D.C.: Cato Institute, March 21, 2002). 13. See, for example, Leonard Weinberg and William Lee Eubank, Leaders and Followers in Italian Terrorist Groups, Terrorism and Political Violence, Vol. 1, No. 2 (April 1989), pp. 156176.

Contending with Terrorism 382

tween the motivations and characteristics of individual operatives, on the one hand, and the means to end their violent attacks, on the other, is implied but not always obvious. Another approach especially favored among terrorism experts is analyzing the organizational dynamics of the group. Important late-twentieth-century research concluded that terrorism is essentially a group activity: by understanding the dynamics of the group, including its shared ideological commitment and group identity, analysts can isolate the means of ending its terrorist attacks.14 The focus is thus on the dynamics of relationships between members as a way of gaining insight into the vulnerability of the groups hierarchy, the weaknesses of its organizational structure, the groups ideology and worldview, and so on, which in turn potentially sheds light on how a group might unravel. Such research analyzes the behavior of the terrorist group from the perspective of the needs of the organization itself, an approach that was particularly inuential in studying the behavior of leftist and ethnonationalist/ separatist groups of the 1970s and 1980s.15 Many analysts, however, question the relevance of this well-established approach in an era of decentralized, nonhierarchical cell structures that are able to exploit information technology and the tools of globalization.16 The internet is emerging as the critical new dimension of twenty-rst-century global terrorism, with websites and electronic bulletin boards spreading ideological messages, perpetuating terrorist networks, providing links between operatives in cyberspace, and sharing violent images to demonstrate ruthlessness and incite followers to action.17 Likewise, a growing emphasis on individual initiative, the presence of mission-driven organizations operating with an understanding

14. For an organizational analysis of terrorist groups, see David C. Rapoport, ed., Inside Terrorist Organizations, 2d ed. (London: Frank Cass, 2001). 15. Martha Crenshaw, Theories of Terrorism: Instrumental and Organizational Approaches, in Rapoport, Inside Terrorist Organizations, pp. 1927. 16. See, for example, Ian O. Lesser, Bruce Hoffman, John Arguilla, David Ronfeldt, and Michele Zanini, Countering the New Terrorism (Santa Monica, Calif.: RAND, 1999); John Arquilla and David F. Ronfeldt, eds., Networks and Netwars: The Future of Terror, Crime, and Militancy (Santa Monica, Calif.: RAND, 2001); and Ray Takeyh and Nikolas Gvosdev, Do Terrorist Networks Need a Home? Washington Quarterly, Vol. 25, No. 3 (Summer 2002), pp. 97108; and Matthew Levitt, Untangling the Terror Web: Identifying and Counteracting the Phenomenon of Crossover between Terrorist Groups, SAIS Review, Vol. 24, No. 1 (WinterSpring 2004), pp. 3348. 17. Steve Coll and Susan B. Glasser, Terrorists Turn to the Web as Base of Operations, Washington Post, August 7, 2005; Craig Whitlock, Briton Used Internet as His Bully Pulpit, Washington Post, August 8, 2005; and Susan B. Glasser, and Steve Coll, The Web as Weapon: Zarqawi Intertwines Acts on Ground in Iraq with Propaganda Campaign on the Internet, Washington Post, August 9, 2005.

How al-Qaida Ends 383

of the commanders intent, and a lack of traditional logistical trails all have implications for analyzing how terrorist groups end. Cells that operate independently are much more difcult to eliminate and can even gain a kind of immortality. Mission-driven groups are designed to be self-perpetuating and may not t traditional organizational models of how terrorism ends. The nature of the grievance that drives a terrorist organization has some bearing on the speed and likelihood of its decline.18 On average, modern terrorist groups do not exist for long. According to David Rapoport, 90 percent of terrorist organizations have a life span of less than one year; and of those that make it to a year, more than half disappear within a decade.19 Whether an organization supports a left-wing, right-wing, or ethnonationlist/separatist cause appears to matter in determining its life span. Of these three, terrorist groups motivated by ethnonationalist/separatist causes have had the longest average life span; their greater average longevity seems to result, at least in part, from support among the local populace of the same ethnicity for the groups political or territorial objectives.20 It is too soon to compile reliable data on the average life span of contemporary terrorist groups motivated by religion (or at least groups that appeal to religious concepts as a mobilizing force); however, the remarkable staying power of early religious terrorist groups such as the Hindu Thugs, in existence for at least 600 years, would seem to indicate the inherent staying power of sacred or spiritually based motivations.21 Finally, because of the degree to which terrorism research has been subsidized by governments and biased by later policy imperatives, the role of
18. United States Institute of Peace, How Terrorism Ends, p. 1. 19. David C. Rapoport, Terrorism, in Mary Hawkesworth and Maurice Kogan, eds., Routledge Encyclopedia of Government and Politics, Vol. 2 (London: Routledge, 1992), p. 1067. This claim admittedly needs to be updated. A good study of the life span of terrorist organizations, including those that gained purchase in the 1990s, is still waiting to be written. 20. In discussing the longevity of terrorist groups, Martha Crenshaw notes only three signicant groups with ethnonationalist ideologies that ceased to exist within ten years of their formation. One of these, the National Organization of Cypriot Fighters (Ethniki Organosis Kyprion Agoniston, or EOKA), disbanded because its goalthe liberation of Cypruswas achieved. By contrast, a majority of the terrorist groups that she lists as having existed for ten years or longer have recognizable ethnonationalist ideologies, including the Irish Republican Army (in its many forms), Sikh separatist groups, Basque Homeland and Freedom (Euskadi Ta Askatasuna, or ETA), various Palestinian nationalist groups, and the Corsican National Liberation Front. See Crenshaw, How Terrorism Declines, pp. 6987. 21. David C. Rapoport, Fear and Trembling: Terrorism in Three Religious Traditions, American Political Science Review, Vol. 78, No. 3 (September 1984), pp. 658677. Rapoport asserts that before the nineteenth century, religion was the only acceptable cause for terrorism, providing a transcendent purpose that rose above the treacherous and petty political concerns of man.

Contending with Terrorism 384

counterterrorism is often overemphasized. With easier access to government data, researchers tend naturally to stress state behavior. The degree to which terrorist groups evolve independent of government action can be underappreciated. The result is a strong bias toward tying the decline of such groups to specic government policies, especially after the fact, even though the relationship between cause and effect may be unclear.22 cycles, stages, waves, and phases Some researchers argue that terrorist attacks conform to a temporal pattern that provides insight into increases and decreases in numbers of attacks. Thus another approach to understanding the life span of a terrorist movement is to search for identiable cycles. Walter Enders and Todd Sandler assert that long-term analysis of terrorism trends during the late twentieth century indicates that transnational terrorist attacks run in cycles, with peaks approximately every two years. Enders and Sandlers cycles are tracked across terrorist groups worldwide, shedding light on the likelihood of an attack coming from someone somewhere; indeed, before September 11 they correctly predicted enhanced danger of a high-casualty terrorist attack. But like strategic intelligence that provided general but not tactical warning of the September 11 attacks, Enders and Sandlers ndings were of limited use in predicting where the attack would occur, by which group, and by what means.23 The apparent existence of global statistical patterns is interesting, but it provides no insight into the decline of specic terrorist groups. In his attempt to use mathematical analysis to determine risk assessment for al-Qaida attacks, Jonathan Farley likewise concluded that while the connections between cells can be quantitatively modeled, assumptions about how individual cells operate may be wrong.24 The usefulness of statistical models based on a large number of assumptions to determine a specic groups decline is limited.
22. Crenshaw, How Terrorism Declines, p. 73. 23. Walter Enders and Todd Sandler, Transnational Terrorism, 19682000: Thresholds, Persistence, and Forecasts, Southern Economic Journal, Vol. 71, No. 3 (January 2004), pp. 467482; Todd Sandler, Walter Enders, and Harvey E. Lapan, Economic Analysis Can Help Fight International Terrorism, Challenge, January/February 1991, pp. 1017; and Walter Enders and Todd Sandler, Is Transnational Terrorism Becoming More Threatening? A Time-Series Investigation, Journal of Conict Resolution, Vol. 44, No. 3 (June 2000), pp. 307332. 24. Jonathan David Farley, Breaking Al-Qaida Cells: A Mathematical Analysis of Counterterrorism Operations (A Guide for Risk Assessment and Decision Making), Studies in Conict and Terrorism, Vol. 26, No. 6 (November/December 2003), pp. 399411. See also Bernard Harris, Mathematical Methods in Combatting Terrorism, Risk Analysis, Vol. 24, No. 4 (August 2004), pp. 985988.

How al-Qaida Ends 385

Other experts have focused on the existence of developmental stages through which all terrorist groups evolve, especially psychological stages of growing alienation or moral disengagement for groups, individuals, or both.25 Leonard Weinberg and Louise Richardson have explored the applicability of a conict theory frameworkincluding stages of emergence, escalation, and de-escalationto the life cycles of terrorist groups. They conclude that the framework is useful in examining terrorist groups originating or operating in Western Europe in the late twentieth century, but urge more research in this area to determine whether it is applicable to other places and periods.26 Still other analysts suggest that specic types of groups may possess their own developmental stages. Ehud Sprinzak, for example, argued that rightwing groups exhibit a unique cyclical pattern. Driven by grievances specic to their particular group, members direct their hostility against enemy segments of the population dened by who they arewith regard to race, religion, sexual preference, ethnicity, and so onnot by what they do. To the extent that the government then defends the target population, the former also becomes a legitimate target. But the cycle of violence reects underlying factors that may continue to exist, and that can experience periods of are-up and remission, depending on the degree to which the government is able to bring campaigns of violence under control.27 Other researchers study the evolution of terrorist groups as types of social movements and are intellectual descendants of Ted Robert Gurr.28 The more highly developed literature on social movements posits, for example, that terrorism may appear at the end of a cycle of the rise and fall of movements of mass protest.29 Social movements may just as easily be drawn toward more
25. For example, referring to the case study of the Weathermen, Ehud Sprinzak argues that leftwing groups evolve through three stages: crisis of condence, conict of legitimacy, and crisis of legitimacy. See Sprinzak, The Psychopolitical Formation of Extreme Left Terrorism in a Democracy: The Case of the Weatherman, in Reich, Origins of Terrorism, pp. 6585. For another argument about the psychological phases for terrorist groups, see Albert Bandura, Mechanisms of Moral Disengagement, in Reich, Origins of Terrorism, pp. 161191. 26. Leonard Weinberg and Louise Richardson, Conict Theory and the Trajectory of Terrorist Campaigns in Western Europe, in Silke, Research on Terrorism, pp. 138160. 27. Ehud Sprinzak, Right-Wing Terrorism in Comparative Perspective: The Case of Split Delegitimization, in Tore Bjorgo, ed., Terrorism from the Extreme Right (London: Frank Cass, 1995), pp. 1743. 28. Ted Robert Gurr, Why Men Rebel (Princeton, N.J.: Princeton University Press, 1970). There is a rich literature from the 1960s and 1970s on political violence, of which terrorism is arguably a subset. Gurr denes political violence as all collective attacks within a political community against the political regime, its actorincluding competing political groups as well as incumbentsor its policies. Ibid., pp. 34. 29. Crenshaw, How Terrorism Declines, p. 82. See also John O. Voll, Bin Laden and the New Age of Global Terrorism, Middle East Policy, Vol. 8, No. 4 (December 2001), pp. 15; and Quintan

Contending with Terrorism 386

positive means, however. Understanding the pattern of mobilization may be important for dissecting the origins of an established group but may not be as revealing of its likely end. On the whole, research on social movements gives more insight into the origins of terrorist groups than it does into their decline. Finally, Rapoport posits another broad hypothesis on the life cycles of terrorist groups. He argues that over the course of modern history, waves of international terrorist activity last about a generation (approximately forty years). These waves are characterized by expansion and contraction and have an international character, with similar activities in several countries driven by a common ideology. Two factors are critical to Rapoports waves: (1) a transformation in communication or transportation patterns, and (2) a new doctrine or culture. Yet although a wave is composed of terrorist organizations and their activities, the two need not exist concurrently. Rapoport argues that because most individual organizations have short life spans, they often disappear before the overarching wave loses force. The current wave of jihadist terrorism may be different, however, because unlike earlier waves of the modern era, this one is driven by a religious (not a secular) cause. Rapoport is therefore reluctant to predict its end.30 comparative counterterrorism cases Cyclical hypotheses are notoriously difcult to formulate and difcult to prove; they can require so much generalization and qualication that their relevance to specic groups becomes remote. As with many international security questions, an alternative approach has been to assemble volumes of comparative case studies that draw parallel lessons about terrorist organizations, including how they declined and ended or were defeated.31 These, too,

Wiktorowicz, Framing Jihad: Intra-movement Framing Contests and al-Qaidas Struggle for Sacred Authority, International Review of Social History, Vol. 49, Supp. 12 (December 2004), pp. 159 177. 30. David C. Rapoport, The Four Waves of Modern Terrorism, in Cronin and Ludes, Attacking Terrorism, pp. 4673; and David C. Rapoport, The Fourth Wave: September 11 in the History of Terrorism, Current History, December 2001, pp. 419424. 31. The best of these is Martha Crenshaws edited book Terrorism in Context (University Park: Pennsylvania State University Press, 1995), because it examines both the evolution of terrorist organizations and the counterterrorist techniques used against them. It explains terrorism as part of broader processes of political and social change. Reecting its own historical context, however, there is a strong bias toward left-wing and ethnonationalist/separatist groups. In the wake of September 11, Yonah Alexander produced an edited volume of comparative case studies, Combating Terrorism: Strategies of Ten Countries (Ann Arbor: University of Michigan Press, 2002), which emphasizes counterterrorist techniques used mainly against European groups. Another promising study is Robert J. Art and Louise Richardson, Democracy and Terrorism: Lessons from the Past (Washington, D.C.: United States Institute of Peace, forthcoming).

How al-Qaida Ends 387

present a host of challenges. First, terrorism studies often look primarily at the attributes of a particular group or at the counterterrorist policies of a state. Rarely are both equally well considered. Because of the heavy state interest in combating terrorism, the emphasis is understandably on a comparison of counterterrorist techniques used by states over the life span of each group, with policy implications for current challenges. Second, with their focus on a relatively narrow functional question, comparative terrorism cases can fall victim to superciality: regional experts can be reluctant to cede ground to strategic studies experts whom they consider interlopers in their geographic/ linguistic/cultural ambit. For this reason, many comparative studies are published as edited collections of articles by regional experts, but these in turn can fail to control relevant variables and to coalesce on a central theme. Third, access to data is a big problem: conducting primary research on contemporary terrorist groups is difcult because making contact with operatives or their targets can be dangerous for both the researchers and their contacts. In addition, governments may restrict access to relevant written sources.32 Fourth, because of the political nature of terrorism, researchers operate at the intersections of sensitive ideas; maintaining objectivity in studying behavior that is deliberately designed to shock can prove challenging. Finally, studying this phenomenon over a range of terrorist groups in different cultural, historical, and political contexts requires generalization and risks the introduction of distortions when making comparisons.33 The best case studies are usually completed years after a group has ceased to exist; as a result, their applicability to current challenges is limited. For any given group, it is vital to identify characteristics that distinguish it from its predecessors and those that do not.

How Other Terrorist Groups Have Ended

There are at least seven broad explanations for, or critical elements in, the decline and ending of terrorist groups in the modern era: (1) capture or killing of the leader, (2) failure to transition to the next generation, (3) achievement of the groups aims, (4) transition to a legitimate political process, (5) undermin32. Crenshaw, How Terrorism Declines, p. 70. 33. Crenshaw, Thoughts on Relating Terrorism to Historical Contexts, in Crenshaw, Terrorism in Context, pp. 324. One interesting comparative study of terrorist groups in two countries, Canada and the United States, concluded that four factors led to terrorisms decline: preemption, deterrence, burnout, and backlash. Its applicability was limited by the small number of cases, however. Jeffrey Ian Ross and Ted Robert Gurr, Why Terrorism Subsides: A Comparative Study of Canada and the United States, Comparative Politics, Vol. 21, No. 4 (July 1989), pp. 405426.

Contending with Terrorism 388

ing of popular support, (6) repression, and (7) transition from terrorism to other forms of violence. The relevant factors can be both internal and external: terrorist groups implode for reasons that may or may not be related to measures taken against them. Nor are they necessarily separate and distinct. Indeed individual case studies of terrorist groups often reveal that more than one dynamic was responsible for their decline. The typical focus on government counterterrorist measures slights the capabilities and dynamics of the group itself and is frequently misguided; even among groups that decline in response to counterterrorist campaigns, the picture remains complex.34 Counterterrorist techniques are often best used in combination, and methods can overlap: frequently more than one technique has been employed to respond to a given group at different times. The goal here is to focus on the historical experience of previous groups and study the commonalities, in both the internal and external variables, so as to determine aspects of the processes of terrorist decline that are relevant to al-Qaida. Although listing these seven key factors separately is admittedly articial, they are analyzed consecutively for the sake of argument and convenience (see Table 1). capture or killing of the leader The effects of capturing or killing a terrorist leader have varied greatly depending on variables such as the structure of the organization, whether the leader created a cult of personality, and the presence of a viable successor. Regardless of whether the removal of a leader results in the demise of the terrorist group, the event normally provides critical insight into the depth and nature of the groups popular support and usually represents a turning point. Recent examples of groups that were either destroyed or deeply wounded by the capture of a charismatic leader include Perus Shining Path (Sendero Luminoso), the Kurdistan Workers Party (PKK), the Real Irish Republican Army (RIRA), and Japans Aum Shinrikyo. The U.S. government designates all four as foreign terrorist organizations.35

34. Crenshaw argues that the decline of terrorism results from the interplay of three factors: the governments response, the choices of the terrorist group, and the organizations resources. See Crenshaw, How Terrorism Declines, p. 80. In another article, she further explores the internal and external factors, pointing to how government strategies such as deterrence, enhanced defense, and negotiations interact with terrorist group success, organizational breakdown, dwindling support, and new alternatives for terrorist organizations. See United States Institute of Peace, How Terrorism Ends, pp. 25. 35. See the annual U.S. Department of State Country Reports on Terrorism (which, beginning in 2004, replaced Patterns of Global Terrorism) at http:/ /www.state.gov/s/ct/rls/crt. In 2003 the PKK was renamed Kongra-Gel, which continues to be a designated group.

How al-Qaida Ends 389

Table 1. How Terrorist Groups Decline and End Key Factors

a

Notable Historical Examples Shining Path Kurdistan Workers Party Real Irish Republican Army Aum Shinrikyo

Capture/Kill leader(s)

Unsuccessful generational transition

Red Brigades Second of June Movement Weather Underground Baader-Meinhof group (Red Army Faction) The Order Aryan Resistance Army Irgun/Stern Gang African National Congress Provisional Irish Republican Army Palestinian Liberation Organization Liberation Tigers of Tamil Eelamc Moro Islamic Liberation Front Real Irish Republican Army Basque Homeland and Freedom (ETA) Shining Path Peoples Will Shining Path Kurdistan Workers Party Abu Sayyaf Revolutionary Armed Forces of Colombia Khmer Rouge Guatemalan Labor Party/Guatemalan National Revolutionary Unit Communist Party of Nepal-Maoists Kashmiri separatist groups (e.g., Lashkare-Toiba and Hizbul Mujahideen) Armed Islamic Group (Algeria) be found in combination.

Achievement of the cause Transition to a legitimate political process/negotiations

Loss of popular support

Repression

Transition out of terrorism: toward criminality toward full insurgency

a The factors listed here are not mutually exclusive and can b These are illustrative examples, not a comprehensive list. c

This peace process is threatened by renewed violence as this article goes to press.

Shining Paths former leader, Manuel Rubn Abimael Guzmn Reynoso (aka Guzmn), was a highly charismatic philosophy professor who built a powerful Marxist movement through a brutal campaign of executing peasant leaders in Perus rural areas during the 1980s and early 1990s. Somewhat ironically, Shining Path, which was founded in the late 1960s, began to engage in violence just after the government undertook extensive land reform and restored

Contending with Terrorism 390

democracy to the country; the earliest attacks involved the burning of rural ballot boxes in the 1980 presidential election.36 Increased popular access to a university education helped Guzmn radicalize a growing cadre of impressionable young followers. He consolidated his power in part by expelling or executing dissenters, resulting in unquestioned obedience but also a highly individualistic leadership. By the early 1990s, Shining Path had pushed Peru into a state of near anarchy. Guzmns capture on September 12, 1992, however, including images of the former leader behind bars recanting and asking his followers to lay down their arms, dealt the group a crushing blow.37 The Kurdistan Workers Party, an ethnonationalist/separatist group founded in 1974 and dedicated to the establishment of a Kurdish state, also suffered the capture of its charismatic leader, Abdullah Ocalan.38 Beginning in 1984, the group launched a violent campaign against the Turkish government that claimed as many as 35,000 lives. Ocalan was apprehended in early 1999 in Kenya (apparently as a result of a tip from U.S. intelligence) and returned to Turkey, where a court sentenced him to death.39 On the day of sentencing, riots and demonstrations broke out among Kurdish populations throughout Europe. Ocalan, whose sentence was later commuted to life imprisonment, advised his followers to refrain from violence.40 Renamed the Kurdistan Freedom and Democracy Congress (KADEK) and then Kongra-Gel, the group remains on the U.S. terrorist list; however, it has subsequently engaged mainly in political activities on behalf of the Kurds.41
36. Audrey Kurth Cronin, Huda Aden, Adam Frost, and Benjamin Jones, Foreign Terrorist Organizations, CRS Report for Congress (Washington, D.C.: Congressional Research Service, February 6, 2004), pp. 103105, Order Code RL32223. 37. Guzmn was sentenced to life in prison in 1992, but the trial was later ruled unconstitutional; he is scheduled to be retried (for a second time). His successor, Oscar Ramrez Duran, also known as Feliciano, continued to direct the movement after Guzmns capture; but as mentioned earlier, the groups membership sharply plummeted. The State Department has estimated its strength at between 400 and 500 members, down from as many as 10,000 in the late 1980s and early 1990s. See U.S. Department of State, Patterns of Global Terrorism, 2003, April 2004, http:/ /www.state.gov/ documents/organization/31912.pdf. Duran was likewise captured in July 1999. Some analysts worry that the group may resurge, with notable attacks in 2002 and 2003. See Cronin et al., Foreign Terrorist Organizations, p. 103. Although possible, resurgence on a large scale is unlikely: the capture of Guzmn was clearly a watershed. See also Art and Richardson, Democracy and Counterterrorism. 38. For an explanation of the meaning of this term and other main typologies of terrorism, see Cronin, Behind the Curve, pp. 3942. 39. U.S. Welcomes Ocalan Capture, Agence France-Presse, February 16, 1999. Ocalans sentence was later commuted to life in prison. 40. Turkey Lifts Ocalan Death Sentence, BBC News, October 3, 2002, http:/ /news.bbc.co.uk/2/ hi/europe/2296679.stm. 41. Cronin et al., Foreign Terrorist Organizations, pp. 5355.

How al-Qaida Ends 391

The Real Irish Republican Army is a splinter group of the Provisional Irish Republican Army that split off in 1997 after refusing to participate in the peace process. It conducted a series of attacks in 1998, including the notorious Omagh bombing, which killed 29 people (including 9 children) and injured more than 200. The Northern Irish community reacted with such outrage that the group declared a cease-re and claimed that its killing of civilians was inadvertent. In 2000 the RIRA resumed attacks in London and Northern Ireland, focusing exclusively on government and military targets. In March 2001 authorities arrested the groups leader, Michael McKevitt. From an Irish prison, he and forty other imprisoned members declared that further armed resistance was futile and that the RIRA was at an end. The group currently has between 100 and 200 active members and continues to carry out attacks; nevertheless, its activities have signicantly declined since McKevitts arrest.42 Aum Shinrikyo (now known as Aleph) is essentially a religious cult founded in 1987 by Shoko Asahara, a half-blind Japanese mystic. Asahara claimed that the world was approaching the apocalypse and used an eclectic blend of Tibetan Buddhist, Hindu, Taoist, and Christian thought to attract an international following, primarily in Japan but also in Australia, Germany, Russia, Sri Lanka, Taiwan, and the United States. Asahara declared that the United States would soon initiate Armageddon by starting World War III against Japan and called on the groups members to take extraordinary measures in preparation for the attack. The notable aspects of this group are its international reach and its use of so-called weapons of mass destruction, particularly anthrax and sarin gas. In March 1995, members of Aum Shinrikyo released sarin gas in the Tokyo subway, resulting in the deaths of 12 people and injuries to another 5,000. Asahara was arrested in May 1995 and sentenced to death in February 2004. The group has shrunk from approximately 45,000 members worldwide in 1995 to fewer than 1,000, many of whom live in Russia.43 These are just a few of the contemporary cases where the capture or killing of the leader of a terrorist organization proved to be an important element in the organizations decline. Other examples include the arrest of leaders in groups as diverse as Frances Direct Action (Action Directe); El Salvadors
42. Ibid., pp. 8889; and U.S. Department of State, Patterns of Global Terrorism, 2003. 43. Cronin et al., Foreign Terrorist Organizations, pp. 1719; and U.S. Department of State, Patterns of Global Terrorism, 2003. According to the State Department, in July 2001 a small group of Aum members was arrested in Russia before it could follow through with its plan to set off bombs near the Imperial Palace in Tokyo, free Asahara, and smuggle him into Russia.

Contending with Terrorism 392

Peoples Liberation Forces (Fuerzas Populares de Liberacin); and the U.S. group known as the Covenant, the Sword, and the Arm of the Lord.44 From a counterterrorism perspective, the killing of a terrorist leader may backre by creating increased publicity for the groups cause and perhaps making the leader a martyr who will attract new members to the organization (or even subsequent organizations). Che Guevera is the most famous example of this phenomenon.45 There is some reason to believe that arresting a leader is more effective in damaging a group than is killing or assassinating him.46 But even a humiliating arrest can backre if the incarcerated leader continues to communicate with his group. Sheikh Omar Abd al-Rahman (the so-called Blind Sheikh), convicted of conspiracy in the 1993 bombing of the World Trade Center, is a notable example.47 In other cases, imprisoned leaders may prompt further violence by group members trying to free them (e.g., the BaaderMeinhof group and, again, al-Rahman).48 Thus, if a leader is captured and jailed, undermining his credibility and cutting off inammatory communications are critical to demoralizing his following.
44. Direct Actions four principal leaders, Joelle Aubron, Georges Cipriani, Nathalie Menigon, and Jean-Marc Rouillan, were arrested in February 1987, followed shortly thereafter by Max Frerot, effectively dismantling the leadership of the group and putting an end to its activities. Members of the Peoples Liberation Forces killed their own deputy leader when she appeared to be interested in negotiating with the Salvadoran government; the leader of the group, Salvador Cayetano Carpio, committed suicide shortly thereafter, resulting in the disintegration of the group and essentially its absorption into a larger organization, the Farabundo Mart National Liberation Front. After a four-day siege of their compound in April 1985, eight leaders of the Covenant, the Sword, and the Arm of the Lord were arrested and imprisoned, effectively ending the group. The Chilean group Manuel Rodriguez Patriotic Front Dissidents essentially ceased to exist because of the arrest of its key leaders in the 1990s. The latter reached a peace agreement with the Salvadoran government in 1991. See the MIPT Terrorism Knowledge Base, a database of domestic and international terrorist organizations, at http:/ /www.tkb.org. 45. Che Guevara was captured and killed by the Bolivian army in October 1967 and subsequently became a legendary gure who inspired leftist and separatist groups in Latin America and throughout the world. Leila Ali Khaled of the Popular Front for the Liberation of Palestine carried the book My Friend Che with her when she hijacked TWA ight 840 in August 1969. In the United States, the Weathermen also organized massive protests on the second anniversary of Ches death. Harvey Kushner, Encyclopedia of Terrorism (Thousand Oaks, Calif.: Sage, 2003), pp. 155156, 372, 406. 46. I am grateful to Mia Bloom for sharing this observation. 47. Sheikh Omar is the leader of the Egyptian al-Gama al-Islamiya, which is closely tied with Egyptian Islamic Jihad. Although imprisoned for life in the United States, he has continued to call on his followers to engage in violence, especially against Jews. He was also convicted for plotting to bomb the Holland and Lincoln Tunnels and the United Nations building and to assassinate Senator Alphonse DAmato and UN Secretary General Boutros Boutros-Ghali. See, for example, Anonymous, Through Our Enemies Eyes: Osama bin Laden, Radical Islam, and the Future of America (Washington, D.C.: Brasseys 2002), p. 274. 48. Members of the Baader-Meinhof group engaged in violence during numerous attempts to free their imprisoned comrades, for example. The Blind Sheikh has also prompted violence aimed at his release. Ibid.

How al-Qaida Ends 393

inability to pass the cause on to the next generation The concept of the failure to transition to the next generation is closely related to theories that posit that terrorist violence is associated with the rise and fall of generations, but here it is applied to individual case studies. As mentioned above, the nature of the groups ideology seems to have relevance to the crossgenerational staying power of that group. The left-wing/anarchistic groups of the 1970s, for example, were notorious for their inability to articulate a clear vision of their goals that could be handed down to successors after the rst generation of radical leaders departed or were eliminated.49 The Red Brigades, the Second of June Movement, the Japanese Red Army, the Weather Underground Organization/Weathermen, the Symbionese Liberation Army, and the BaaderMeinhof group are all examples of extremely dangerous, violent groups in which a leftist/anarchist ideology became bankrupt, leaving no possibility to transition to a second generation.50 Right-wing groups, which draw their inspiration from fascist or racist concepts, can also have difculty persisting over generations, though, as Martha Crenshaw observes, this may reect the challenges of tracking them over time rather than their actual disintegration.51 Examples include the numerous neoNazi groups in the United States and elsewhere.52 Still, the racist causes of many of these groups can persist long after the disappearance of the group itself; their movement underground,53 or their reemergence under a different name or structure, is common.54 Extensive examinations by academic experts and the Federal Bureau of Investigation of right-wing groups in the United States during the 1990s, especially after the 1995 Oklahoma City bombing, revealed their tendency to operate according to a common modus operandi, ideology, or intent; this includes the so-called leaderless resistance, which involves individual operatives or small cells functioning independently in
49. For an explanation of the major types of terrorist organizations, see Cronin, Behind the Curve, pp. 3942. 50. The Red Army Faction, a successor of the Baader-Meinhof group, arguably continued for some years and transitioned to what it called its third generation, with claims of attacks in the name of the RAF during the 1980s and early 1990s. The degree to which it truly was the same group is debatable. In any case, the dissolution of the Soviet Union severely undermined its ideology. See entry in MIPT Terrorism Knowledge Base. 51. Crenshaw, How Terrorism Declines, p. 78. On right-wing groups, see Cronin, Behind the Curve, pp. 3942. 52. These include the Christian Patriots, the Aryan Nations, the Ku Klux Klan, and The Order (a short-lived faction of Aryan Nations) in the United States, as well as the Anti-Zionist Movement in Italy and the National Warriors of South Africa. 53. In the United States, the Ku Klux Klan is a notable example. 54. Some groups, such as The Order (active between 1982 and 1984), have been idolized by their admirers and continue to exercise inuence.

Contending with Terrorism 394

pursuit of an understood purpose.55 Such organizational decentralization complicates conclusions about beginnings and endings of right-wing groups, but it may also militate against truly effective generational transition. Furthermore, to support their activities, some right-wing groups engage in criminal behavior such as the robbing of banks and armored cars, racketeering, and counterfeiting, which, in the United States, has provided evidence trails for federal authorities and undermined group longevity. The internal process that occurs during the transition from rst- to secondgeneration terrorist leaders is very sensitive. Failure to pass the legacy to a new generation is a common historical explanation for a terrorist groups decline or end. achievement of the cause Some terrorist organizations cease to exist once they have fullled their original objective. Two examples are the Irgun Zvai Leumi (National Military Organization, also known either by its Hebrew acronym ETZEL or simply as Irgun), founded in 1931 to protect Jews with force and to advance the cause of an independent Jewish state, and the African National Congress (ANC). As head of the Irgun, Menachem Begin, who would later become prime minister of Israel, ordered the 1946 bombing of the King David Hotel, headquarters of British rule in Palestine. The attack killed 92 people and hastened Britains withdrawal. Irgun disbanded with the creation of the state of Israel, when its members transitioned to participation in the new government. The ANC was created in 1912 and turned to terrorist tactics in the 1960s. Its attacks were met with an extremely violent campaign of right-wing counterstrikes as the apartheid regime began to wane.56 ANC leader Nelson Mandela, imprisoned for terrorist acts from 1964 to 1990, was elected South Africas rst president following the end of apartheid.57 The last ANC attack occurred in 1989, and the organization became a legal political actor in 1990, having achieved its objective of ending the apartheid regime. Walter Laqueur divides terrorist groups that attained their objectives into
55. The concept of leaderless existence apparently originated with Col. Ulius Louis Amoss, who wrote an essay with this as the title in 1962. It was later popularized by Louis R. Beam Jr., Aryan Nations leader and former Texas Ku Klux Klan Grand Dragon leader. See Beam, Leaderless Resistance, The Seditionist, No. 12 (February 1992). 56. During much of this period, the African National Congress was labeled a terrorist organization by the U.S. Department of Defense but not by the U.S. State Department. 57. The South-West African Peoples Organization underwent a similar transition: from orchestrating bombings in banks, stores, schools, and service stations to governing Namibia.

How al-Qaida Ends 395

three categories: (1) those with narrow, clearly dened aims that were realistically attainable; (2) those with powerful outside protectors; and (3) those facing imperial powers that were no longer willing or able to hold on to their colonies or protectorates.58 In the context of twenty-rst-century terrorism, additional categories are possible. Although it happens in a minority of cases, using terrorism to achieve an aim does sometimes succeed; to recognize this reality is not to condone the tactic and may even be a prerequisite to effectively countering it. negotiations toward a legitimate political process The opening of negotiations can be a catalyst for the decline or end of terrorist groups, potentially engendering a range of effects. Groups have transitioned to political legitimacy and away from terrorist behavior after the formal opening of a political process. Examples include the Provisional Irish Republican Army, whose participation in the multiparty talks with the British and Irish governments was crucial to the 1998 Good Friday agreement; the Palestine Liberation Organization (PLO), which entered a peace process with Israel during the 1990s; and the Liberation Tigers of Tamil Eelam (LTTE, or Tamil Tigers), which began talks with the Sri Lankan government, brokered by the Norwegian government, in 2002.59 But the typical scenario for a terrorist groups decline is usually much more complicated than simply the pursuit or achievement of a negotiated agreement. Despite the successful negotiated outcomes that can result between the major parties, a common effect of political processes is the splintering of groups into factions that support the negotiations (or their outcome) and those that do not. For example, the IRA splintered into the Real Irish Republican Army;60 and the Popular Front for the Liberation of Palestine (PFLP), Democratic Front for the Liberation of Palestine, and PFLPGeneral Command (GC) split with the PLO over the Israeli-Palestinian peace process. From a counterterrorist perspective, dividing groups can be a purpose of the negotiations process, as it isolates and potentially strangles the most radical factions. But such splinter58. For example, among those that had outside protectors, Laqueur includes the Palestinian Arab groups and the Croatian Ustasha. For those facing imperial powers not able to hold on to colonies, he includes the IRA (Britain and Ireland after World War I), the Irgun (Britain and the Palestine Mandate after World War II), and the EOKA (Britain and Cyprus, also after World War II). Walter Laqueur, Terrorism (London: Weidenfeld and Nicolson, 1977), p. 118. 59. The talks have stopped and started several times, and at this writing are threatened by renewed violence; however, it looks as if they will result in limited Tamil autonomy. 60. Other splinter groups include the Continuity IRA and the Irish National Liberation Army.

Contending with Terrorism 396

ing can also occur on the status quo (or, usually, pro-government) side, as happened in South Africa (with the Afrikaner white power group Farmers Force, or Boermag) and in Northern Ireland (with the Ulster Volunteer Force). Governments confront huge difculties when negotiating with organizations against which they are still ghting in either a counterterrorism campaign or a traditional war.61 The most extreme case of counterproductive splintering of status quo factions is Colombia, where the signing of the peace accords between the Colombian government and the Popular Liberation Army (Ejrcito Popular de Liberacin, or EPL) in 1984 resulted in the formation of right-wing paramilitary groups that disagreed with the granting of political status to the EPL. Before long, leftist groups, paramilitary units, and the Colombian army stepped up their attacks, unraveling the peace, increasing the violence, and further fractionating the political actors. Worse, splinter groups are often more violent than the mother organization, responding to the imperative to demonstrate their existence and signal their dissent. Splinter groups can be seen as engaging in a new layer of terrorism with respect to the original group or their own government. This can also be the case, for example, when groups enter elections and take on a governing role. In such situations, the long-term goal (a viable political outcome) and the short-term goal (the reduction in violence) may be at odds.62 A wide range of variables can determine the broader outcome of negotiations to end terrorism, including the nature of the organization of the group (with hierarchical groups having an advantage over groups that cannot control their members actions), the nature of the leadership of the group (where groups with strong leaders have an advantage over those that are decentralized), and the nature of public support for the cause (where groups with ambivalent constituencies may be more likely to compromise). There must also be negotiable aims, which are more likely to exist with territorially based groups than with those that follow left-wing, right-wing, or religious/spiritualist ideologies. Determining the degree to which opening a political dialogue with a terrorist group is a likely avenue for the decline of the group and a reduction in violence is a highly differentiated calculation. Negotiations, however, need not be a formalized process and need not occur
61. For a discussion of this problem, see Fred Charles Ilk, Every War Must End, 2d ed. (New York: Columbia University Press, 2005), pp. 84105. 62. Negotiations, however, complicate a terrorist organizations efforts to perpetuate its own absolutist perspective in justifying the use of terrorist violence. See Adrian Guelke, The Age of Terrorism and the International Political System (London: I.B. Tauris, 1998), pp. 162181.

How al-Qaida Ends 397

only with the leadership of a group. Arguably, a form of negotiation with a terrorist organization, or more precisely with its members, is the offer of amnesty to those willing to stop engaging in violence and come forth with information about their fellow operatives. The classic case of a successful amnesty is the Italian governments 1979 and 1982 repentance legislation and the Red Brigades.63 In another case, the government of Alberto Fujimori in Peru offered amnesty to members of Shining Path, both after Guzmns capture and during the waning days of the group. As Robert Art and Louise Richardson point out in their comparative study of state counterterrorism policies, an amnesty may be most successful when an organization is facing defeat and its members have an incentive to seek a way out of what they see as a losing cause.64 diminishment of popular support Terrorist groups are strategic actors that usually deliberate about their targets and calculate the effects of attacks on their constituent populations. Miscalculations, however, can undermine a groups cause, resulting in plummeting popular support and even its demise. Terrorist groups generally cannot survive without either active or passive support from a surrounding population. Examples of active support include hiding members, raising money, and, especially, joining the organization. Passive support, as the term implies, is more diffuse and includes actions such as ignoring obvious signs of terrorist group activity, declining to cooperate with police investigations, sending money to organizations that act as fronts for the group, and expressing support for the groups objectives. Popular support for a terrorist group can dissipate for a number of reasons. First, people who are not especially interested in its political aims may fear government counteraction. Apathy is a powerful force; all else being equal, most people naturally prefer to carry on their daily lives without the threat of being targeted by counterterrorism laws, regulations, sanctions, and raids. Sometimes even highly radicalized populations can pull back active or passive

63. Franco Ferracuti, Ideology and Repentance: Terrorism in Italy, in Reich, Origins of Terrorism, pp. 5964; and Leonard Weinberg, Italy and the Red Brigades, in Art and Richardson, Democracy and Counterterrorism. 64. Other examples cited by Art and Richardson include amnesties or other incentives given to members of the ETA, the Shining Path, the FALN, the IRA, and the Tamil Tigers, with degrees of success related to whether members of a group perceived it as likely to prevail. See Conclusion, in Art and Richardson, Democracy and Counterterrorism.

Contending with Terrorism 398

support for a group, especially if the government engages in strong repressive measures and people simply become exhausted. The apparent loss of local popular support for Chechen terrorist groups is a good example. Second, the government may offer supporters of a terrorist group a better alternative. Reform movements, increased spending, and creation of jobs in underserved areas are all tactics that can undermine the sources of terrorist violence. They can also result, however, in increased instability and a heightened sense of opportunitysituations that in the past have led to more terrorist acts. Evidence suggests that the extent to which societal conditions lead to a sense of indignation or frustrated ambition among certain segments of society during a period of transition might be a crucial factor for the decision to turn to terrorist violence. Sometimes terrorist attacks are seen as an effort to nudge the ow of history further in ones direction.65 Third, populations can become uninterested in the ideology or objectives of a terrorist group; events can evolve independently such that the groups aims become outdated or irrelevant. A sense of historical ripeness or opportunity may have been lost. Examples include many of the Marxist groups inspired by communist ideology and supported by the Soviet Union. This is arguably a major reason why the nature of international terrorism has evolved beyond primary reliance on state sponsorship toward a broader range of criminal or entrepreneurial behavior. Fourth, a terrorist groups attacks can cause revulsion among its actual or potential public constituency.66 This is a historically common strategic error and can cause the group to implode. Independent of the counterterrorist activity of a government, a terrorist group may choose a target that a wide range of its constituents consider illegitimate. This occurred, for example, with the Omagh bombing. Despite hasty subsequent statements by RIRA leaders that they did not intend to kill innocent civilians, the group never recovered in the eyes of the community.67 Other examples of strategic miscalculation abound.

65. Peter A. Lupsha, Explanation of Political Violence: Some Psychological Theories versus Indignation, Politics and Society, Vol. 2, No. 1 (Fall 1971), pp. 89104; and Luigi Bonanate, Some Unanticipated Consequences of Terrorism, Journal of Peace Research, Vol. 16, No. 3 (May 1979), pp. 197211. 66. Terrorist operations, albeit shocking and tragic, at least as often increase the level of public support for the cause and indeed are designed to do so. Examples include the PLO, the Internal Macedonian Revolutionary Organization, the IRA, and nineteenth-century Russian terrorist groups. See Laqueur, Terrorism. 67. See Sean Boyne, The Real IRA: After Omagh, What Now? Janes Intelligence Review, October 1998, http:/ /www.janes.com/regional_news/europe/news/jir/jir980824_1_n.shtml.

How al-Qaida Ends 399

In February 1970 the PFLP-GC sabotaged a Swissair plane en route to Tel Aviv, resulting in the deaths of all 47 passengers, 15 of whom were Israelis. The PFLP-GC at rst took responsibility but then tried unsuccessfully to retract its claim when popular revulsion began to surface.68 Similarly, there has been revulsion among the Basque population in Spain to attacks by the separatist group Basque Fatherland and Liberty (Euzkadi Ta Askatasuna, or ETA), which some observers credit with the declining popularity of the group.69 Public revulsion was a factor in the undermining of support for Sikh separatism in India, a movement directed at establishing an independent state of Khalistan that killed tens of thousands between 1981 and 1995 and was responsible for the assassination of Indian Prime Minister Indira Gandhi on October 31, 1984.70 Popular revulsion against terrorist attacks can have immediate effects. Arguably the most well developed and broadly based conduit for resource collection in the world is the connection between the Tamil Tigers and the dispersed Tamil diaspora. The LTTEs desire to avoid the terrorist organization label in the postSeptember 11 world and shore up its base of popular support was an element in the groups December 2001 decision to pursue a negotiated solution.71 Likewise, a state-sponsored terrorist group can lose support when the state decides that it is no longer interested in using terrorism, responds to pressure from other states, has more important competing goals, or loses the ability to continue its support.72 Libyas expulsion of the Palestinian terrorist Abu Nidal and cutting off of support to Palestinian groups such as the Palestine Islamic Jihad and the PFLP-GC are notable examples.

68. MIPT database incident prole, http:/ /64.233.167.104/search?q?cache:vBXLiW4gwekJ:www .tkb.org/Incident.jsp%3FincID%3D372 popular revulsion terrorist group&hl en. The PFLPGC has ties with Syria and Libya and may have had a role in the bombing of Pan Am 103 over Lockerbie, Scotland. See Cronin et al., Foreign Terrorist Organizations, pp. 8082; and Kenneth Katzman, The PLO and Its Factions, CRS Report for Congress (Washington, D.C.: Congressional Research Service, June 10, 2002), Order Code RS21235. The state sponsorship of this group from the 1980s may have reduced its dependency on a local constituency. 69. See, for example, Goldie Shabad and Francisco Jos Llera Ramo, Political Violence in a Democratic State: Basque Terrorism in Spain, in Crenshaw, Terrorism in Context, pp. 410469, especially pp. 455462. 70. Mark Juergensmeyer, Terror in the Mind of God: The Global Rise of Religious Violence (Berkeley: University of California Press, 2000), pp. 84101. See also C. Christine Fair, Diaspora Involvement in Insurgencies: Insights from the Khalistan and Tamil Eelam Movements, Nationalism and Ethnic Politics, Vol. 11, No. 1 (Spring 2005), pp. 125156. 71. Fair, Diaspora Involvement in Insurgencies. 72. For more on state sponsorship, see Daniel Byman, Deadly Connections: States That Sponsor Terrorism (Cambridge: Cambridge University Press, 2005).

Contending with Terrorism 400

military force and the repression of terrorist groups The use of military force has hastened the decline or ended a number of terrorist groups, including the late-nineteenth-century Russian group Narodnaya Volya, Shining Path, and the Kurdistan Workers Party. From the states perspective, military force offers a readily available means that is under its control. Although terrorism is indeed arguably a form of war, terrorists use asymmetrical violence, by denition, because they are unable or unwilling to meet a status quo government on the battleeld. Shifting the violence to a form that is familiar and probably advantageous for the state is an understandable response. In some circumstances, it is also successful. Historically, military force has taken two forms: intervention, when the threat is located mainly beyond the borders of the target state (as with Israels 1982 involvement in Lebanon); or repression, when the threat is considered mainly a domestic one (as with the PKK). More typically, the state will use some combination of the two (as in Colombia).73 The effects of the use of repressive military force in some cases may prove to be temporary or counterproductive; in other cases, it may result in the export of the problem to another country. The classic contemporary case is the Russian counterterrorism campaign in Chechnya.74 Russian involvement in the second Chechen war appears to have produced a transition in the Chechen resistance, with more terrorist attacks in the rest of Russia, greater reliance on suicide bombers, and the growing inuence of militant Islamic ghters. To the extent that the Chechens originally engaged in a classic insurgency rather than in terrorism, they have since 2002 altered their tactics toward increasing attacks on Russian civilians. The strong repressive response by the Russian government has apparently facilitated the spread of the conict to neighboring areas, including Ingushetia and Dagestan. And there seems to be no end in sight, given the increasing radicalization and identication of some Chechen factions with the al-Qaida movement.75
73. On intervention, see Adam Roberts, The War on Terror in Historical Perspective, Survival, Vol. 47, No. 2 (Summer 2005), pp. 115121. 74. For recent work on this subject, see Mark Kramer, The Perils of Counterinsurgency: Russias War in Chechnya, International Security, Vol. 29, No. 3 (Winter 2004/05), pp. 563; and Mark Kramer, Guerrilla Warfare, Counterinsurgency, and Terrorism in the North Caucasus: The Military Dimension of the Russian-Chechen Conict, Europe-Asia Studies, Vol. 57, No. 2 (March 2005), pp. 209290. 75. The International Institute for Strategic Studies (IISS) claims that Russian forces suffered 4,749 casualties between August 2002 and August 2003, the highest gure since the second Chechen conict began in 1999. International Institute for Strategic Studies, The Military Balance, 2003/2004 (London: IISS, 2003), pp. 8687. For more on the Chechen case, see Audrey Kurth Cronin, Russia and Chechnya, in Art and Richardson, Democracy and Counterterrorism.

How al-Qaida Ends 401

Democracies or liberal governments face particular difculties in repressing terrorist groups. Because military or police action requires a target, the use of force against operatives works best in situations where members of the organization can be separated from the general population. This essentially forces proling or some method of distinguishing members from nonmembersalways a sensitive issue, particularly when the only available means of discrimination relate to how members are dened (race, age, religion, nationality, etc.) rather than to what they do (or are planning to do). Excellent intelligence is essential for the latter (especially in advance of an attack), but even in the best of situations, it is typically scarce. Repressive measures also carry high resource and opportunity costs. Long-term repressive measures against suspected operatives may challenge civil liberties and human rights, undermine domestic support, polarize political parties, and strain the fabric of the state itself, further undercutting the states ability to respond effectively to future terrorist attacks. transition to another modus operandi In some cases, groups can move from the use of terrorism toward either criminal behavior or more classic conventional warfare. The transition to criminal behavior implies a shift away from a primary emphasis on collecting resources as a means of pursuing political ends toward acquiring material goods and prot that become ends in themselves. Groups that have undertaken such transitions in recent years include Abu Sayyaf in the Philippines and arguably all of the major so-called narco-terrorist groups in Colombia.76 Beginning in 2000, Abu Sayyaf shifted its focus from bombings and targeted executions to the taking of foreign hostages and their exchange for millions of dollars in ransom.77 The Revolutionary Armed Forces of Colombia uses a variety of mechanisms to raise fundsincluding kidnapping for ransom, extortion, and especially drug trafckingrunning operations that yield as much as $1 billion annually.78 Terrorist groups can also escalate to insurgency or even conventional war, especially if they enjoy state support. Notable examples include the Kashmiri separatist groups, the Khmer Rouge, and the Communist Party of Nepal

76. These would include the Revolutionary Armed Forces of Colombia, the National Liberation Army, and the United Self-Defense Forces of Colombia. 77. See Larry Niksch, Abu Sayyaf: Target of Philippine-U.S. Antiterrorism Cooperation, CRS Report for Congress (Washington, D.C.: Congressional Research Service, updated April 8, 2003), Order Code RL31265; and Cronin et al., Foreign Terrorist Organizations, pp. 47. 78. Cronin et al., Foreign Terrorist Organizations, p. 92.

Contending with Terrorism 402

Maoists. Transitions in and out of insurgency are especially common among ethnonationalist/separatist groups, whose connection to a particular territory and grounding in an ethnic population provide a natural base; in these situations, the evolution involves changes in size or type of operations (do they operate as a military unit and attack mainly other military targets?) and whether or not the organization holds territory (even temporarily). Terrorism and insurgency are not the same, but they are related. Very weak, territorially based movements may use terrorist attacks and transition to insurgency when they gain strength, especially when their enemy is a state government (as was the case for most groups in the twentieth century). One example is Algerias Armed Islamic Group, which massacred tens of thousands of civilians in the civil war that followed the Islamic Salvation Fronts victory in the 1991 parliamentary elections. The key in understanding the relationship between the two tactics is to analyze the groups motivation, its attraction to a particular constituency, its strength, and the degree to which its goals are associated with control of a piece of territory. Transitions to full-blown conventional war, on the other hand, can occur when the group is able to control the behavior of a state according to its own interests, or even when an act of terrorism has completely unintended consequences.79

Is al-Qaida Unique among Terrorist Organizations?

Four characteristics distinguish al-Qaida from its predecessors in either nature or degree: its uid organization, recruitment methods, funding, and means of communication. fluid organization The al-Qaida of September 2001 no longer exists. As a result of the war on terrorism, it has evolved into an increasingly diffuse network of afliated groups, driven by the worldview that al-Qaida represents. In deciding in 1996 to be, essentially, a visible organization, running training camps and occupying territory in Afghanistan, al-Qaida may have made an important tactical error; this, in part, explains the immediate success of the U.S.-led coalitions war
79. As Adam Roberts notes, the outbreak of World War I is a principal example: the Bosnian-Serb student who killed Archduke Franz Ferdinand in July 1914 had no intention of setting off an international cataclysm. See Ratko Parezanin, Mlada Bosna I prvi svetski rat [Young Bosnia and the First World War] (Munich: Iskra, 1974), cited in Roberts, The War on Terror in Historical Perspective, p. 107 n. 15.

How al-Qaida Ends 403

in Afghanistan.80 Since then, it has begun to resemble more closely a global jihad movement, increasingly consisting of web-directed and cyber-linked groups and ad hoc cells.81 In its evolution, al-Qaida has demonstrated an unusual resilience and international reach. It has become, in the words of Porter Goss, only one facet of the threat from a broader Sunni jihadist movement.82 No previous terrorist organization has exhibited the complexity, agility, and global reach of al-Qaida, with its uid operational style based increasingly on a common mission statement and objectives, rather than on standard operating procedures and an organizational structure.83 Al-Qaida has been the focal point of a hybrid terrorist coalition for some time, with ties to inspired freelancers and other terrorist organizations both old and new.84 Some observers argue that considering al-Qaida an organization is misleading; rather it is more like a nebula of independent entities (including loosely associated individuals) that share an ideology and cooperate with each other.85 The original umbrella group, the International Islamic Front for Jihad against Jews and Crusaders, formed in 1998, included not only alQaida but also groups from Algeria, Bangladesh, Egypt, and Pakistan. A sampling of groups that are connected in some way includes the Moro Islamic Liberation Front (Philippines), Jemaah Islamiyah (Southeast Asia), Egyptian Islamic Jihad (which merged with al-Qaida in 2001), al-Ansar Mujahidin (Chechnya), al-Gamaa al-Islamiyya (primarily Egypt, but has a worldwide presence), Abu Sayyaf (Philippines), the Islamic Movement of Uzbekistan, the Salast Group for Call and Combat (Algeria), and Harakat ul-Mujahidin (Pakistan/Kashmir). Some experts see al-Qaidas increased reliance on connections to other groups as a sign of weakness; others see it as a worrisome indicator of growing strength, especially with groups that formerly focused on local

80. Rapoport, The Four Waves of Modern Terrorism, p. 66. 81. Coll and Glasser, Terrorists Turn to the Web as Base of Operations. 82. Porter J. Goss, director of central intelligence, testimony before the Senate Select Committee on Intelligence, Global Intelligence Challenges, 2005: Meeting Long-Term Challenges with a LongTerm Strategy, 109th Cong., 2d sess., February 16, 2005, http:/ /www.cia.gov/cia/public_affairs/ speeches/2004/Goss_testimony_02162005.html. 83. Possible exceptions include the international anarchist movement, which was conned to Russia, Europe, and the United States. 84. David Johnston, Don Van Natta Jr., and Judith Miller, Qaidas New Links Increase Threats from Far-Flung Sites, New York Times, June 16, 2002. Many commentators have pointed out that al-Qaida is translated as the base, the foundation, or even the method. It was not intended at its founding to be a single structure. 85. Xavier Raufer, Al-Qaeda: A Different Diagnosis, Studies in Conict and Terrorism, Vol. 26, No. 6 (November/December 2003), pp. 391398.

Contending with Terrorism 404

issues and now display evidence of convergence on al-Qaidas Salast, antiU.S., anti-West agenda.86 The nature, size, structure, and reach of the coalition have long been subject to debate. Despite claims of some Western experts, no one knows how many members al-Qaida has currently or had in the past. U.S. intelligence sources place the number of individuals who underwent training in camps in Afghanistan from 1996 through the fall of 2001 at between 10,000 and 20,000;87 the gure is inexact, in part, because of disagreement over the total number of such camps and because not all attendees became members.88 The International Institute for Strategic Studies in 2004 estimated that 2,000 al-Qaida operatives had been captured or killed and that a pool of 18,000 potential al-Qaida operatives remained.89 These numbers can be misleading, however: it would be a mistake to think of al-Qaida as a conventional force, because even a few trained ghters can mobilize many willing foot soldiers as martyrs. methods of recruitment The staying power of al-Qaida is at least in part related to the way the group has perpetuated itself; in many senses, al-Qaida is closer to a social movement than a terrorist group.90 Involvement in the movement has come not from pressure by senior al-Qaida members but mainly from local volunteers competing to win a chance to train or participate in some fashion.91 The process seems to be more a matter of joining than being recruited,92 and thus the tra86. For an early analysis of the arguments of both sides, see Audrey Kurth Cronin, Al-Qaeda after the Iraq Conict, CRS Report for Congress (Washington, D.C.: Congressional Research Service, May 23, 2003), Order Code RS21529. 87. Central Intelligence Agency, Afghanistan: An Incubator for International Terrorism, CTC 0140004, March 27, 2001; and Central Intelligence Agency, Al-Qaida Still Well Positioned to Recruit Terrorists, July 1, 2002, p. 1, both cited in Final Report of the National Commission on Terrorist Attacks upon the United States (New York: W.W. Norton, 2004), p. 67 n. 78. 88. Anonymous, Through Our Enemies Eyes, especially pp. 126, 130131; Marc Sageman, Understanding Terror Networks (Philadelphia: University of Pennsylvania Press, 2004), especially p. 121; and David Johnson and Don Van Natta Jr., U.S. Ofcials See Signs of a Revived al-Qaida, New York Times, May 17, 2003. 89. International Institute for Strategic Studies, Strategic Survey, 2003/4 (London: Routledge, 2005), p. 6. 90. See also The Worldwide Threat, 2004: Challenges in a Changing Global Context, testimony of Director of Central Intelligence George J. Tenet before the Senate Select Committee on Intelligence, 108th Cong., 2d sess., February 24, 2004, http:/ /www.cia.gov/cia/public_affairs/speeches/ 2004/dci_speech_02142004.html. 91. According to Sagemans analysis, operatives have gathered in regional clusters, which he labels the central staff of al-Qaida, the Southeast Asian cluster, the Maghreb cluster, and the core Arab cluster. Sageman, Understanding Terror Networks. 92. Ibid.; and Jeffrey Cozzens, Islamist Groups Develop New Recruiting Strategies, Janes Intelligence Review, February 1, 2005.

How al-Qaida Ends 405

ditional organizational approach to analyzing this group is misguided. But the draw of al-Qaida should also not be overstated: in the evolving pattern of associations, attraction to the mission or ideology seems to have been a necessary but not sufcient condition. Exposure to an ideology is not enough, as reected in the general failure of al-Qaida to recruit members in Afghanistan and Sudan, where its headquarters were once located.93 As psychiatrist Marc Sageman illustrates, social bonds, not ideology, apparently play a more important role in al-Qaidas patterns of global organization.94 Sagemans study of established links among identied al-Qaida operatives indicates that they joined the organization mainly because of ties of kinship and friendship, facilitated by what he calls a bridging person or entry point, perpetuated in a series of local clusters in the Maghreb and Southeast Asia, for example. In recent years, operatives have been connected to al-Qaida and its agenda in an even more informal way, having apparently not gone to camps or had much formal training: examples include those engaged in the London bombings of July 7 and 21, 2005, the Istanbul attacks of November 15 and 20, 2003, and the Casablanca attacks of May 16, 2003.95 This loose connectedness is not an accident: bin Laden describes al-Qaida as the vanguard of the Muslim nation and does not claim to exercise command and control over his followers.96 Although many groups boast of a connection to al-Qaidas ideology, there are often no logistical trails and thus no links for traditional intelligence methods to examine.97 This explains, for example, the tremendous difculty in establishing connections between a radical mosque, bombers, bomb makers, supporters, and al-Qaida in advance of an attack (not to mention after an attack). Another concern has been the parallel development of Salast networks apparently drawing European Muslims into combat against Western forces in Iraq. The European Unions counterterrorism coordinator, Gijs de Vries, for example, has cautioned that these battle-hardened veterans of the Iraq conict
93. Sageman, Understanding Terrorist Networks, p. 119. Sagemans book analyzes the patterns of growth of cells involved in past attacks, including vigorous refutation of the thesis that exposure to ideology alone explains the growth of al-Qaida. 94. Ibid., p. 178. 95. Jack Kalpakian, Building the Human Bomb: The Case of the 16 May 2003 Attacks in Casablanca, Studies in Conict and Terrorism, Vol. 28, No. 2 (MarchApril 2005), pp. 113127. 96. Al Jazeera TV, September 10, 2003, cited in Michael Scheuer, Coalition Warfare: How alQaeda Uses the World Islamic Front against Crusaders and Jews, Part I, Terrorism Focus, Vol. 2, No. 7, March 31, 2005, http:/ /www.jamestown.org/terrorism/news/article.php?articleid 2369530. 97. Paul Pillar, Counterterrorism after al-Qaida, Washington Quarterly, Vol. 27, No. 3 (Summer 2004), pp. 101113.

Contending with Terrorism 406

will return to attack Western targets in Europe. The Ansar al-Islam plot to attack the 2004 NATO summit in Turkey was, according to Turkish sources, developed in part by operatives who had fought in Iraq.98 A proportion of those recently drawn to the al-Qaida movement joined after receiving a Salast message disseminated over the internet. Such direct messages normally do not pass through the traditional process of vetting by an imam. European counterterrorism ofcials thus worry about members of an alienated diaspora sometimes second- and third-generation immigrantswho may be vulnerable to the message because they are not thoroughly trained in fundamental concepts of Islam, are alienated from their parents, and feel isolated in the communities in which they nd themselves. The impulse to join the movement arises from a desire to belong to a group in a context where the operative is excluded from, repulsed by, or incapable of successful integration into a Western community.99 Thus, with al-Qaida, the twentieth-century focus on structure and function is neither timely nor sufcient. Tracing the command and control relationships in such a dramatically changing movement is enormously difcult,100 which makes comparisons with earlier, more traditional terrorist groups harder but by no means impossible; one detects parallels, for example, between al-Qaida and the global terrorist movements that developed in the late nineteenth century, including anarchist and social revolutionary groups.101 means of support Financial support of al-Qaida has ranged from money channeled through charitable organizations to grants given to local terrorist groups that present promising plans for attacks that serve al-Qaidas general goals.102 The majority of its operations have relied at most on a small amount of seed money provided by the organization, supplemented by operatives engaged in petty crime and fraud.103 Indeed, beginning in 2003, many terrorism experts agreed that al-Qaida could best be described as a franchise organization with a
98. Cozzens, Islamist Groups Develop New Recruiting Strategies. 99. Sageman, Understanding Terrorist Networks, especially p. 92. 100. On this point, see Bruce Hoffman, The Changing Face of al-Qaida and the Global War on Terrorism, Studies in Conict and Terrorism, Vol. 27, No. 6 (December 2004), p. 556. 101. On this subject, see Laqueur, Terrorism, especially pp. 320; and James Joll, The Second International, 18891914 (London: Weidenfeld and Nicolson, 1955). 102. Bruce Hoffman, The Leadership Secrets of Osama bin Laden: The Terrorist as CEO, Atlantic Monthly, April 2003, pp. 2627. 103. Sageman, Understanding Terrorist Networks, pp. 5051.

How al-Qaida Ends 407

marketable brand.104 Relatively little money is required for most al-Qaidaassociated attacks. As the International Institute for Strategic Studies points out, the 2002 Bali bombing cost less than $35,000, the 2000 USS Cole operation about $50,000, and the September 11 attacks less than $500,000.105 Another element of support has been the many autonomous businesses owned or controlled by al-Qaida; at one point, bin Laden was reputed to own or control approximately eighty companies around the world. Many of these legitimately continue to earn a prot, providing a self-sustaining source for the movement. International counterterrorism efforts to control al-Qaida nancing have reaped at least $147 million in frozen assets.106 Still, cutting the nancial lifeline of an agile and low-cost movement that has reportedly amassed billions of dollars and needs few resources to carry out attacks remains a formidable undertaking. Choking off funds destined for al-Qaida through regulatory oversight confronts numerous challenges. Formal banking channels are not necessary for many transfers, which instead can occur through informal channels known as alternative remittance systems, informal value transfer systems, parallel banking, or underground banking. Examples include the much-discussed hawala or hundi transfer networks and the Black Market Peso Exchange that operate through family ties or unofcial reciprocal arrangements.107 Value can be stored in commodities such as diamonds and gold that are moved through areas with partial or problematical state sovereignty. Al-Qaida has also used charities to raise and move funds, with a relatively small proportion of gifts being siphoned off for illegitimate purposes, often without the knowledge of donors.108 Yet efforts to cut off charitable ows to impoverished areas may

104. See Douglas Farah and Peter Finn, Terrorism, Inc.: Al-Qaida Franchises Brand of Violence to Groups across World, Washington Post, November 21, 2003; Daniel Benjamin, Are the Sparks Catching? Washington Post, November 23, 2003; and Sebastian Rotella and Richard C. Paddock, Experts See Major Shift in al-Qaidas Strategy, Los Angeles Times, November 19, 2003. 105. International Institute for Strategic Studies, Strategic Survey, 2003/4, p. 8. 106. U.S.: Terror Funding Stymied, CBS News, January 11, 2005, http:/ /www.cbsnews.com/ stories/2005/01/11/terror/main666168.shtml. 107. These are methods of transferring value anonymously, without electronic traceability or paper trails. See Rensselaer Lee, Terrorist Financing: The U.S. and International Response, CRS Report for Congress (Washington, D.C.: Congressional Research Service, December 6, 2002), pp. 1113, Order Code RL31658. 108. For information on informal nancing mechanisms and U.S. efforts to control them, see testimony of Assistant Secretary Juan Carlos Zarate for Terrorist Financing and Financial Crimes, U.S. Department of the Treasury, before the House Financial Services Committees Subcommittee on Oversight and Investigations, 109th Cong., 2d sess., February 16, 2005, http:/ /www.ustreas.gov/ press/releases/js2256.htm.

Contending with Terrorism 408

harm many genuinely needy recipients and could result in heightened resentment, which in turn may generate additional political support for the movement.109 Al-Qaidas scal autonomy makes the network more autonomous than its late-twentieth-century state-sponsored predecessors. means of communication The al-Qaida movement has successfully used the tools of globalization to enable it to communicate with multiple audiences, including potential new members, new recruits, active supporters, passive sympathizers, neutral observers, enemy governments, and potential victims. These tools include mobile phones, text messaging, instant messaging, and especially websites, email, blogs, and chat rooms, which can be used for administrative tasks, fund-raising, research, and logistical coordination of attacks.110 Although alQaida is not the only terrorist group to exploit these means, it is especially adept at doing so. A crucial facilitator for the perpetuation of the movement is the use of websites both to convey messages, fatwas, claims of attacks, and warnings to the American public, as well as to educate future participants, embed instructions to operatives, and rally sympathizers to the cause. The internet is an important factor in building and perpetuating the image of al-Qaida and in maintaining the organizations reputation. It provides easy access to the media, which facilitates al-Qaidas psychological warfare against the West. Indoctrinating and teaching new recruits is facilitated by the internet, notably through the dissemination of al-Qaidas widely publicized training manual (nicknamed The Encyclopedia of Jihad) that explains how to organize and run a cell, as well as carry out attacks.111 Websites and chat rooms are used to offer practical advice and facilitate the fraternal bonds that are crucial to al-

109. For more on illegal and legal globalized networks of nancing, see Terrorist Financing, Report of an Independent Task Force Sponsored by the Council on Foreign Relations (New York: Council on Foreign Relations Press, October 2002); Update on the Global Campaign against Terrorist Financing, Second Report of an Independent Task Force on Terrorist Financing Sponsored by the Council on Foreign Relations (New York: Council on Foreign Relations Press, June 15, 2004); Loretta Napoleoni, Modern Jihad: Tracing the Dollars behind the Terror Networks (London: Pluto, 2003); Douglas Farah, Blood from Stones: The Secret Financial Network of Terror (New York: Broadway Books, 2004); and Thomas J. Biersteker and Sue E. Eckert, eds., Countering the Financing of Global Terrorism (London: Routledge, forthcoming). 110. For more on this phenomenon, see Cronin, Behind the Curve, pp. 3058. 111. For an analysis of current terrorist use of the internet, see Gabriel Weimann, www.terror.net: How Modern Terrorism Uses the Internet, Special Report, No. 116 (Washington, D.C.: United States Institute of Peace, March 2004).

How al-Qaida Ends 409

Qaida. In a sense, members of the movement no longer need to join an organization at all, for the individual can participate with the stroke of a few keys. The debate over the size, structure, and membership of al-Qaida may be a quaint relic of the twentieth century, displaced by the leveling effects of twenty-rst-century technology. The new means of communication also offer practical advantages. Members of al-Qaida use the web as a vast source of research and data mining to scope out future attack sites or develop new weapons technology at a low cost and a high level of sophistication. On January 15, 2003, for example, U.S. Secretary of Defense Donald Rumsfeld quoted an al-Qaida training manual retrieved by American troops in Afghanistan that advised trainees that at least 80 percent of the information needed about the enemy could be collected from open, legal sources.112

Earlier Terrorist Groups, al-Qaida, and U.S. Policy Implications

Al-Qaidas uid organization, methods of recruitment, funding, and means of communication distinguish it as an advancement in twenty-rst-century terrorist groups. Al-Qaida is a product of the times. Yet it also echoes historical predecessors, expanding on such factors as the international links and ideological drive of nineteenth-century anarchists, the open-ended aims of Aum Shinrikyo, the brilliance in public communications of the early PLO, and the taste for mass casualty attacks of twentieth-century Sikh separatists or Hezbollah. Al-Qaida is an amalgam of old and new, reecting twenty-rstcentury advances in means or matters of degree rather than true originality; still, most analysts miss the connections with its predecessors and are blinded by its solipsistic rhetoric. That is a mistake. The pressing challenge is to determine which lessons from the decline of earlier terrorist groups are relevant to al-Qaida and which are not. First, past experience with terrorism indicates that al-Qaida will not end if Osama bin Laden is killed. There are many other reasons to pursue him, including bringing him to justice, removing his leadership and expertise, and increasing esprit de corps on the Western side (whose credibility is sapped because of bin Ladens enduring elusiveness). The argument that his demise will end al-Qaida is tinged with emotion, not dispassionate analysis. Organi112. Ibid., p. 7.

Contending with Terrorism 410

zations that have been crippled by the killing of their leader have been hierarchically structured, reecting to some degree a cult of personality, and have lacked a viable successor. Al-Qaida meets neither of these criteria: it has a mutable structure with a strong, even increasing, emphasis on individual cells and local initiative. It is not the rst organization to operate in this way; and the demise of similar terrorist groups required much more than the death of one person.113 Unlike the PKK and Shining Path, al-Qaida is not driven by a cult of personality; despite his astonishing popularity, bin Laden has deliberately avoided allowing the movement to revolve around his own persona, preferring instead to keep his personal habits private, to talk of the insignicance of his own fate, and to project the image of a humble man eager to die for his beliefs. As for a viable replacement, bin Laden has often spoken openly of a succession plan, and that plan has to a large degree already taken effect. Furthermore, his capture or killing would produce its own countervailing negative consequences, including (most likely) the creation of a powerful martyr. On balance, the removal of bin Laden would have important potential benets, but to believe that it would kill al-Qaida is to be ahistorical and naive. That al-Qaida is already dead. Second, although there was a time when the failure to transition to a new generation might have been a viable nale for al-Qaida, that time is long past. Al-Qaida has transitioned to a second, third, and arguably fourth generation. The reason relates especially to the second distinctive element of al-Qaida: its method of recruitment or, more accurately, its attraction of radicalized followers (both individuals and groups), many of whom in turn are connected to existing local networks. Al-Qaidas spread has been compared to a virus or a bacterium, dispersing its contagion to disparate sites.114 Although this is a seductive analogy, it is also misleading: the perpetuation of al-Qaida is a sentient process involving well-considered marketing strategies and deliberate tactical decisions, not a mindless disease process; thinking of it as a disease shores up the unfortunate American tendency to avoid analyzing the mentality of the enemy. Al-Qaida is operating with a long-term strategy and is certainly not following the left-wing groups of the 1970s in their failure to articulate a coherent ideological vision or the peripatetic right-wing groups of the twentieth
113. Comparable cell-based immortal terrorist networks have included the social revolutionary and anarchist movements of the late nineteenth and early twentieth centuries. I am indebted to Timothy Hoyt for this observation. 114. See, for example, Corine Hegland, Global Jihad, National Journal, May 8, 2004, pp. 1396 1402.

How al-Qaida Ends 411

century. It has transitioned beyond its original structure and now represents a multigenerational threat with staying power comparable to the enthonationalist groups of the twentieth century. Likewise, arguments about whether alQaida is best described primarily as an ideology or by its opposition to foreign occupation of Muslim lands are specious: al-Qaidas adherents use both rationales to spread their links. The movement is opportunistic. The challenge for the United States and its allies is to move beyond rigid mind-sets and paradigms, do more in-depth analysis, and be more nimble and strategic in response to al-Qaidas uid agenda. The third and fourth models of a terrorist organizations endachievement of the groups cause and transition toward a political role, negotiations, or amnestyalso bear little relevance to al-Qaida today. It is hard to conceive of al-Qaida fully achieving its aims, in part because those aims have evolved over time, variably including achievement of a pan-Islamic caliphate, the overthrow of non-Islamic regimes, and the expulsion of all indels from Muslim countries, not to mention support for the Palestinian cause and the killing of Americans and other so-called indels. Historically, terrorist groups that have achieved their ends have done so by articulating clear, limited objectives. Al-Qaidas goals, at least as articulated over recent years, could not be achieved without overturning an international political and economic system characterized by globalization and predominant U.S. power. As the historical record indicates, negotiations or a transition to a legitimate political process requires feasible terms and a sense of stalemate in the struggle. Also, members of terrorist groups seeking negotiations often have an incentive to nd a way out of what they consider a losing cause. None of this describes bin Ladens al-Qaida. This points to another issue. As al-Qaida has become a hybrid or virtual organization, rather than a coherent hierarchical organization, swallowing its propaganda and treating it as a unied whole is a mistake. It is possible that bin Laden and his lieutenants have attempted to cobble together such disparate entities (or those entities have opportunistically attached themselves to alQaida) that they have stretched beyond the point at which their interests can be represented in this movement. Some of the local groups that have recently claimed an association with al-Qaida have in the past borne more resemblance to ethnonationalist/separatist groups such as the PLO, the IRA, and the LTTE. Examples include local afliates in Indonesia, Morocco, Tunisia, and Turkey. This is not to argue that these groups aims are rightful or that their tactics are legitimate. Rather, because of its obsession with the notion of a monolithic al-

Contending with Terrorism 412

Qaida, the United States is glossing over both the extensive local variation within terrorist groups and their different goals; these groups have important points of divergence with al-Qaidas agenda, and the United States does no one any favors by failing to seriously analyze and exploit those local differences (except perhaps al-Qaida). The U.S. objective must be to enlarge the movements internal inconsistencies and differences. Al-Qaidas aims have become so sweeping that one might wonder whether they genuinely carry within them the achievement of specic local grievances. There is more hope of ending such groups through traditional methods if they are dealt with using traditional tools, even including, on a case-by-case basis, concessions or negotiations with specic local elements that may have negotiable or justiable terms (albeit pursued through an illegitimate tactic). The key is to emphasize the differences with al-Qaidas agenda and to drive a wedge between the movement and its recent adherents. The historical record of other terrorist groups indicates that it is a mistake to treat alQaida as a monolith, to lionize it as if it is an unprecedented phenomenon with all elements equally committed to its aims, for that eliminates a range of proven counterterrorist tools and techniques for ending it. It is also a mistake to nurture al-Qaidas rallying point, a hatred of Americans and a resentment of U.S. policies (especially in Iraq and between Israel and the Palestinians), thereby conveniently facilitating the glossing over of differences within. Fifth, reducing popular support, both active and passive, is an effective means of hastening the demise of some terrorist groups. This technique has received much attention among critics of George W. Bush and his administrations policies, many of whom argue that concentrating on the roots of terrorism is a necessary alternative to the current policy of emphasizing military force. This is a supercial argument, however, as it can be countered that a roots approach is precisely at the heart of current U.S. policy: the promotion of democracy may be seen as an idealistic effort to provide an alternative to populations in the Muslim world, frustrated by corrupt governance, discrimination, unemployment, and stagnation. But the participants in this debate are missing the point.115 The problem is timing: democratization is a decades-long approach to a short-term and immediate problem whose solution must also be

115. As Jonathan Monten demonstrates, democratization is a long-standing goal of U.S. foreign policy; with respect to terrorism, however, both the vindicationists and the exemplarists are missing the point. Monten, The Roots of the Bush Doctrine: Power, Nationalism, and Democracy Promotion in U.S. Strategy, International Security, Vol. 29, No. 4 (Spring 2005), pp. 112156.

How al-Qaida Ends 413

measured in months, not just years. The efforts being undertaken are unlikely to have a rapid enough effect to counter the anger, frustration, and sense of humiliation that characterize passive supporters. As for those who actively sustain the movement through terrorist acts, it is obviously a discouraging development that many recent operatives have lived in, or even been natives of, democratic countriesthe March 2004 train attacks in Madrid and 2005 bombings in London being notable examples.116 The history of terrorism provides little comfort to those who believe that democratization is a good method to reduce active and passive support for terrorist attacks. There is no evidence that democratization correlates with a reduction in terrorism; in fact, available historical data suggest the opposite. Democratization was arguably the cause of much of the terrorism of the twentieth century. Moreover, democracy in the absence of strong political institutions and civil society could very well bring about radical Islamist governments over which the West would have little inuence. There are much worse things than terrorism, and it might continue to be treated as a regrettable but necessary accompaniment to change were it not for two potentially serious developments in the twenty-rst century: (1) the use of increasingly destructive weapons that push terrorist attacks well beyond the nuisance level, and (2) the growing likelihood that terrorism will lead to future systemic war. In any case, the long-term, idealistic, and otherwise admirable policy of democratization, viewed as Americanization in many parts of the world, does not represent a sufciently targeted response to undercutting popular support for alQaida. There are two vulnerabilities, however, where cutting the links between al-Qaida and its supporters hold promise: its means of funding and its means of communication. Efforts to cut off funding through traditional avenues have had some results. But not nearly enough attention is being paid to twenty-rst-century communications, especially al-Qaidas presence on the internet. The time has come to recognize that the stateless, anarchical realm of cyberspace requires better tools for monitoring, countermeasures, and, potentially, even control. Previous leaps in cross-border communication such as the telegraph, radio, and telephone engendered counteracting developments in

116. Nor will withdrawal from occupation in Iraq and Palestine, for example, solve this problem: al-Qaida-associated suicide attacks in Egypt, Morocco, Saudi Arabia, and Turkey belie that argument. See Robert A. Pape, Dying to Win: The Strategic Logic of Suicide Terrorism (New York: Random House, 2005).

Contending with Terrorism 414

code breaking, monitoring, interception, and wiretapping. This may seem a heretical suggestion for liberal states, especially for a state founded on the right of free speech; however, the international community will inevitably be driven to take countermeasures in response to future attacks. It is time to devote more resources to addressing this problem now. Western analysts have been misguided in focusing on the potential use of the internet for so-called cyberterrorism (i.e., its use in carrying out attacks); the internet is far more dangerous as a tool to shore up and perpetuate the al-Qaida movements constituency.117 Preventing or interdicting al-Qaidas ability to disseminate its message and draw adherents into its orbit is crucial. Countering its messages in serious ways, not through the outdated and stilted vehicles of government websites and ofcial statements but through sophisticated alternative sites and images attractive to a new generation, is an urgent priority. As for the opponents marketing strategy, al-Qaida and its associates have made serious mistakes of timing, choice of targets, and technique; yet the United States and its allies have done very little to capitalize on them. In particular, the United States tends to act as if al-Qaida is essentially a static enemy that will react to its actions, but then fails to react effectively and strategically to the movements missteps. The Bali attacks, the May 2003 attacks in Saudi Arabia, the Madrid attacks, the July 2005 London attacksall were immediately and deliberately trumpeted by al-Qaida associates. Where was the coordinated counterterrorist multimedia response? There is nothing so effective at engendering public revulsion as images of murdered and maimed victims, many of whom resemble family members of would-be recruits, lying on the ground as the result of a terrorist act. Outrage is appropriate. Currently, however, those images are dominated by would-be family members in Iraq, the Palestinian territories, Abu Ghraib, and Guantanamo Bay. The West is completely outanked on the airwaves, and its countermeasures are virtually nonexistent on the internet. But as the RIRA, PFLP-GC, and ETA cases demonstrate, the al-Qaida movement can undermine itself, if it is given help. A large part of this war is arguably being fought not on a battleeld but in cyberspace. The time-honored technique of undermining active and passive support for a terrorist group through in-depth analysis, agile responses to missteps,

117. On the broader signicance of using the internet as a means of mobilization, see Audrey Kurth Cronin, Cyber-mobilization: The New Leve en Masse, Parameters, Vol. 36, No. 2 (Summer 2006), pp. 7787.

How al-Qaida Ends 415

carefully targeted messages, and cutting-edge technological solutions is a top priority. This is a crucial moment of opportunity. Polls indicate that many of alQaidas potential constituents have been deeply repulsed by recent attacks. According to the Pew Global Attitudes Project, publics in many predominantly Muslim states increasingly see Islamic extremism as a threat to their own countries, express less support for terrorism, have less condence in bin Laden, and reect a declining belief in the usefulness of suicide attacks.118 In these respects, there is a growing range of commonality in the attitudes of Muslim and non-Muslim publics; yet the United States focuses on itself and does little to nurture cooperation. American public diplomacy is the wrong concept. This is not about the United States and its ideals, values, culture, and image abroad: this is about tapping into a growing international norm against killing innocent civilianswhether, for example, on vacation or on their way to work or schoolmany of whom are deeply religious and many of whom also happen to be Muslims. If the United States and its allies fail to grasp this concept, to work with local cultures and local people to build on common goals and increase their alienation from this movement, then they will have missed a long-established and promising technique for ending a terrorist group such as al-Qaida. There is little to say about the sixth factor, the use of military repression, in ending al-Qaida. Even though the U.S. military has made important progress in tracking down and killing senior operatives, the movements ability to evolve has demonstrated the limits of such action, especially when poorly coordinated with other comparatively underfunded approaches and engaged in by a democracy. Although apparently effective, the Turkish governments repression of the PKK and the Peruvian governments suppression of Shining Path, for example, yield few desirable parallels for the current counterterrorist campaign. Transitioning out of terrorism and toward either criminality or full insurgency is the nal, worrisome historical precedent for al-Qaida. In a sense, the network is already doing both. Efforts to cut off funding through the formal banking system have ironically heightened the incentive and necessity to en-

118. See Islamic Extremism: Common Concern for Muslim and Western Publics, Support for Terror Wanes among Muslim Publics, July 14, 2005, http:/ /pewglobal.org/reports/display .php?ReportID 248.

Contending with Terrorism 416

gage in illicit activities, especially narcotics trafcking. With the increasing amount of poppy seed production in Afghanistan, al-Qaida has a natural pipeline to riches. This process is well under way. As for al-Qaida becoming a full insurgency, some analysts believe this has already occurred.119 Certainly to the extent that Abu Musab al-Zarqawi and his associates in Iraq truly represent an arm of the movement (i.e., al-Qaida in Iraq), that transition is likewise well along. The alliance negotiated between bin Laden and al-Zarqawi is another example of an effective strategic and public relations move for both parties, giving new life to the al-Qaida movement at a time when its leaders are clearly on the run and providing legitimacy and fresh recruits for the insurgency in Iraq. As many commentators have observed, Iraq is an ideal focal point and training ground for this putative global insurgency. The glimmer of hope in this scenario, however, is that the foreigners associated with al-Qaida are not tied to the territory of Iraq in the same way the local population is, and the tensions that will arise between those who want a future for the nascent Iraqi state and those who want a proving ground for a largely alien ideology and virtual organization are likely to increaseespecially as the victims of the civil war now unfolding there continue increasingly to be Iraqi civilians. The counter to al-Zarqawis al-Qaida in Iraq, as it is for other areas of the world with local al-Qaida afliates, is to tap into the long-standing and deep association between peoples and their territory and to exploit the inevitable resentment toward foreign terrorist agendas, while scrupulously ensuring that the United States is not perceived to be part of those agendas. If the United States continues to treat al-Qaida as if it were utterly unprecedented, as if the decades-long experience with ghting modern terrorism were totally irrelevant, then it will continue to make predictable and avoidable mistakes in responding to this threat. It will also miss important strategic opportunities. That experience points particularly toward dividing new local afliates from al-Qaida by understanding and exploiting their differences with the movement, rather than treating the movement as a monolith. It is also crucial to more effectively break the political and logistical connections between the movement and its supporters, reinvigorating time-honored counterterrorism tactics targeted at al-Qaidas unique characteristics, including the perpetuation of its message, its funding, and its communications. Al-Qaida continues to exploit what is essentially a civil war within the Muslim world, attracting alien-

119. Among them is Michael Scheuer, author of Anonymous, Imperial Hubris.

How al-Qaida Ends 417

ated Muslims around the globe to its rage-lled movement. Al-Qaida will end when the West removes itself from the heart of this ght, shores up international norms against terrorism, undermines al-Qaidas ties with its followers, and begins to exploit the movements abundant missteps.

Conclusion
Major powers regularly relearn a seminal lesson of strategic planning, which is that embarking on a long war or campaign without both a grounding in previous experience and a realistic projection of an end state is folly. This is just as true in response to terrorism as it is with more conventional forms of political violence. Terrorism is an illegitimate tactic that by its very nature is purposefully and ruthlessly employed.120 At the heart of a terrorists plan is seizing and maintaining the initiative. Policymakers who have no concept of a feasible outcome are unlikely to formulate clear steps to reach it, especially once they are compelled by the inexorable action/reaction, offense/defense dynamic that all too often drives terrorism and counterterrorism. Although history does not repeat itself, ignoring history is the surest way for a state to be manipulated by the tactic of terrorism. At the highest levels, U.S. counterterrorism policy has been formulated organically and instinctively, in reaction to external stimuli or on the basis of unexamined assumptions, with a strong bias toward U.S. exceptionalism. Sound counterterrorism policy should be based on the full range of historical lessons learned about which policies have worked, and under which conditions, to hasten terrorisms decline and demise. Treating al-Qaida as if it were sui generis is a mistake. As I have argued here, while there are unique aspects to this threat, there are also connections with earlier threats. Speaking of an unprecedented jihadist threat, while arguably resonating in a U.S. domestic context, only perpetuates the image and perverse romanticism of the al-Qaida movement abroad, making its ideology more attractive to potential recruits. Such an approach also further undermines any inclination by the United States to review and understand the relationship between historical instances of terrorism and the contemporary plotting of a strategy for accelerating al-Qaidas demise. In short, formulating U.S. counterterrorism strategy as if no other state has ever faced an analogous threat is a serious blunder.
120. I have dealt at length with the question of the denition of terrorism elsewhere and will not rehash those arguments here. See, for example Cronin, Behind the Curve, pp. 3233.

Contending with Terrorism 418

Comparatively speaking, the United States has not had a great deal of experience with terrorism on its territory. In this respect, its response to the shock of al-Qaidas attacks is understandable. But the time for a learning curve is past. Intellectually, it is always much easier to over- or underreact to terrorist attacks than it is to take the initiative and think through the scenarios for how a terrorist group, and a counterterrorist effort, will wind down. Short-term reactive thinking is misguided for two reasons: the extraordinary and expensive effort to end terrorism will be self-perpetuating, and the inept identication of U.S. aims will ensure that the application of means is unfocused and ill informed by past experience. Failing to think through al-Qaidas termination, and how U.S. policy either advances or precludes it, is an error not only for the Bush administration, criticized by some for allegedly wanting an excuse to hype a permanent threat, but also for any administration of either political party that succeeds or replaces it. Terrorism, like war, never ends; however, individual terrorist campaigns and the groups that wage them always do. A vague U.S. declaration of a war on terrorism has brought with it a vague concept of the closing stages of alQaida rather than a compelling road map for how it will be reduced to the level of a minor threat. The only outcome that is inevitable in the current U.S. policy is that militarily focused efforts will end, because of wasteful or counterproductive effort and eventual exhaustion. The threat is real and undeniable, but continuing an ahistorical approach to effecting al-Qaidas end is a recipe for failure, the further alienation of allies, and the squandering of U.S. power.

International Security
The Robert and Rene Belfer Center for Science and International Affairs John F. Kennedy School of Government Harvard University All of the articles in this reader were previously published in International Security, a quarterly journal sponsored and edited by the Robert and Rene Belfer Center for Science and International Affairs at Harvard Universitys John F. Kennedy School of Government, and published by MIT Press Journals. To receive journal subscription information or to nd out more about other readers in our series, please contact MIT Press Journals at 238 Main Street, suite 500, Cambridge, MA 02142, or on the web at http:// www.mitpress.mit.edu.